#LyX 2.3 created this file. For more info see http://www.lyx.org/ \lyxformat 544 \begin_document \begin_header \save_transient_properties true \origin unavailable \textclass article \begin_preamble \pagenumbering{roman} \end_preamble \use_default_options true \begin_modules customHeadersFooters minimalistic todonotes figs-within-sections \end_modules \maintain_unincluded_children false \language british \language_package default \inputencoding utf8 \fontencoding global \font_roman "default" "default" \font_sans "default" "default" \font_typewriter "default" "default" \font_math "auto" "auto" \font_default_family default \use_non_tex_fonts false \font_sc false \font_osf false \font_sf_scale 100 100 \font_tt_scale 100 100 \use_microtype true \use_dash_ligatures true \graphics default \default_output_format default \output_sync 0 \bibtex_command biber \index_command default \paperfontsize default \spacing single \use_hyperref true \pdf_title "Net-zero Cable Repair Ship" \pdf_author "Andy Pack" \pdf_bookmarks true \pdf_bookmarksnumbered false \pdf_bookmarksopen false \pdf_bookmarksopenlevel 1 \pdf_breaklinks false \pdf_pdfborder true \pdf_colorlinks false \pdf_backref false \pdf_pdfusetitle true \papersize default \use_geometry true \use_package amsmath 1 \use_package amssymb 1 \use_package cancel 1 \use_package esint 1 \use_package mathdots 1 \use_package mathtools 1 \use_package mhchem 1 \use_package stackrel 1 \use_package stmaryrd 1 \use_package undertilde 1 \cite_engine biblatex \cite_engine_type authoryear \biblio_style plain \biblio_options urldate=long \biblatex_bibstyle authoryear \biblatex_citestyle authoryear \use_bibtopic false \use_indices false \paperorientation portrait \suppress_date true \justification false \use_refstyle 1 \use_minted 0 \index Index \shortcut idx \color #008000 \end_index \leftmargin 2cm \topmargin 2cm \rightmargin 3.5cm \bottommargin 2cm \secnumdepth 3 \tocdepth 3 \paragraph_separation skip \defskip medskip \is_math_indent 0 \math_numbering_side default \quotes_style british \dynamic_quotes 0 \papercolumns 1 \papersides 1 \paperpagestyle fancy \bullet 1 0 9 -1 \bullet 2 0 24 -1 \tracking_changes false \output_changes false \html_math_output 0 \html_css_as_file 0 \html_be_strict false \end_header \begin_body \begin_layout Standard \begin_inset CommandInset toc LatexCommand tableofcontents \end_inset \begin_inset Newpage newpage \end_inset \end_layout \begin_layout Standard \begin_inset FloatList table \end_inset \end_layout \begin_layout Standard \begin_inset FloatList figure \end_inset \end_layout \begin_layout Standard \begin_inset Newpage newpage \end_inset \end_layout \begin_layout List of TODOs \end_layout \begin_layout Standard \begin_inset Newpage newpage \end_inset \end_layout \begin_layout Right Footer Andy Pack \end_layout \begin_layout Left Footer January 2021 \end_layout \begin_layout Left Header Sustainable Cable Ship - Group 1 \end_layout \begin_layout Standard \lang english \begin_inset ERT status open \begin_layout Plain Layout \backslash pagenumbering{arabic} \end_layout \begin_layout Plain Layout \backslash setcounter{page}{1} \end_layout \end_inset \end_layout \begin_layout Section Introduction \end_layout \begin_layout Subsection Sustainability \end_layout \begin_layout Part Vessel Study \end_layout \begin_layout Section Propulsion \end_layout \begin_layout Subsection Power Requirements \end_layout \begin_layout Subsubsection Hotel Load [AP] \end_layout \begin_layout Standard The hotel load is defined as the energy usage not related to the propulsion including lighting and power outlets. An estimation of this load for the vessel was modelled and a breakdown can be seen in figure \begin_inset CommandInset ref LatexCommand ref reference "fig:hotel-load" plural "false" caps "false" noprefix "false" \end_inset . The total load was estimated at 680 kWh per day. As can be seen the oven and food refrigeration make up the majority and so 3 methods for refrigeration were investigated in order to find the most energy efficient solution. These included a collection of standard upright fridges, all-in-one cold rooms and a bespoke cold room. All-in-one cold-rooms were the most energy efficient and were selected as a result. \end_layout \begin_layout Standard \begin_inset Float figure wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Graphics filename hotel-load.png lyxscale 40 width 90col% \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout Breakdown of hotel load energy for the surface vessel, totals to \begin_inset Formula $\thickapprox$ \end_inset 680 kW \begin_inset CommandInset label LatexCommand label name "fig:hotel-load" \end_inset \end_layout \end_inset \end_layout \end_inset \end_layout \begin_layout Section Efficiency Investigations \end_layout \begin_layout Subsection Solar [AP] \end_layout \begin_layout Standard The scope of the vessel's solar energy capabilities was investigated with the intention of supplementing the chemical energy of the ammonia fuel cells. The capabilities of photovoltaic cells covering an area of the vessel's footprint are considered and compared with both the financial and carbon cost in an effort to determine whether the proposal would be effective in achieving the goal of net-zero operations. \end_layout \begin_layout Standard The expected available area capable of hosting photovoltaic cells was estimated to be 26m x 30m or 780 m \begin_inset script superscript \begin_layout Plain Layout 2 \end_layout \end_inset . Using the same solar panel modelling of GEORGE, this would equate to 441 panels with a max power of 176 kW, an order of magnitude smaller than the ammonia fuel cell capabilities. This would represent 123 tonnes of embodied carbon which could be reduced to 65 tonnes by recycling following decommission. \end_layout \begin_layout Standard As a result of this projection, it is proposed that the vessel is not fitted with solar panels. As the other energy being generated via the ammonia cells is already carbon-zer o (assuming the use of \emph on green ammonia \emph default ), the embodied carbon of the panels would not be offset through their own use. This carbon cost would need to be offset via the project-level offsetting processes, see JASON. The benefits would be limited to the financial savings of reducing fuel usage, limiting carbon cost is more of a priority for this project. \end_layout \begin_layout Section Energy Storage [AP] \begin_inset CommandInset label LatexCommand label name "sec:Energy-Storage" \end_inset \end_layout \begin_layout Standard The use of ammonia fuel cells for power generation on the vessel provides the opportunity to eliminate direct CO \begin_inset script subscript \begin_layout Plain Layout 2 \end_layout \end_inset emissions from the vessel; when produced using renewable energy ( \emph on green ammonia \emph default ), the entire fuel supply chain from production to use can be made carbon-neutra l. From an electrical perspective, however, the current-voltage characteristics of such a system must be considered. \end_layout \begin_layout Standard Figure \begin_inset CommandInset ref LatexCommand ref reference "fig:fuel-cell-iv" plural "false" caps "false" noprefix "false" \end_inset presents the I-V characteristics for a typical fuel cell, it can be seen that drawing more current from a cell reduces its voltage. As \begin_inset Formula $P=IV$ \end_inset , this inverse relationship results in an optimum current draw to operate with the highest efficiency or power density. Operating outside of this area will accentuate losses, the dominant effects of each operating region can be seen in figure \begin_inset CommandInset ref LatexCommand ref reference "fig:fuel-cell-iv-losses" plural "false" caps "false" noprefix "false" \end_inset . Comparing the two graphs, it can be seen that the optimum operating state would be in R-2; in fact drawing excess current and pushing into R-3 can damage the cell, \begin_inset CommandInset citation LatexCommand citep key "elec-a2z-fuel-cell-iv" literal "false" \end_inset . \end_layout \begin_layout Standard \begin_inset Float figure wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Graphics filename fuel-cell-i-v.gif width 60col% \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout Current-Voltage characteristics for a typical fuel cell, rated operating point highlighted, \begin_inset CommandInset citation LatexCommand cite key "strathclyde-fuel-cell-efficiency" literal "false" \end_inset \begin_inset CommandInset label LatexCommand label name "fig:fuel-cell-iv" \end_inset \end_layout \end_inset \end_layout \end_inset \end_layout \begin_layout Standard \begin_inset Float figure wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Graphics filename fuel-cell-iv-a2z.jpg lyxscale 50 width 60col% \end_inset . \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout Current-Voltage characteristics for a fuel cell with dominant losses highlighted in each operating region, \begin_inset CommandInset citation LatexCommand cite key "elec-a2z-fuel-cell-iv" literal "false" \end_inset \begin_inset CommandInset label LatexCommand label name "fig:fuel-cell-iv-losses" \end_inset \end_layout \end_inset \end_layout \begin_layout Plain Layout \end_layout \end_inset \end_layout \begin_layout Standard From these figures, fuel cells could be described as being sensitive to a noisy or dynamic load such as changes in thrust and therefore the required power can vary quickly. For example, when using dynamic positioning in a high sea state. Ideally, the use of more cells operating in their optimum state would be preferred over increasing the draw on a smaller population \begin_inset Flex TODO Note (Margin) status open \begin_layout Plain Layout Is this valid? \end_layout \end_inset . However, this increase in active cells is not an instantaneous operation and cells require time to reach their optimum state. To allow this focus on efficiency, the load including hotel and propulsion power should be decoupled from the fuel cells with an electrical storage buffer in between. This will allow the buffer to absorb spikes in load draw and allow the fuel cells to increase power generation by increasing active cells instead of individual draw. \end_layout \begin_layout Standard The following section outlines solutions for this described buffer, rechargeable batteries are the natural option and as such this is considered first. Other, innovative solutions are also outlined before the implementation of a suitable solution is presented along with the safety and financial implications of such a system. \end_layout \begin_layout Subsection Rechargeable Battery Chemistry \end_layout \begin_layout Standard There are many different methods for constructing a traditional rechargeable battery or \emph on secondary cell \emph default ; the chemistry of the reactants determines the characteristics of the system as well as having drastic implications on the safety and sustainability. Secondary cells are a consumable item, their components degrade with usage and this lifespan will be reduced if not constructed and maintained correctly. This only accentuates the importance of the solution's sustainability as it constitutes a significant amount of material which will periodically require replacing and disposal. \end_layout \begin_layout Standard \begin_inset Float table wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Tabular \begin_inset Text \begin_layout Plain Layout \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \series bold NiCd \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \series bold NiMH \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \series bold Lead Acid \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \series bold Li-ion \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \series bold Li-ion Polymer \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \series bold Reusable Alkaline \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Gravimetric Energy Density \begin_inset Newline newline \end_inset \series bold (Wh/kg) \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 45 - 80 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 60 - 120 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 30 - 50 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 110 - 160 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 100 - 130 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 80 (initial) \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Cycle Life \begin_inset Newline newline \end_inset \series bold (to 80% of initial) \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 1500 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 300 - 500 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 200 - 300 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 500 - 