Method of operating an electrochemical cell system

Abstract

A method of operating an electrochemical cell system. The system comprising a plurality of electrochemical cell arrays, each said electrochemical cell array having an expected array lifetime divided into a set of predefined operational time intervals, each said operational time interval having an associated array replacement rule and system control strategy. The method comprises: performing at least periodic detections for a failure condition amongst the plurality of electrochemical cell arrays; upon detecting a failure condition in an electrochemical cell array, determining in which of said operational time intervals the electrochemical cell array having the failure condition is in; and for the electrochemical cell array having the failure condition, following the associated array replacement rule and system control strategy for that operational time interval.

Description

[0001] PN853570GB 1

[0002] METHOD OF OPERATING AN ELECTROCHEMICAL CELL SYSTEM

[0003] Field of the Invention

[0004] The present invention relates to an electrochemical cell system comprising a plurality of electrochemical cell units, and the operation thereof. The electrochemical cell system may be an electrolyser system or a fuel cell system.

[0005] Background to the Invention

[0006] Electrochemical cell units are commonly referred to as fuel cell units or electrolyser cell units, and in some instances their names are interchangeable as some fuel cell units can work as electrolyser cell units and some electrolyser cell units can operate as fuel cell units, each either as a producer of electricity or in a regenerative mode - electrolyzing a fluid to electrochemically split it into two or more component parts. For example, some fuel cell units can produce electricity by using an electrochemical conversion process that oxidises fuel to produce electricity. Some fuel cell units can also, or instead, operate as regenerative fuel cells (or reverse fuel cells) units, often known as electrolyser cell units, for example to separate hydrogen and oxygen from water, carbon monoxide and oxygen from carbon dioxide, or nitrogen monoxide and oxygen from nitrogen dioxide.

[0007] Electrochemical cell units may be tubular or planar in configuration. Planar cell units may be arranged overlying one another in a stack arrangement, for example 100-400 cell units in a stack, with the individual fuel cell units arranged, for example, electrically in series. Tubular cell units may be arranged in groups, stacks or coils thereof.

[0008] A solid oxide fuel cell (SOFC) unit that produces electricity is based upon a solid oxide electrolyte that conducts negative oxygen ions from an oxygen electrode to a fuel electrode located on opposite sides of the electrolyte. For this, a fuel, or reformed fuel, contacts the fuel electrode and an oxidant, such as air or an oxygen rich fluid, contacts the air electrode. A solid oxide electrolyser cell (SOEC) may have the same structure as an SOFC but is essentially that SOFC operating in reverse, or in its regenerative mode, to achieve the electrolysis of fuel, for example water and / or carbon dioxide, by input of electrical energy and using the solid oxide electrolyte to produce hydrogen gas and / or carbon monoxide and oxygen. Conventional ceramic-supported (e.g. fuel electrode-supported) SOFCs and SOECs have low mechanical strength and are vulnerable to fracture. Hence, metal-supported SOFCs and SOECs have been developed which have the active cell component layer supported on a metal substrate. In these cells, the ceramic layers can be very thin since they only perform an electrochemical function: that is to say, the ceramic layers are not self-supporting but rather are thin coatings / films laid down on and supported by the metal substrate. Such metal-supported SOEC stacks are more robust, lower cost, have better thermal properties than ceramic-supported SOECs and can be manufactured using conventional metal welding techniques.

[0009] Electrolyser cell stacks commonly operate at elevated temperatures. For example, intermediate or high temperature electrolysers (such as SOEC and alkaline electrolyte cells) have operational

[0010] 15188541-1 PN853570GB 2 temperatures in excess of 400 °C, typically 450 °C to 700 °C for an intermediate temperature electrolyser such as one based on an MS-SOEC, and above 700 °C for high temperature electrolysers.

[0011] Electrolyser cell units, or more usually stacks thereof, can be assembled within a housing, and can by combined with heat exchangers and fluid delivery pipework, amongst other equipment, to form an electrolyser system, and the electrolyser system will be configured to supply input gas to and exhaust off-gas from the electrolyser cell stack(s) at the required operational temperatures. Fuel cell units can be similarly configured within a fuel cell system, and supplied fuel, with off-gas then being exhausted.

[0012] It is known that electrochemical cell systems can suffer component failures and / or degradation. Cell units and stacks can fail or deteriorate during their working lives (i.e. prior to their expected lifetime or planned obsolescence) for many reasons, for example because of manufacturing tolerances or defects, fluid contamination within the system, or use of parts of the system outside given temperature or pressure tolerances, or even simply due to the gradual deterioration of the electrochemically active compounds within the system. Such systems however typically have high capital outlay, and in order to be economically viable must maintain a minimum output performance over their lifetime. Operators of these systems will thus inevitably want to service or repair these systems. However, such service or maintenance will generally involve shutting down the entire electrochemical cell system, and then cooling and disconnecting electrically and fluidically the electrochemical cell array, the cell units, the stack or the stacks that are faulty, and then replacing them (or repairing them if possible). The system can then be reconnected and ramped up to an operating temperature again to allow recommencement of operations. These steps thus lead to significant downtime, and loss of production, and may not be optimal in all circumstances. It is therefore difficult to deal with component failure of such a system, while minimizing the economic impact.