1000 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 300 - 500 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 50 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Self-discharge / Month \begin_inset Newline newline \end_inset \series bold (room temperature) \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 20% \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 30% \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 5% \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 10% \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout ~10% \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 0.3% \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Load Current \begin_inset Newline newline \end_inset \series bold (Peak) \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 20C \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 5C \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 5C \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout >2C \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout >2C \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 0.5C \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Load Current \begin_inset Newline newline \end_inset \series bold (Ideal) \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 1C \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \begin_inset Formula $\leq$ \end_inset 0.5C \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 0.2C \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \begin_inset Formula $\leq$ \end_inset 1C \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \begin_inset Formula $\leq$ \end_inset 1C \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \begin_inset Formula $\leq$ \end_inset 0.2C \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Operating Temperature \begin_inset Newline newline \end_inset \series bold (discharge only) \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout -40 - 60°C \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout -20 - 60°C \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout -20 - 60°C \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout -20 - 60°C \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 0 - 60°C \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 0 - 65°C \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Commercial Use Since \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 1950 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 1990 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 1970 (sealed lead acid) \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 1991 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 1999 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 1992 \end_layout \end_inset \end_inset \end_layout \begin_layout Plain Layout \noindent \align center \begin_inset VSpace smallskip \end_inset \end_layout \begin_layout Plain Layout \begin_inset CommandInset citation LatexCommand citep key "battery-uni-chemistry-types" literal "false" \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout Comparison of physical characteristics for common rechargeable battery chemistry \begin_inset CommandInset label LatexCommand label name "tab:battery-chemistry-stats" \end_inset \end_layout \end_inset \end_layout \end_inset \end_layout \begin_layout Standard Table \begin_inset CommandInset ref LatexCommand ref reference "tab:battery-chemistry-stats" plural "false" caps "false" noprefix "false" \end_inset outlines the relevant characteristics for the most common configurations of rechargeable battery. As can be seen, Lithium-ion technology leads the other solutions in most of the categories. While Nickel-Cadmium has a higher lifespan than Li-ion there are other factors that led to this being discounted. NiCd suffers from the \emph on memory effect \emph default , where frequent charge/discharge cycles lead to the battery \emph on remembering \emph default the point at which charging began and experiencing a drop in voltage past this point. Additionally, Cadmium is a highly toxic heavy metal, requiring specialist containment; in fact, many types of Cadmium battery are now banned in the EU. \end_layout \begin_layout Standard Lithium-ion batteries are a mature domain and one of active research; they are essentially the standard for portable electronics and the growing electric vehicle market. \end_layout \begin_layout Subsection Innovative Solutions \end_layout \begin_layout Standard Traditional rechargeable batteries of varying chemistry are currently the standard for this domain. However, other systems utilising different technologies were also considered. \end_layout \begin_layout Subsubsection Flow Battery \end_layout \begin_layout Standard A redox flow battery is a type of electrochemical cell where the energy is stored in two chemicals brought together at a membrane in order to facilitat e ion exchange and create a potential difference or voltage. \end_layout \begin_layout Standard This can be structured to function like a rechargeable battery as the chemical reaction is reversible. \end_layout \begin_layout Standard There are a number of advantages to a system like this, for example it is less sensitive than Lithium-ion to overcharge and overdischarge with no need for equalisation charging. \end_layout \begin_layout Standard \begin_inset Float table wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Tabular \begin_inset Text \begin_layout Plain Layout \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Energy Density (WhL \begin_inset script superscript \begin_layout Plain Layout -1 \end_layout \end_inset ) \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Power Density (WL \begin_inset script superscript \begin_layout Plain Layout -1 \end_layout \end_inset ) \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Bromine-polysulphide \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 20 - 35 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 60 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Vanadium-vanadium \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 20 - 35 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 60 - 100 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Iron-chromium \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 20 - 35 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 6 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Vanadium-bromine \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 20 - 35 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 50 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Zinc-/bromine \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 20 - 35 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 40 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Zinc-cerium \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 20 - 35 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 50 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Soluble lead-acid \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 20 - 35 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 25 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Conventional lead-acid \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 60 - 80 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 230 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Lithium-ion \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 150 - 200 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 275 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Nickel-metal-hydride \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 100 - 150 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 330 \end_layout \end_inset \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout Energy and power densities for typical redox flow battery chemistry (top) compared to traditional rechargeable cells (bottom), \begin_inset CommandInset citation LatexCommand citep key "flow-battery-energy-density" literal "false" \end_inset \begin_inset CommandInset label LatexCommand label name "tab:flow-battery-energy-density" \end_inset \end_layout \end_inset \end_layout \end_inset \end_layout \begin_layout Standard The main disadvantage relevant to the required applications is the required space and weight. Flow batteries typically have a much lower energy and power density than traditional rechargeable batteries, a comparison of values for various structures can be seen in table \begin_inset CommandInset ref LatexCommand ref reference "tab:flow-battery-energy-density" plural "false" caps "false" noprefix "false" \end_inset . Both values are critical for the vessel; as a buffer for absorbing large peaks from the propulsion, it is key that a high power can be drawn from the solution. From the presented values, Lithium-ion can provide between about 3 and 7 times as much power draw than redox flow batteries. Looking to energy or capacity density, Li-ion is roughly 4 to 10 times higher. \begin_inset Flex TODO Note (Margin) status open \begin_layout Plain Layout Sentence on energy density \end_layout \end_inset \end_layout \begin_layout Standard To achieve the capacity and power requirements defined by the propulsion system, a flow battery would likely need vastly more space and weigh significan tly more. The weight penalty would prove more damaging as this would require more fuel for propulsion and lower the efficiency of the vessel. \end_layout \begin_layout Standard Although flow batteries have found applications in large-capacity applications including grid services from load balancing to peak shaving, these are typically stationary without as much of a volume restriction. \end_layout \begin_layout Subsubsection Solid-State \end_layout \begin_layout Standard The previously described traditional rechargeable batteries have a liquid electrolyte in which the electrodes are submerged. This solution allows ions to pass between the electrodes during charge and discharge. Systems in which this liquid electrolyte is instead a solid are called solid-state batteries (SSB). This provides an advantage in that these liquid electrolytes are typically one of the key causes of safety concerns as they are flammable and sometimes toxic. These are important considerations for this domain. With less of a concern regarding high operating temperatures, this also allows SSBs to be charged faster. \end_layout \begin_layout Standard Additionally, the system has the opportunity to increase energy density, thereby reducing the required space while increasing the available capacity. \end_layout \begin_layout Standard Unfortunately, however, there are a number of considerations that, currently, make it unsuitable for the required application. As an active area of research without much commercial availability, the price of solid-state batteries is much higher than that of other formats, for comparison, the \begin_inset Flex TODO Note (Margin) status open \begin_layout Plain Layout Price \end_layout \end_inset . Additionally, concerns associated with Lithium-ion batteries including overheating and explosion are not completely removed by transitioning to an SSB. Dendrites are structures of Lithium that can form during charging and dischargi ng as a result of electrodeposition. While this should occur evenly across the electrode, if uneven it can cause columns to grow towards the separator. As these grow they can penetrate through this separator and make contact with the cathode. This will cause a short circuit, rapidly increasing heat and potentially causing fire and explosion. \end_layout \begin_layout Standard \begin_inset Float figure wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Graphics filename dendrite.jpg width 40col% \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout Dendrites growing between the Lithium battery electrodes, \begin_inset CommandInset citation LatexCommand citep key "mse-supplies-dendrite" literal "false" \end_inset \begin_inset CommandInset label LatexCommand label name "fig:Dendrites" \end_inset \end_layout \end_inset \end_layout \end_inset \end_layout \begin_layout Standard \begin_inset CommandInset citation LatexCommand cite key "4-ssb-challenges-article" literal "false" \end_inset summarises four current challenges to scaling up solid-state batteries as investigated by \begin_inset CommandInset citation LatexCommand cite key "SSB-challenges-paper" literal "false" \end_inset : \end_layout \begin_layout Enumerate Stability of the electrode-electrolyte interface \end_layout \begin_layout Enumerate Characterising and analysing this interface in a now opaque structure \end_layout \begin_layout Enumerate Sustainable manufacturing processes \end_layout \begin_layout Enumerate Designing for recyclability \end_layout \begin_layout Standard These last two are critical for the project objectives. With these considerations it is suggested that solid-state batteries will not be ready at the time of construction to supply the scale of storage array required. \end_layout \begin_layout Subsection Proposed Solution \begin_inset CommandInset label LatexCommand label name "subsec:Proposed-UUV-Battery-Solution" \end_inset \end_layout \begin_layout Standard For this project, Lithium-ion chemistry was proposed as the solution for the vessel energy storage. As previously mentioned, the domain is an area of fervent research as a result of its importance to consumer electronics and electric vehicles. An important factor in the decision is the scale of system required, this will have significant impacts on the ability to source and dispose of a system as well as the financial and safety implications. \end_layout \begin_layout Standard There are a variety of chemical compositions for Lithium-ion batteries depending on the other materials used in the electrodes including Lithium Manganese Oxide and Lithium Cobalt Oxide. Each have specific benefits and associated applications, for our purposes Lithium Nickel Manganese Cobalt Oxide will be used as it is a common chemistry with high specific energy, \begin_inset CommandInset citation LatexCommand cite key "batt-uni-lithium-chemistry" literal "false" \end_inset . \end_layout \begin_layout Standard There are many standard Lithium-ion standard cell formats from flat pouches and prismatic cells designed for mobile phones, to the more standard cylindrica l cells. For this application, cylindrical cells are a suitable choice as compactness and thinness are not critical design parameters. \end_layout \begin_layout Standard The 18650 cell is a mature cylindrical cell with good reliability records and high rates of use among medical equipment, drones and electric vehicles; Tesla uses battery packs composed of 18650 cells. \end_layout \begin_layout Standard As with other battery cells, the voltage is a characteristic of the chemistry, for Lithium this is around 3.6 V. The key parameters that vary amongst producers are the capacity and charge/disc harge C-rates. In order to estimate the cell specification for use in this project, the existing range of available cells was taken into account. Typical, mid-range 18650 cells can vary between 2500 - 3000 mAh capacity; the highest energy density can currently extend this to 3500 - 3600 mAh. As technology improves, it is expected that by the point of construction this higher range will be more accessible and reliable, and as such, 3500 mAh is used as the cell capacity for further calculations. \end_layout \begin_layout Standard The 18650 cell specifications being used herein are described in table \begin_inset CommandInset ref LatexCommand ref reference "tab:18650-specs" plural "false" caps "false" noprefix "false" \end_inset . \end_layout \begin_layout Standard \begin_inset Float table wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Tabular \begin_inset Text \begin_layout Plain Layout \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \series bold 18650 Cell \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Voltage, ( \begin_inset Formula $V$ \end_inset ) \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 3.6 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Capacity, ( \begin_inset Formula $mAh$ \end_inset ) \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 3500 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Ideal Discharge C-Rate, ( \begin_inset Formula $h^{-1}$ \end_inset ) \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 1 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Ideal Charge C-Rate, ( \begin_inset Formula $h^{-1}$ \end_inset ) \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 0.5 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Weight, ( \begin_inset Formula $g$ \end_inset ) \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 48 \end_layout \end_inset \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout General specifications for 18650 Lithium-ion cells \begin_inset CommandInset label LatexCommand label name "tab:18650-specs" \end_inset \end_layout \end_inset \end_layout \end_inset \end_layout \begin_layout Subsubsection Configuration \end_layout \begin_layout Standard The quantity of required cells for the battery system was calculated using the expected propulsion power requirements in conjunction with the expected generation capabilities of the ammonia fuel cells. The specifics for the calculations can be seen in appendix \begin_inset CommandInset ref LatexCommand ref reference "sec:Battery-Cell-Calculations" plural "false" caps "false" noprefix "false" \end_inset ; to summarise, the quantity of required cells was calculated from the required power draw of the battery and the characteristics of the 18650 Lithium cell being used. The result was 193,600 cells. These cells are arranged into a matrix of parallel and series blocks, all the series blocks connected in parallel must be of the same length \begin_inset Flex TODO Note (Margin) status open \begin_layout Plain Layout Figure? \end_layout \end_inset . \end_layout \begin_layout Standard The balance of parallel to series blocks is not a critical parameter for the application and can instead be tuned for efficiency; as the propulsion units require AC power, transformers can be used to select a desired voltage and current from a given power value. For high power applications high voltage is typically preferred to high current which corresponds to a higher weighting of series length. \end_layout \begin_layout Subsubsection Challenges \end_layout \begin_layout Paragraph Self-discharge \end_layout \begin_layout Standard \begin_inset Flex TODO Note (inline) status open \begin_layout Plain Layout Self discharge per month, 1-10%? \end_layout \end_inset \end_layout \begin_layout Paragraph Limited Lifespan \end_layout \begin_layout Standard Traditional rechargeable battery cells are a consumable item with the capacity and performance decreasing over extended use for a number of reasons. These include electrode corrosion, reduced porosity or a reduction in Lithium ions as a result of side reactions, \begin_inset CommandInset citation LatexCommand citep key "li-ion-degradation" literal "false" \end_inset . \end_layout \begin_layout Standard When the cells begin to perform below a defined acceptable level, they will require replacement. This poses both financial and environmental implications. As will be discussed in section \begin_inset CommandInset ref LatexCommand ref reference "subsec:Extending Lifespan" plural "false" caps "false" noprefix "false" \end_inset this is a significant amount of money. Additionally, the scale of required cells means that the disposal protocols are critical. Options for such disposal procedures are discussed as part of the life cycle analysis in section \begin_inset CommandInset ref LatexCommand ref reference "subsec:Life-cycle-Analysis" plural "false" caps "false" noprefix "false" \end_inset . \end_layout \begin_layout Paragraph Safety \end_layout \begin_layout Standard Although Lithium-ion batteries are typically safe and stable if stored and used correctly, abuse can cause severe safety issues. As previously mentioned, the liquid organic electrolyte is flammable, and combined with the high energy density of Li-ion batteries can lead to thermal runaway and eventually fire if not handled correctly. \end_layout \begin_layout Standard There are a couple of causes for such a thermal runaway, these include physical damage, short circuits, overcharging and exposure to high temperature, \begin_inset CommandInset citation LatexCommand citep key "washington-lithium-safety" literal "false" \end_inset . \end_layout \begin_layout Standard Multiple measures must be implemented to ensure safety, \begin_inset CommandInset citation LatexCommand citep key "washington-lithium-safety" literal "false" \end_inset : \end_layout \begin_layout Itemize Reputable manufacturers must be used as defects during construction can cause or exacerbate faults. \end_layout \begin_layout Itemize The battery system should be located far from combustible materials \end_layout \begin_deeper \begin_layout Itemize Contextually this would primarily be the fuel tanks. \end_layout \end_deeper \begin_layout Itemize The cell's temperature should be monitored and controlled. \end_layout \begin_layout Itemize Cells of the same age must be grouped and used together \end_layout \begin_layout Itemize Cells should be charged intelligently in order to mitigate overcharge and implement equalisation charging, see \begin_inset CommandInset ref LatexCommand ref reference "subsec:Safety-Circuitry" plural "false" caps "false" noprefix "false" \end_inset \end_layout \begin_layout Itemize The battery system should be stored in an water/air-tight container \end_layout \begin_layout Itemize This container should be able to safely vent gases \end_layout \begin_deeper \begin_layout Itemize In the event of an emergency, cells can release toxic gases (CO \begin_inset script subscript \begin_layout Plain Layout 2 \end_layout \end_inset , CO, HF) \end_layout \end_deeper \begin_layout Standard With regards to the described container, as previously outlined, standard shipping containers will be used \begin_inset Flex TODO Note (Margin) status open \begin_layout Plain Layout Container spec? \end_layout \end_inset . This provides a good base to modify in order to meet the above, i.e. air/water-tight and venting. \end_layout \begin_layout Subsubsection Charging & Safety Circuitry \begin_inset CommandInset label LatexCommand label name "subsec:Safety-Circuitry" \end_inset \end_layout \begin_layout Standard As previously described, Lithium-ion cells are sensitive to stressful electrical conditions such as overcharging, deep discharging and excessive current draw. \end_layout \begin_layout Standard In order to protect and enforce operating conditions for the Lithium cells, a battery management system or BMS is used. A BMS implements safety protocols to mitigate the above effects as well as providing information to the load and other monitoring services such as state-of-charge information (remaining capacity), \begin_inset CommandInset citation LatexCommand cite key "batt-uni-bms" literal "false" \end_inset . \end_layout \begin_layout Subsubsection Extending Lifespan \begin_inset CommandInset label LatexCommand label name "subsec:Extending Lifespan" \end_inset \end_layout \begin_layout Standard Lithium-ion cells are a consumable item that degrades. The environmental and financial cost of replacement creates a significant incentive to extend this as much as possible as long as this does not inhibit the operating capabilities beyond the specification. \end_layout \begin_layout Standard As previously mentioned, the temperature of the cells is a key parameter affecting both performance and lifespan, \begin_inset CommandInset citation LatexCommand citep key "batt-uni-discharge-temp" literal "false" \end_inset . Although operating at a higher temperature increases performance it also decreases lifespan. Temperature control is already critical for safety purposes. 20°C provides the ideal temperature for prolonging lifespan and as such will be set as the target temperature, \begin_inset CommandInset citation LatexCommand citep key "batt-uni-discharge-temp" literal "false" \end_inset . \end_layout \begin_layout Standard Another important aspect to the lifespan of Lithium-ion batteries is the depth-of-discharge (DOD) which determines the number of charge cycles that the battery will last for. The depth-of-discharge describes the amount of capacity used each cycle before recharging. Lithium batteries are able to handle moderate DOD without significantly affecting the lifespan, however, frequent deep discharge cycles, completely emptying the battery, will shorten its life, \begin_inset CommandInset citation LatexCommand citep key "bat-uni-prolong-liion" literal "false" \end_inset . \end_layout \begin_layout Standard Similar to frequently completely discharging the battery, storing a battery fully charged for long periods of time can also shorten its lifespan. This can be seen presented in figure \begin_inset CommandInset ref LatexCommand ref reference "fig:charge-lifespan" plural "false" caps "false" noprefix "false" \end_inset where higher cell charge voltages can be seen to reduce capacity much faster as time or charge cycles increases \begin_inset Foot status open \begin_layout Plain Layout Worth noting that the capacity is initially higher for higher voltages, it is the lifespan that can be extended by picking a more reserved value. \end_layout \end_inset . \end_layout \begin_layout Standard \begin_inset Float figure wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Graphics filename lithium-overcharge-capacity.jpg width 55col% \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout The lifespan of Lithium-ion cells described by the max capacity as charge cycles increase for various charge voltages, \begin_inset CommandInset citation LatexCommand citep key "bat-uni-prolong-liion" literal "false" \end_inset \begin_inset CommandInset label LatexCommand label name "fig:charge-lifespan" \end_inset \end_layout \end_inset \end_layout \end_inset \end_layout \begin_layout Standard Both of the above points, tempering DOD and charge voltage rely on the BMS. To prevent deep discharges, more ammonia fuel cells should start up as the capacity decreases in order to balance the workload meaning that the ammonia management system will need to interface with the BMS. Additionally, it would be the responsibility of the BMS to charge cells to a reasonable voltage, 3.9 - 4.1 V can provide a balance between higher capacity and longer lifespan, \begin_inset CommandInset citation LatexCommand citep key "bat-uni-prolong-liion" literal "false" \end_inset . \end_layout \begin_layout Standard \begin_inset Float figure wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Graphics filename discharge-voltage-temperature.jpg lyxscale 60 width 60col% \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout Voltage-Capacity characteristics for a 18650 Li-ion cell at varying temperatures , \begin_inset CommandInset citation LatexCommand citep key "batt-uni-discharge-temp" literal "false" \end_inset \begin_inset CommandInset label LatexCommand label name "fig:temperature-characteristics" \end_inset \end_layout \end_inset \end_layout \begin_layout Plain Layout \end_layout \end_inset \end_layout \begin_layout Standard \begin_inset Flex TODO Note (inline) status open \begin_layout Plain Layout Define a replacement interval \end_layout \end_inset \end_layout \begin_layout Subsubsection Financial \begin_inset CommandInset label LatexCommand label name "subsec:batteryFinancial" \end_inset \end_layout \begin_layout Standard As described, a battery array lasts approximately 2.3 years, after which, a new set must be sourced. 18650 cells vary in price across manufacturers and distributors, a range of RRPs were taken from UK distributors, \begin_inset CommandInset citation LatexCommand cite key "18650-ecolux,18650.uk,18350-fogstar" literal "false" \end_inset . In general, a cell's RRP ranged between £4 and £7, and so it is assumed that a procurement department would secure a unit price at the lower end of this for the scale of order required, £5 is used. \end_layout \begin_layout Standard A set of 193,600 cells at this cost totals to £968,000 which averages to £420,869 a year for modelling purposes. \end_layout \begin_layout Standard Costing a battery management system is complicated by the scale of battery being proposed. A mega-watt scale battery will require a complex BMS without publicly available prices and an estimation must be made. \begin_inset CommandInset citation LatexCommand cite key "bms-cost-article,bms-cost-report" literal "false" \end_inset suggests that cells constitute 60% of the total system price. Using this estimate the BMS and pack is estimated to cost £645,000. \end_layout \begin_layout Subsection Life-cycle Analysis \begin_inset CommandInset label LatexCommand label name "subsec:Life-cycle-Analysis" \end_inset \end_layout \begin_layout Standard The life-cycle analysis (LCA) of Lithium-ion batteries is a complicated process for a couple of reasons. As repeatedly stated, Li-ion batteries have been critical to the explosion of mobile consumer electronics; the development of the fabrication process and the associated environmental effects has changed dramatically. More recent LCAs and meta-analyses of previous data are considered in order to account for this. Additionally, as a global product the values for various greenhouse gas (GHG) and other emissions is contingent on the country within which the cells are made. \end_layout \begin_layout Standard Both the cumulative energy demand (CED) and the GHG emissions are considered. Cumulative energy demand allows abstraction of the specific method of energy production and the associated emissions. \end_layout \begin_layout Standard Many of the LCA studies on Lithium-ion batteries consider a cradle-to-gate scope without including use or end-of-life. Two end-of-life procedures are considered as well as practices to improve usage lifetime. \end_layout \begin_layout Subsubsection Cradle-to-Gate \end_layout \begin_layout Standard \begin_inset Float figure wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Graphics filename battery-breakdown-mj-kwh.png lyxscale 50 width 75col% \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout CED breakdown for a NCM11 battery pack (MJ/kWh), \begin_inset CommandInset citation LatexCommand citep key "circular-energy-li-lca,argonne-li-ion-lca" literal "false" \end_inset \begin_inset CommandInset label LatexCommand label name "fig:battery-ced-breakdown" \end_inset \end_layout \end_inset \end_layout \end_inset \end_layout \begin_layout Standard Figure \begin_inset CommandInset ref LatexCommand ref reference "fig:battery-ced-breakdown" plural "false" caps "false" noprefix "false" \end_inset outlines the cumulative energy demand for the major elements of a Nickel/Cobalt /Manganese cathode (NCM11) battery. As \begin_inset CommandInset citation LatexCommand cite key "circular-energy-li-lca" literal "false" \end_inset points out, cathodes are tending towards a higher cathode composition of Nickel, however the general proportions are relevant to other chemistry. It can be seen that the production of the cells constitutes the majority of the required CED at 75% of the total. As this is also the element of the battery that requires periodic replacement following degradation, a closer look at the contributing stages should be considered. \end_layout \begin_layout Standard \begin_inset Float figure wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Graphics filename cell-breakdown-mj-kwh.png lyxscale 50 width 75col% \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout CED breakdown for a NCM11 cell without BMS or pack (MJ/kWh), \begin_inset CommandInset citation LatexCommand citep key "circular-energy-li-lca,argonne-li-ion-lca" literal "false" \end_inset \begin_inset CommandInset label LatexCommand label name "fig:cell-ced-breakdown" \end_inset \end_layout \end_inset \end_layout \end_inset \end_layout \begin_layout Standard This further breakdown can be seen in figure \begin_inset CommandInset ref LatexCommand ref reference "fig:cell-ced-breakdown" plural "false" caps "false" noprefix "false" \end_inset . The precursors, Li \begin_inset script subscript \begin_layout Plain Layout 2 \end_layout \end_inset CO \begin_inset script subscript \begin_layout Plain Layout 3 \end_layout \end_inset and cathode production constitute almost 50% of the cell's CED, these are the Lithium intensive processes. By using recycled Lithium, this major contributor can be reduced. As such, a cell manufacturer using recycled Lithium should be identified and used. \end_layout \begin_layout Standard \begin_inset Float figure wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Graphics filename battery-breakdown-c02e-kwh.png lyxscale 40 width 75col% \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout Equivalent carbon breakdown for a NCM11 battery pack (kg CO2e/kWh), \begin_inset CommandInset citation LatexCommand citep key "circular-energy-li-lca,argonne-li-ion-lca" literal "false" \end_inset \begin_inset CommandInset label LatexCommand label name "fig:battery-co2e-breakdown" \end_inset \end_layout \end_inset \end_layout \end_inset \end_layout \begin_layout Standard The contributions of each stage to the embodied carbon can be seen in figure \begin_inset CommandInset ref LatexCommand ref reference "fig:battery-co2e-breakdown" plural "false" caps "false" noprefix "false" \end_inset . As previously shown, the Lithium-containing NCM powder constitutes the largest contributor. The 73 kg CO2e/kWh quoted value is dependent on the location of manufacture, a slightly more conservative 80 kg CO2e/kWh will be used for calculating the battery systems embodied carbon from production. \end_layout \begin_layout Standard Using this value, a set of cells represents 191 tonnes of embodied carbon. This will be included in the total embodied for the project requiring offset. \end_layout \begin_layout Subsubsection Use \end_layout \begin_layout Standard The use of Lithium batteries does not inherently incur a Carbon cost; the associated cost of energy stored is accounted for by the source of this energy, in this case ammonia fuel cells \begin_inset Flex TODO Note (Margin) status open \begin_layout Plain Layout solar panels? \end_layout \end_inset . \end_layout \begin_layout Standard The use of the batteries does require analysis, however. The source and end-of-life procedures for a battery pack is a carbon intensive operation and the ability to extend the time in-between replacement will improve the environmental impact overall. \end_layout \begin_layout Standard There are a number of ways to increase the lifespan of a battery pack, these have been outlined in section \begin_inset CommandInset ref LatexCommand ref reference "subsec:Extending Lifespan" plural "false" caps "false" noprefix "false" \end_inset and are critical for reducing the environmental impact of the system. \end_layout \begin_layout Subsubsection End-of-Life \end_layout \begin_layout Standard There are two main approaches to sustainable end-of-life processing for Lithium-ion processing, second-use and recycling. \end_layout \begin_layout Standard Second-use describes the use of a battery for new applications after the performance is deemed too low for the vessel's buffer. By doing so the lifespan of the batteries can be extended and reducing the amount being constructed. \end_layout \begin_layout Standard There are many methods for recycling Lithium batteries, and it is important to identify a process that will not use more energy than that required to mine virgin materials. \begin_inset CommandInset citation LatexCommand cite key "circular-energy-li-lca" literal "false" \end_inset summarises that up to 48% of the CED and CO \begin_inset script subscript \begin_layout Plain Layout 2 \end_layout \end_inset e could be saved. They identify \emph on direct recycling \emph default as the best option for this. This method allows the electrodes to retain their composition as opposed to breaking it down into constituent parts, \begin_inset CommandInset citation LatexCommand cite key "li-direct-recycling" literal "false" \end_inset . This has applicability across Lithium-ion chemistry including the form used herein, NCM. \end_layout \begin_layout Standard For the vessel battery, the use of an intelligent BMS and tight integration with the ammonia cells should allow it to be treated well such that it is kept in comparably good condition. As such it is proposed that they would be well suited for second-use applicatio ns such as energy storage. The battery will not be used until failure but instead until it cannot be used for the high-draw requirements of the buffer. They would likely have good applicability to purposes without such focus on high draw but that will make use of the remaining capacity. \end_layout \begin_layout Subsection Sustainability \begin_inset CommandInset label LatexCommand label name "subsec:SurfaceVesselBatterySustainability" \end_inset \end_layout \begin_layout Standard Although many of the important environmental aspects of sustainability are covered by a life-cycle analysis, there are other elements to consider regarding sustainability. One of the most important aspects is a social one, that of the mining of Lithium and Cobalt. The majority of both minerals are located in two areas of the global south where resource shortages and unethical mining practices lead to dangerous and damaging results both socially and environmentally. \end_layout \begin_layout Subsubsection Lithium \end_layout \begin_layout Standard The majority of global Lithium deposits can be found in an area of South America referred to as the \emph on Lithium Triangle \emph default covering areas of Chile, Argentina and Bolivia. The area has been estimated to constitute between 54 and 70% of the world's deposits, \begin_inset CommandInset citation LatexCommand citep key "wired-lithium,resourceworld-54-lithium" literal "false" \end_inset . The extraction is a water-intensive process in an area already without an adequate supply; in Chile this is as much as 65% of the area's water or 500,000 gallons per tonne of Lithium, \begin_inset CommandInset citation LatexCommand citep key "wired-lithium" literal "false" \end_inset . \end_layout \begin_layout Standard The processing can also include dangerous chemicals including various acids that can pollute local water supplies as a result of leaks, leaching and emissions, \begin_inset CommandInset citation LatexCommand citep key "wired-lithium" literal "false" \end_inset . \end_layout \begin_layout Subsubsection Cobalt \end_layout \begin_layout Standard Over half of the world's Cobalt deposits are found in the Democratic Republic of Congo, \begin_inset CommandInset citation LatexCommand citep key "wired-lithium,ethical-consumer-conflict-materials" literal "false" \end_inset . \end_layout \begin_layout Standard Although not officially designated as such, there are efforts to class Cobalt as a conflict mineral as its importance grows to one of the most notorious countries for other such minerals including Gold and Tungsten. \end_layout \begin_layout Standard 20% of the exported cobalt has been estimated to come from artisanal mines, \begin_inset CommandInset citation LatexCommand citep key "ethical-consumer-conflict-materials" literal "false" \end_inset . These are small-scale mines known for a lack of safety standards including minimal personal protective equipment, structural requirements and child labour, \begin_inset CommandInset citation LatexCommand cite key "wef-cobalt-mining" literal "false" \end_inset . \end_layout \begin_layout Subsubsection Summary \end_layout \begin_layout Standard The above emphasises the need to identify Lithium cell manufacturers using recycled materials in order to reduce the amount of virgin material being mined and assembled. \end_layout \begin_layout Subsection Time-dependent Modelling \end_layout \begin_layout Standard In order to validate the buffer configuration, a model was constructed to visualise the capacity of the battery system while in use. Explanations as to the specific behaviour and assumptions made can be seen in appendix \begin_inset CommandInset ref LatexCommand ref reference "sec:Time-dependent-Power-Modelling" plural "false" caps "false" noprefix "false" \end_inset . An example model describing the vessel's dynamic positioning above a cable fault can be seen in figure \begin_inset CommandInset ref LatexCommand ref reference "fig:dyn-pos-power-model" plural "false" caps "false" noprefix "false" \end_inset . \end_layout \begin_layout Standard \begin_inset Float figure wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Graphics filename ../maths/final_report_res/dynpos1.png lyxscale 20 width 80col% \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout Power model describing the vessel dynamic positioning on mission \begin_inset CommandInset label LatexCommand label name "fig:dyn-pos-power-model" \end_inset \end_layout \end_inset \end_layout \end_inset \end_layout \begin_layout Standard The top sub-figure describes the power in from the ammonia fuel cells and the load power drawn from the hotel and propulsion systems. The middle figure describes the effect this has on the capacity of the battery. The bottom figure describes the efficiency of this system, the \emph on unused \emph default power describes when more fuel cells than required are turned on and power is drawn below the most efficient state. The \emph on unavailable \emph default power describes when the fuel cells in their most efficient state and the battery combined cannot meet the requirement. As a result, extra power would be drawn from both pulling them into an inefficient or damaging state. \end_layout \begin_layout Standard \begin_inset Float figure wide false sideways false status open \begin_layout Plain Layout \noindent \align center second \begin_inset Graphics filename ../maths/final_report_res/mission3.png lyxscale 20 width 80col% \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout Power model describing one mission including travelling and dynamic positioning \begin_inset CommandInset label LatexCommand label name "fig:mission-power-model" \end_inset \end_layout \end_inset \end_layout \end_inset \end_layout \begin_layout Standard This model was extended to simulate an entire mission as described in section \begin_inset Flex TODO Note (Margin) status open \begin_layout Plain Layout NICKS SECTION \end_layout \end_inset (figure \begin_inset CommandInset ref LatexCommand ref reference "fig:mission-power-model" plural "false" caps "false" noprefix "false" \end_inset ) including: \end_layout \begin_layout Itemize Outbound journey (3 days) \end_layout \begin_layout Itemize Manoeuvring to the fault (1 day) \end_layout \begin_layout Itemize Dynamic positioning while completing the first splice (2 days) \end_layout \begin_layout Itemize Manoeuvring to the other half of the cable (1 day) \end_layout \begin_layout Itemize Dynamic positioning while completing the second splice (2 days) \end_layout \begin_layout Itemize Homeward journey (3 days) \end_layout \begin_layout Standard \begin_inset Flex TODO Note (Margin) status open \begin_layout Plain Layout Probably covered by Nick \end_layout \end_inset \end_layout \begin_layout Standard From these models, the amount of battery charge cycles was estimated to be 2 per day. Extrapolating this to a yearly usage value using the expected vessel usage, a battery array was estimated to last 2.3 years. This is a typical value for the lifespan of Lithium-ion batteries. \end_layout \begin_layout Subsection Summary \end_layout \begin_layout Standard The proposed buffer solution includes 193,600 NCM Lithium-ion cells requiring replacement every 2.3 years. As a result of this replacement rate, it is stipulated that the battery be re-appropriated for second-use such as energy storage following decommission in order to extend their life and reduce the environmental impact. Manufacturers using recycled materials should also be identified in order to reduce the impact of mining virgin materials. \end_layout \begin_layout Standard It is worth noting that this is the proposed solution based on the current state of energy storage. One of the only advantages of rechargeable batteries being a consumable item requiring replacement is that when this occurs, it is an opportunity to re-evaluate the system design. With an expected project lifespan of 30 years, it is highly unlikely that the use of Li-ion cells will remain the best option. \end_layout \begin_layout Standard Systems based on a solid-state chemistry will likely become more stable and less expensive to the point that the advantages in safety and energy density can be fully utilised without the heavy downsides. \end_layout \begin_layout Section Mission Ops \end_layout \begin_layout Subsection Grapnel-based Operations [AP] \end_layout \begin_layout Standard \begin_inset Flex TODO Note (Margin) status open \begin_layout Plain Layout Possibly covered by Nick \end_layout \end_inset While the use of robotics has made sub-sea cable repair operations more efficient, there are situations where this is not available and it is worth briefly outlining how grapnels are used in repair operations. \end_layout \begin_layout Standard Grapnels are specialised tools attached to lengths of chain which trail the stern of the ship. For cable repair operations, a cut & hold grapnel is used \begin_inset CommandInset citation LatexCommand citep key "cut-and-hold-paper,cut-and-hold-eta-product" literal "false" \end_inset . With knowledge of the path of the subject cable and the location of the fault, the grapnel is lowered before the boat makes a pass perpendicular to the cable. As the grapnel makes contact it is able to both cut and grip the cable before being raised to the surface vessel. \begin_inset Flex TODO Note (Margin) status open \begin_layout Plain Layout Disadvantages \end_layout \end_inset \end_layout \begin_layout Subsection Unmanned Underwater Vehicle Operations [AP] \end_layout \begin_layout Standard The following section outlines how the use of an unmanned underwater vehicle (UUV) can make mission operations more efficient and precise. The state of current UUV usage throughout cable repair operations is outlined in order to identify the critical capabilities, requirements and advantages over traditional grapnel operations. The future of the domain is then explored and the challenges in applying these developments to sub-sea cable repair are identified before exploring how these can be overcome in order to meet the determined requirements. Prior to this, the domain of UUVs as a whole is described in order to outline the scope of available vehicles. \end_layout \begin_layout Subsubsection UUV Classes \end_layout \begin_layout Paragraph ROVs and AUVs \end_layout \begin_layout Standard UUVs can be divided into two categories based on their control scheme. Remotely operated underwater vehicles (ROV) and autonomous underwater vehicles (AUV) are distinguished by whether a human is controlling the vehicle or whether it operates independently; as such they have different applications. ROVs have been the vehicle class of choice where complex intervention and actuation is required such as offshore oil and gas operations and cable repair. A human operator controls the vehicle from the surface vessel; bi-directional communication including data, control, video and power are transmitted through an umbilical cord tether between the two vessels. AUVs on the other hand have primarily been used for survey and research purposes. \begin_inset Flex TODO Note (Margin) status open \begin_layout Plain Layout No umbilical cord? \end_layout \end_inset \end_layout \begin_layout Standard This distinction in responsibilities is not static, however. Like other robotics domains such as auto-mobiles and ships, autonomy is a rapidly developing area of research and development; newer vehicles are able to complete more complex operations without human intervention and with longer endurance. \end_layout \begin_layout Paragraph Physical Configuration \end_layout \begin_layout Standard The physical layout of a UUV can generally be described by one of two classes, box frames or torpedo shaped. \begin_inset Flex TODO Note (Margin) status open \begin_layout Plain Layout determined by the size and range of the vehicle. \end_layout \end_inset Box frame UUVs are typically larger with more space for instruments and actuators but are not expected to make longer distance journeys as a result of their poor hydrodynamic profile. Torpedo shaped vehicles tend to be smaller without actuators; their hydrodynami c profile makes them well suited for faster, longer distance missions however this comes at the cost of reduced stability and control. \end_layout \begin_layout Subsubsection Current ROV Usage \end_layout \begin_layout Standard Cable repair operations are currently undertaken, where possible, with human-con trolled ROVs. With visual contact and direct actuation at the seabed, the ROV is used to identify, cut and grip the cable for retrieval to the surface-vessel. In doing so the need for repeated motions of the ship across the cable is removed, saving time and fuel. Instead, the surface vessel uses dynamic positioning in order to maintain its position above the ROV and cable. \end_layout \begin_layout Standard While this finer control is a key benefit for ROV use over grapnels, one of the most important benefits is the ability to bury repaired cables in the sea floor using high-powered water jets. 70% of cable damage is caused by man-made activity, of which over a third is a result of fishing activity; another quarter is as a result ship anchors \begin_inset CommandInset citation LatexCommand citep key "ultra-map-cable-damage-causes" literal "false" \end_inset . As such, the ability to protect sub-sea cables in shallower waters by burying them from human intervention is a key parameter in protecting cables from further damage and extending the time between repairs. While this can be completed with a separate plough, this would require more deck space and motion of the surface vessel. \end_layout \begin_layout Standard The need for fine movement control and actuators with which to manipulate cables has led to box frame vehicles dominating this field, figure \begin_inset CommandInset ref LatexCommand ref reference "fig:The-HECTOR-7-ROV" plural "false" caps "false" noprefix "false" \end_inset shows SIMEC Technology's HECTOR-7 ROV, a typical design for sub-sea cable repair vehicles. \end_layout \begin_layout Standard \begin_inset Float figure wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Graphics filename hector-7.jpg lyxscale 50 width 50col% \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout SIMEC Technology's HECTOR-7 ROV used on Orange Marine's Pierre de Fermat, \begin_inset CommandInset citation LatexCommand citep key "rov-hector-7-datasheet" literal "false" \end_inset \begin_inset CommandInset label LatexCommand label name "fig:The-HECTOR-7-ROV" \end_inset \end_layout \end_inset \end_layout \end_inset \end_layout \begin_layout Standard \begin_inset Float table wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Tabular \begin_inset Text \begin_layout Plain Layout \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \series bold HECTOR-7 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \series bold Atlas \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \series bold ST200 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \series bold QTrencher 600 \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \series bold Company \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout SIMEC Technology \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Global Marine \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout SMD \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \series bold Vessel \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Pierre de Fermat \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Wave Sentinel \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout Cable Innovator \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout N/A \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout C.S Sovereign \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \series bold Depth Rating \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 3,000 m \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 2,000 m \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 2,500 m \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 3,000 m \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \series bold Weight in Air \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 9 t \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 10.6 t \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 6.5 t \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 11 t \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \series bold Power \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 300 kW \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 300 kW \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout - \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 450 kW \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout \series bold Burial Depth \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 2 m \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 2 m \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 1.5 m \end_layout \end_inset \begin_inset Text \begin_layout Plain Layout 3 m \end_layout \end_inset \end_inset \begin_inset VSpace smallskip \end_inset \end_layout \begin_layout Plain Layout \begin_inset CommandInset citation LatexCommand citep key "rov-hector-7-datasheet,global-marine-atlas-data-sheet,glboal-marine-st200-datasheet,smd-qtrencher-600-datasheet" literal "false" \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout Relevant specifications and operating capabilities for sub-sea cable repair ROVs \begin_inset CommandInset label LatexCommand label name "tab:ROV-specs" \end_inset \end_layout \end_inset \end_layout \end_inset \end_layout \begin_layout Standard Table \begin_inset CommandInset ref LatexCommand ref reference "tab:ROV-specs" plural "false" caps "false" noprefix "false" \end_inset lists the specifications for the ROVs currently being used as part of the ACMA cable repair agreement along with similarly classed vehicles from other providers. As can be seen, current ROVs for this domain have a maximum working depth of about 3 km. This poses a problem to cable repair operations where, further out to sea, the sea floor can extend much further, see figure \begin_inset Flex TODO Note (Margin) status open \begin_layout Plain Layout bathymetry chart? \end_layout \end_inset . While ROVs could be capable, in theory, of reaching lower depths it is important to balance these working capabilities with other considerations such as price and weight. In practice, while this working depth is a reasonable range to work within, it could be argued that the most important capability of current ROVs is their ability to re-bury the cable post-repair. As described previously, this is in order to protect the cable from human intervention including fishing and anchor operations. These incidents are more prevalent in shallower waters within the operating range of the ROV, therefore it is acceptable to use a grapnel outside of this operating range where burying the cable is less important. \end_layout \begin_layout Paragraph Requirements Specification \end_layout \begin_layout Standard Using this information, the requirements for a cable repair UUV could be described as the following, \end_layout \begin_layout Enumerate The UUV should have actuators in order to both cut and grip cables \end_layout \begin_layout Enumerate The UUV should be able to operate to at least 2 km of depth \end_layout \begin_layout Enumerate The UUV should be able to locate the cable without visual information i.e. electromagnetically \end_layout \begin_deeper \begin_layout Enumerate In shallower water the cable is buried and will not be able to be visually identified \end_layout \end_deeper \begin_layout Enumerate The UUV should be able to re-bury the cable in shallower waters \end_layout \begin_deeper \begin_layout Enumerate This should provide more protection to the cable from interference including fishing operations \end_layout \end_deeper \begin_layout Subsubsection Current AUV Usage \end_layout \begin_layout Standard Autonomous underwater vehicles are well suited to survey and research operations ; without human intervention they sweep a given area collecting data for analysis. This can include bathymetry \begin_inset Foot status open \begin_layout Plain Layout Measurement of the depth of a body of water \end_layout \end_inset , surveys and chemical composition investigations such as pH and toxin levels. \end_layout \begin_layout Standard \begin_inset Float figure wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Graphics filename hugin-superior.jpg lyxscale 30 width 60col% \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout Kongsberg Maritime's HUGIN Superior AUV, \begin_inset CommandInset citation LatexCommand citep key "auv-hugin-superior-datasheet" literal "false" \end_inset \end_layout \end_inset \end_layout \begin_layout Plain Layout \end_layout \end_inset \end_layout \begin_layout Subsubsection Advantages \end_layout \begin_layout Standard An advantage of using an autonomous vehicle would be the lack of need for the surface vessel to maintain position directly above the ROV and fault; instead the surface vessel would stay within a larger area only to maintain contact with the UUV. This could reduce the required power directed to dynamic positioning which in higher sea states can become a significant draw. Additionally, as the UUV can move independently, the surface vehicle would not need to directly track the vehicles movement; for example, when the UUV is re-burying the repaired cable in shallower waters. This would, again, lower the required propulsion power used by the surface vessel. \end_layout \begin_layout Standard Another advantage could be a reduction in risk during mission operations. With a traditional tethered ROV, should the umbilical cable be broken the vehicle would likely lose functionality and require specialist recovery. This break could occur as a result of a fault in the tether management system, high storm activity causing too much tension on the system, or in less likely scenarios, animal intervention. An autonomous vehicle has no tether to break and a hybrid ROV/AUV could likely be instructed to take control and return home should the tether break during missions involving direct human control \begin_inset Flex TODO Note (Margin) status open \begin_layout Plain Layout No longer valid, battery is removable when using TMS \end_layout \end_inset . \end_layout \begin_layout Subsubsection Domain Challenges \end_layout \begin_layout Paragraph Navigation \end_layout \begin_layout Standard As mentioned, one of the main advantages of using an autonomous vehicle for sub-sea cable repairs would be the physical de-coupling of the vehicles, however this also poses the most significant challenge. In typical ROV operations, the operator has knowledge of the location of the ROV relative to the surface vessel. As the surface vessel is GNSS \begin_inset Foot status open \begin_layout Plain Layout Global Navigation Satellite System, the generic term for satellite aided global navigation of which the American GPS, Russian GLONASS and European Galileo systems are examples \end_layout \end_inset -enabled (Likely GPS) it has knowledge of its position in world co-ordinates and the operator can use this to reduce the ROV's cable search space. \begin_inset Flex TODO Note (Margin) status open \begin_layout Plain Layout Diagram? \end_layout \end_inset \end_layout \begin_layout Standard Decoupling the vehicles introduces complications that are not necessarily typical to the existing use cases for AUVs. The frequency of EM waves used by GNSS systems do not penetrate deep through the water and an AUV must be able to operate without world co-ordinates provided in this manner. As such, navigation systems used by AUVs are typically \emph on dead reckoning \emph default systems. This is a form of navigation that operates relative to a known fixed point (where a UUV is deployed for example) as opposed to one relative to world co-ordinates. \end_layout \begin_layout Standard With an accurate system, this will satisfy many surveying and research use cases where relative location data can be transformed to world-coordinates after the fact. This will prove less effective when the vehicle is expected to autonomously navigate to a specific location (the cable fault). A dead reckoning system as described above uses relative sensors to measure speed and infer the current location however these relative sensors have associated measurement errors which accumulate over time. This would be more pronounced under the water where sea currents are liable to accentuate these errors, the efficacy of an AUV's fault location capabilitie s may be reduced to the point of unacceptability. \begin_inset Flex TODO Note (Margin) status open \begin_layout Plain Layout Kalman filter now? \end_layout \end_inset \end_layout \begin_layout Paragraph Launch & Recovery \end_layout \begin_layout Standard By allowing the UUV to operate untethered underwater, complications are introduced to the method by which it is launched and recovered to the surface vessel. A tethered ROV of the size being considered typically has a top-hat tether management system (TMS) responsible for controlling the slack in the umbilical cable, an example can be seen in figure \begin_inset CommandInset ref LatexCommand ref reference "fig:top-hat-tether" plural "false" caps "false" noprefix "false" \end_inset . Smaller vehicles can also use a garage-style TMS where the tether attaches to a box-like cradle that houses the UUV within. \end_layout \begin_layout Standard \begin_inset Float figure wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Graphics filename tophat.png lyxscale 50 width 40col% \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout A top-hat tether management system attached to the top of an ROV, \begin_inset CommandInset citation LatexCommand cite key "tophat-tms" literal "false" \end_inset \begin_inset CommandInset label LatexCommand label name "fig:top-hat-tether" \end_inset \end_layout \end_inset \end_layout \end_inset \end_layout \begin_layout Standard This results in the UUV being under control by the surface vessel as it is being lifted from the water, especially during the \emph on splash zone \emph default , the area surrounding the average water level. As the UUV is lifted or dropped through this area and it beings to make contact with the water, the weight load on the crane can change dramatically, \begin_inset CommandInset citation LatexCommand cite key "splash-zone" literal "false" \end_inset . During this period, the UUV is most at risk of damage as wind and sea forces can make it swing towards the surface vessel. \end_layout \begin_layout Standard The TMS and LARS system together aim to protect both vessels as the UUV descends through the splash zone by dampening the lateral movement of the UUV and by limiting the amount of umbilical slack; this keeps it away from the surface vessel's thrusters, \begin_inset CommandInset citation LatexCommand cite key "rov-tms-splash-zone" literal "false" \end_inset . \end_layout \begin_layout Standard These methods are effective in protecting tethered UUVs during launch and recovery, the challenge comes in defining how the UUV will be deployed when operating autonomously without a tether. \end_layout \begin_layout Subsubsection Proposed Design \end_layout \begin_layout Standard The vehicle will be designed for hybrid ROV/AUV operations. The vehicle should be able to complete missions independently of the surface vessel with the ability to operate in a similar fashion to existing ROVs (human controller, tethered power and data connection). This will have a number of benefits, primarily that the vehicle should be able to benefit from autonomous operation where possible with the ability for direct human control in missions deemed too complex for autonomous control. \end_layout \begin_layout Standard The existing remit of AUV operations is primarily survey, inspection and light intervention, it is likely that the autonomous capabilities of this vehicle would not be capable of conducting all existing cable repair missions which involve more intervention. It is important that enabling autonomous operations does not ultimately reduce its operating capabilities. \end_layout \begin_layout Standard As previously described, box frame UUVs are well suited to sub-sea cable operations where fine movement control and space for actuators are critical. As such a box frame of similar specifications to those currently used, \begin_inset CommandInset citation LatexCommand citep key "global-marine-atlas-data-sheet,rov-hector-7-datasheet" literal "false" \end_inset will be used. The vehicle will likely be at the larger and heavier end of existing ROVs as the vehicle must now have the onboard energy capabilities to complete a mission without a constant power supply from the surface vessel. \end_layout \begin_layout Subsubsection Communication \end_layout \begin_layout Standard As the UUV is now expected to operate independently of the surface vessel, it should have the ability to bi-directionally, wirelessly communicate with the surface vessel. Uses for such a communications channel include the UUV reporting its mission status and the surface vessel providing high-level instructions such as \emph on return home \emph default orders. When operating underwater, acoustic signals are the primary medium for wireless communication. \end_layout \begin_layout Subsubsection Navigation \begin_inset CommandInset label LatexCommand label name "subsec:Navigation" \end_inset \end_layout \begin_layout Standard As previously described, the navigation system will primarily be built on the principle of \emph on dead reckoning \emph default using an inertial navigation system (INS). An INS uses input from many types of sensor such as accelerometers and gyroscopes to measure the movement of the vehicle and hence infer its location. None of these could individually provide an accurate determination of location and as such \emph on sensor fusion \emph default is employed. Each sensor has an associated measurement uncertainty which compounds over time, sensor fusion allows all the sensor measurements to be combined in such a way as to produce a single output measurement with an uncertainty smaller than any of each sensor individually. A common method for implementing sensor fusion is using a \emph on Kalman filter \emph default \begin_inset Flex TODO Note (Margin) status open \begin_layout Plain Layout reference, explain? \end_layout \end_inset . \end_layout \begin_layout Standard However, despite the use of a Kalman filter allowing more precise approximations of the vehicle's relative location, the lack of external calibration means that the overall uncertaintiy will still increase over time. In land-based robotics this is mitigated through the use of periodic GPS measurements which have low, constant uncertainty and help to place an upper bound on the overall error. As previously mentioned, GNSS systems do not work deep underwater and as such, another method for providing these external updates must be used. \end_layout \begin_layout Standard The following proposes methods for providing global positioning to the UUV without a traditional GNSS system. This will be completed in two stages, the first being to provide the UUV with the ability to measure the location of a fixed point relative to itself. In parallel, the global co-ordinates of this fixed point will be communicated to the UUV in order to infer its own global location. \end_layout \begin_layout Paragraph Underwater Acoustic Positioning \end_layout \begin_layout Standard Alongside the use of acoustic signals for communications it will also be employed for positioning. One application for this is underwater acoustic positioning which employs the use of time-of-flight measurements to beacons of a known location to triangulate an object's location. There are different configurations for such a system depending on how these beacons are laid out, \emph on long-baseline \emph default (LBL) systems involve beacons located on the sea floor. Spreading these beacons around the working area of an ROV widens the baseline of the system and provides higher accuracy when triangulating. This configuration is best suited to static areas of research such as ship wrecks where an initial time devoted to deploying and calibrating these underwater beacons is a reasonable expense to pay for the required high accuracy. This is not the case for sub-sea cable repairs where the deployment, calibratio n and recovery of beacons on the seabed would be prohibitively complex and add significant time to the duration of a mission. \end_layout \begin_layout Standard \emph on Short-baseline \emph default (SBL) systems involve a number of beacons placed at the furthest corners of the surface vessel, this has the benefit of requiring little set-up and pack-down at the cost of reduced accuracy. Relative to the UUV these beacons are all on a similar bearing when operating at a distance, as a result changes in the vehicle's location would be reflected in similar changes to the measurements from all of the beacons. Previously, with a long-baseline, the beacons are ideally surrounding the UUV's working area and changes in its location are reflected in different distance deltas for each beacon allowing tighter triangulation. Accuracy can be improved by extending the beacons away from the vessel to extend the baseline as far as possible. \begin_inset Flex TODO Note (Margin) status open \begin_layout Plain Layout such as on a boom? \end_layout \end_inset \end_layout \begin_layout Standard One method to mitigate the drawbacks of both described methods is by using GPS Intelligent Buoys (GIBs). This configuration, also referred to as an \emph on inverted long-baseline \emph default , allows a much wider baseline than the surface-vessel-mounted beacons by deploying a group of \emph on smart buoys \emph default around the expected working area of the UUV. The use of buoys as opposed to beacons on the sea-floor significantly decreases the preparation and clean-up mission phases. \end_layout \begin_layout Standard Of these methods it is proposed that the surface vessel be equipped with a short-baseline beacon array as well as a population of GIBs. This will allow the choice between higher accuracy or faster mission turnaround be decided by mission conditions as well as providing redundancy for either system. In shallower waters, the accuracy of the onboard SBL may be deemed sufficient however in deeper water where the UUV is operating far further from the surface vessel, the compactness of the SBL baseline may require the higher accuracy of the GIBs \begin_inset Foot status open \begin_layout Plain Layout In practice the two could be used in conjunction for efficiency. As the UUV is deployed it initially uses the onboard SBL array while the surface vessel makes a pass around the working area deploying GIBs for use as the UUV gets deeper \end_layout \end_inset . The GIBs would be considered additional accuracy, the SBL would be used alongside the GIBs and act as an extra node in the array. Additionally the weather and sea conditions could play a factor in the decision. In higher sea states and stormy weather, the deployment and recovery of GIBs may be deemed too risky and the SBL could be used alone. \end_layout \begin_layout Paragraph Global Calibration \end_layout \begin_layout Standard The above underwater acoustic positioning system will allow the UUV to keep track of its position relative to known points at the surface, however this alone will not provide the UUV with its global location. In order for the UUV to calibrate its local map to global co-ordinates, the global position of these surface points must be provided. This will be conducted over the previously described acoustic communication channel. As it could be expected that this channel has a low bandwidth, these updates need not be excessively frequent. \end_layout \begin_layout Paragraph Acoustic Doppler Current Profiling \end_layout \begin_layout Standard While accelerometers and gyroscopes would be expected components of any mobile dead reckoning navigation system, additional sensors well-suited to sub-sea localisation will allow the vessel's movement to be more precise. One such sensor is a \emph on Doppler velocity log \emph default (DVL) which estimates the vessel's velocity by tracking the seabed. DVLs apply the broader concept of \emph on acoustic Doppler current profiling \emph default which measures the velocity of water by measuring the change in frequency caused by the Doppler effect. Combined with depth measurements calculated from the signal's echo time, this can be used to estimate the vessel's velocity. DVLs are crucial to a sub-sea INS as, like GPS, their error does not grow when employed correctly unlike other relative sensors. As described in section \begin_inset CommandInset ref LatexCommand ref reference "subsec:Navigation" plural "false" caps "false" noprefix "false" \end_inset , a sensor who's measurement error does not compound and grow but stays constant is important as it places an upper bound on the overall error and allows the system to maintain accuracy over time. \end_layout \begin_layout Subsubsection Power \end_layout \begin_layout Standard The ability to operate autonomously without an umbilical cord implies that the UUV must have an onboard power supply. \end_layout \begin_layout Standard As mentioned, much of the vehicle specification is being inherited from existing ROV technology and this would include expected operating power. The expansion of the UUV's capabilities to include autonomous operation would primarily be completed through software and not significantly alter the required power. \end_layout \begin_layout Standard 300 kW was used as the required max power to calculate the energy storage capabilities, an operating time of 10 hours was also defined. An average draw of 50% max power was used to calculate 1.5 MWh of required storage. \end_layout \begin_layout Standard The previously described 18650 cells (section \begin_inset CommandInset ref LatexCommand ref reference "subsec:Proposed-UUV-Battery-Solution" plural "false" caps "false" noprefix "false" \end_inset ) will be used for the UUV's battery pack, this will allow a single process for sourcing and end-of-life processing and increase efficiency by utilising the economy of scale. As such, the previously mentioned notes on sustainability including processes for second-use and recycling would apply to the UUVs battery pack. \begin_inset Flex TODO Note (Margin) status open \begin_layout Plain Layout As described in the sustainability, operating at scale has allowed the carbon cost of cells to go down, this is the same thing \end_layout \end_inset Lithium-polymer batteries have found usage in AUVs as a result of their lighter weight than Lithium-ion batteries. While this will increase efficiency, it is proposed that the use of a single supply chain will improve sustainability, a key parameter for this project. \end_layout \begin_layout Standard The cell voltage (3.6 V) and capacity (3.5 Ah) were multiplied for 12.6 Wh of power capacity per cell. This would require 119,048 cells to meet the capacity requirements. \end_layout \begin_layout Standard The battery system constitutes an extra 5,700 kg of extra weight for the UUV, it is important that the battery be removable for tethered operation in order to increase efficiency when independent operation is not required. \end_layout \begin_layout Standard As this battery array will experience less usage than the surface vessel's set, it is expected that it will last longer. Despite undergoing less charge cycles, Lithium-ion cells still have a finite lifespan and would not be expected to last beyond 4 years. As such this is defined as the battery replacement time period. \end_layout \begin_layout Subsubsection LARS \end_layout \begin_layout Standard The proposal for hybrid AUV/ROV capabilities somewhat simplifies the required LARS methods. As the vehicle must still be able to operate while tethered, from the surface vessel's perspective a traditional LARS system with a top-hat TMS will be used. When operating as an ROV, the underwater vehicle will not have the battery system mounted on the top and the tether will directly connect to the vehicle. In these scenarios the LARS processes will be as previously described, see figure \begin_inset CommandInset ref LatexCommand ref reference "fig:tethered tms" plural "false" caps "false" noprefix "false" \end_inset . \end_layout \begin_layout Standard \begin_inset Float figure wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Graphics filename tms-lars.png lyxscale 50 width 40col% \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout TMS and UUV structure when operating tethered \begin_inset CommandInset label LatexCommand label name "fig:tethered tms" \end_inset \end_layout \end_inset \end_layout \begin_layout Plain Layout \end_layout \end_inset \end_layout \begin_layout Standard When operating autonomously, the battery system will be mounted on the top of the vehicle. It is proposed that the battery system have an interface for the top-hat TMS on the top, see figure \begin_inset CommandInset ref LatexCommand ref reference "fig:untethered tms" plural "false" caps "false" noprefix "false" \end_inset . An acoustic location and communication beacon will be mounted on the underside of the TMS in order to provide a reference to the UUV for navigating towards and docking, \begin_inset CommandInset citation LatexCommand cite key "hugin-lars-article" literal "false" \end_inset . \end_layout \begin_layout Standard \begin_inset Float figure wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Graphics filename tms-lars-auv.png lyxscale 50 width 30col% \end_inset \begin_inset Graphics filename tms-lars-disconnected.png lyxscale 50 width 50col% \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout TMS and UUV structure when operating autonomously, indicating mount point for re-housing \begin_inset CommandInset label LatexCommand label name "fig:untethered tms" \end_inset \end_layout \end_inset \end_layout \end_inset \end_layout \begin_layout Standard This allows the actual LARS processes to be conducted in the traditional secure manor through the splash zone and then allow detachment for autonomous operation when safely under the water. \end_layout \begin_layout Subsubsection Sustainability \end_layout \begin_layout Standard With regards to the battery system, the same principles as described for the surface vessel battery in sections \begin_inset CommandInset ref LatexCommand ref reference "subsec:Life-cycle-Analysis" plural "false" caps "false" noprefix "false" \end_inset and \begin_inset CommandInset ref LatexCommand ref reference "subsec:SurfaceVesselBatterySustainability" plural "false" caps "false" noprefix "false" \end_inset apply including procurement, and lifespan extension. \end_layout \begin_layout Standard However, the difference in usage patterns for the UUV may make the battery's applicability to second-usage less viable. This would be because it would be expected to sustain higher depth-of-discharge without the ability to activate more ammonia cells and keep the capacity higher. As such the direct recycling processes previously described may be more appropriate as an end-of-life procedure. \end_layout \begin_layout Standard Using the previous per kilowatt embodied carbon value, a UUV battery set represents 120 tonnes of carbon requiring offset to achieve net-zero. \end_layout \begin_layout Subsubsection Financials \end_layout \begin_layout Standard UUVs and TMSs are typically bespoke, purpose-built projects without publicly available prices. As such estimations were made in order to cost the entire system. UUVs with publicly available prices tend to be light intervention and autonomou s survey vehicles with a max depth of 300m, \begin_inset CommandInset citation LatexCommand cite key "price-amron-rov,price-deep-trekker" literal "false" \end_inset . These vehicles tend to cost between £24,000 and £50,000. The proposed vehicle is far more powerful and is not the same class of vehicle. \end_layout \begin_layout Standard Including the advancements of autonomous capabilities it is proposed that an industrial vehicle of the requirements described would cost £2,000,000. Additionally it is proposed that the tether management system would cost £400,000. \end_layout \begin_layout Standard As described, a single type of Lithium cell are being used in order to benefit from the economy of scale. Using the previously specified £5 unit price, a set of UUV battery cells will cost £595,240. Modelling this across its lifespan averages to £148,810 per year. For costing the associated BMS, the previously mentioned 60% cell cost contribution would suggest a price of £397,000. \end_layout \begin_layout Subsubsection Summary \end_layout \begin_layout Standard The proposed UUV describes an extension to existing ROV capabilities by allowing untethered autonomous operations. This requires a wireless communication channel between the surface and underwater vessel which will be completed using acoustic waves. The UUV will effectively be GNSS-enabled by proxy from the surface vessel, allowing it to navigate to the fault locations. In order to operate autonomously, the UUV will require onboard power, a battery system of suitable scale was described along with protocols for decommissioning sustainably. \end_layout \begin_layout Part Digitalisation \end_layout \begin_layout Standard Before discussing how this project aims to leverage \emph on digitalisation \emph default it is worth defining the term and the adjacent term \emph on digitisation \emph default . Digitisation describes the transforming of data or a process from an analogue system to a digital one, \begin_inset CommandInset citation LatexCommand citep key "workingmouse-digitalisation" literal "false" \end_inset . It is a value-neutral term in of itself and could have positive or negative effects. A simple example would be transitioning from working with pen-and-paper forms to digital documents and PDFs. \end_layout \begin_layout Standard Digitalisation describes the use of digitisation to increase efficiency and access new value-producing business opportunities, \begin_inset CommandInset citation LatexCommand citep key "workingmouse-digitalisation,gartner-digitalization" literal "false" \end_inset . To follow the above example, digitalisation could include using groupware in order to collaboratively work on a cloud document as opposed to delivering hard copy revisions of a document between locations. \end_layout \begin_layout Standard As a broad concept, there are many ways that the concept of digitalisation could be applied to this project, as a whole, though, the initiative could be described as a \emph on smart ship \emph default . Many of the features rely on interconnected sites, the internet network topology can be seen in figure \begin_inset CommandInset ref LatexCommand ref reference "fig:Network-topology" plural "false" caps "false" noprefix "false" \end_inset . \end_layout \begin_layout Standard The surface vessel will be connected to the internet via two gateways. While berthed, the vessel should be able to connect to the depot via Ethernet which can be run alongside the shore power line. For internet connectivity while at sea, the vessel will be equipped with satellite internet apparatus. \end_layout \begin_layout Standard \begin_inset Float figure wide false sideways false status open \begin_layout Plain Layout \noindent \align center \begin_inset Graphics filename ../network/NetworkDiagramJointDepot.png lyxscale 30 width 75col% \end_inset \end_layout \begin_layout Plain Layout \begin_inset Caption Standard \begin_layout Plain Layout Network topology across the depot, vessel and cloud environment; main services highlighted \begin_inset CommandInset label LatexCommand label name "fig:Network-topology" \end_inset \end_layout \end_inset \end_layout \end_inset \end_layout \begin_layout Standard With a fully connected environment, head-office and the depot will be able to monitor and control aspects of the vessel. In theory, head-office would be able to remotely command both the vessel and UUV, although proper authorisation, safety and business practices would need to be defined for where this is appropriate. \end_layout \begin_layout Standard Live access will be available for information about the vessels including location data, course information, battery capacity information and remaining fuel levels. This could be used in order to sync depot operations with the vessels mission status, for example by preparing the local electrical supply for cold-ironing the vessel. \end_layout \begin_layout Standard Within the vessels, machine learning (ML) and AI will have varying applicability. The UUV, for example, would likely apply both for applications such as image recognition from the visual cameras. Kalman filters have already been discussed for calculating the vessel's location. \end_layout \begin_layout Standard \begin_inset Newpage newpage \end_inset \end_layout \begin_layout Standard \begin_inset CommandInset bibtex LatexCommand bibtex btprint "btPrintCited" bibfiles "references" options "bibtotoc" \end_inset \end_layout \begin_layout Standard \begin_inset Newpage newpage \end_inset \end_layout \begin_layout Section \start_of_appendix Battery Cell Calculations \begin_inset CommandInset label LatexCommand label name "sec:Battery-Cell-Calculations" \end_inset \end_layout \begin_layout Standard \begin_inset Flex TODO Note (inline) status open \begin_layout Plain Layout C rates \end_layout \end_inset \end_layout \begin_layout Section Time-dependent Power Modelling \begin_inset CommandInset label LatexCommand label name "sec:Time-dependent-Power-Modelling" \end_inset \end_layout \begin_layout Subsection Power In \end_layout \begin_layout Standard As described in the premise for why a buffer is needed (section \begin_inset CommandInset ref LatexCommand ref reference "sec:Energy-Storage" plural "false" caps "false" noprefix "false" \end_inset ), the power from the ammonia fuel cells will primarily be varied by changing the population of active cells as opposed to drawing varied power from each individually. This allows the cells to operate as much as possible in their most efficient state. However, cells require time to turn on and reach this efficient state, about 20 minutes. This was modelled by having the input power move in discrete steps as cells are turned on and off. This step value was defined as 200kW, the most efficient state for a single fuel cell. \end_layout \begin_layout Standard As time increases, every twenty minutes the amount of active cells can be incremented or decremented. This over-simplifies the actual behaviour as this would, in reality, be a gradual process as opposed to one of discrete steps, however, it was deemed acceptable in order to enforce the time penalty in changing the number of powering cells. \end_layout \begin_layout Subsection Power Out \end_layout \begin_layout Standard In order to model the load draw from the propulsion and hotel load, a random differential was applied each second. This was done in order to provide a dynamic environment, were the load power to stay the same the battery would either charge or discharge entirely and then stay in this state. A random change each second more closely matches the expected power requirement s as the wind and currents are also dynamic. \end_layout \begin_layout Standard The max load differential was defined as 10 kW. This means that each second the load could change by a maximum of \begin_inset Formula $\pm$ \end_inset 10kW as a random number between -1 and 1 was generated and used as a coefficient. \end_layout \begin_layout Standard The different stages of a mission were defined as having a maximum and minimum load power which the random function was able to fluctuate between. When dynamic positioning it could be expected that more power would be used than when completing either the out or home-bound journey. \end_layout \begin_layout Subsection Efficiencies \end_layout \begin_layout Standard The charging and discharging of the battery is not a completely efficient process. In order to take this into account, both processes were treated as 80% efficient. \end_layout \begin_layout Subsection Validity \end_layout \begin_layout Standard In terms of applicability, the model provides a good high-level approximation of the relationship between the fuel cells and the battery usage. In reality, however, the system would be far more complex. For example, the model only increments or decrements the active fuel cells by one at each twenty minute interval when in reality many could be activated or deactivated simultaneously. \end_layout \end_body \end_document