[0013] It would be desirable to provide a method of operating an electrochemical cell system that can help to reduce or overcome one or more of these issues.

[0014] SUMMARY OF THE INVENTION

[0015] According to a first aspect of the present invention there is provided a method of operating an electrochemical cell system comprising a plurality of electrochemical cell arrays, each said electrochemical cell array having an expected array lifetime divided into a set of predefined operational time intervals, each said operational time interval having an associated array replacement rule and system control strategy, the method comprising: performing at least periodic detections for a failure condition amongst the plurality of electrochemical cell arrays; upon detecting a failure condition in an electrochemical cell array, determining in which of said operational time intervals the electrochemical cell array having the failure condition is in; and for the electrochemical cell array having the failure condition, following the associated array replacement rule and system control strategy for that operational time interval.

[0016] 15188541-1 PN853570GB 3

[0017] As noted above, a failure or a significant degradation of one or more of the cell arrays before the planned end of life of the system will impact performance of the system, but maintenance or replacement of components within the system typically results in extensive downtime of all of the cell arrays in the system, making determining the best course of action for any detected such failure condition a potentially difficult process. Aspects of the present invention allow an optimised or preferred course of action to be determined quickly and easily, without requiring expert analysis or calculation at the time of detecting the failure condition, or during the subsequent maintenance / rectification process.

[0018] In some embodiments, the method claim including an initial step of calculating or determining time periods based on stack performance criteria, and / or operating limits of the electrochemical cell arrays, and / or operational conditions of the electrochemical cell arrays.

[0019] In some embodiments, the step of determining in which of said operational time intervals the electrochemical cell array having the failure condition is in is based on at least one of a) stack performance criteria, b) operating limits or c) array conditions.

[0020] In some embodiments, a step of dividing the expected array lifetime into the set of predefined operational time intervals is based on at least one of a) stack performance criteria, b) operating limits or c) array conditions.

[0021] In some embodiments, the electrochemical arrays within the system operate with a common fuel or steam supply.

[0022] In some embodiments, the electrochemical cell arrays within the system all operate with a substantially common temperature and pressure. This may be achieved by using a control device, and optionally by operating them in a common volume of an enclosure. In some embodiments, the common temperature may change over time. For example, during normal operation, the properties of a stack will tend to slowly change. The most significant change is usually that the electrical resistance of the stack increases over time. In some embodiments, to compensate for this, in normal operation a control system may slowly increase an operational temperature of the cell arrays either intermittently or regularly throughout a working life of the system. This can then maintain other criteria or parameters at a constant level - for example, the increase in temperature may reduce the electrical resistance, and thus maintain a constant voltage. This in turn can lead to a consistent hydrogen production rate from an electrolyser system that is converting steam into hydrogen and oxygen

[0023] In some embodiments the electrochemical cell arrays each have independent electrical control.

[0024] In some embodiments, each electrochemical cell array comprises a plurality of electrolyser cell units.

[0025] In some embodiments, each electrochemical cell array comprises at least one stack of electrochemical cell units. In some embodiments, each electrochemical cell array comprises two

[0026] 15188541-1 PN853570GB 4 or more stacks of electrochemical cell units. In some embodiments, each electrochemical cell array comprises at least one pair of stacks of electrochemical cell units.

[0027] In some embodiments, the electrochemical cell arrays are contained within an enclosure.

[0028] In some embodiments, the enclosure is pressurisable - i.e. it is a pressure vessel. This allows the system to operate at elevated pressures (i.e. above environmental pressures surrounding the enclosure). This can improve an efficiency of the electrochemical cell system, and in particular the efficiency of a downstream off-gas delivery system.

[0029] Typically, an environmental pressure surrounding the electrochemical cell system is around 1 bar. However, off-gases from an electrochemical cell system - particularly from an electrolyser system (i.e. off-gases such as hydrogen and oxygen), may want to be collected and stored at a higher than ambient pressure. Therefore, by having the electrochemical cell system operating at a pressure higher than ambient (i.e. higher than that of the environment surrounding the enclosure), the off-gas can also be exhausted from the system at that higher pressure. This in turn means that downstream equipment will have less work to do to recondition the off-gases to storage pressures.

[0030] In some embodiments, the array replacement rule is whether or not to replace or repair the electrochemical cell array having the failure condition. Another array replacement rule may be whether or not to replace or repair a failed cell unit, a failed stack, or a failed pair of stacks, or more than 2 stacks, within the electrochemical cell array having the failure condition, and this may be the same or different dependent upon whether one or both or all cell units or stacks within the electrochemical cell array has failed.

[0031] In some embodiments, the system control strategy involves controlling certain parameters of the system. For example, the system may monitor steam or fuel utilisation within the system and control them within limits set by the associated system control strategy. For example, this may be a percentage of water converted to steam.

[0032] In some embodiments, the system may monitor and control steam flow, i.e. a rate of steam provided to the cell arrays within the system. Typically an electrochemical cell system in the form of an electrolyser system is designed to provide (as close as possible to) even distribution of steam to all stacks in the system, and this can be controlled by manifolding. However, upon detection of a failure condition, the method may have an option within its system control strategy to vary steam supply to respective cell arrays, or to different stacks or pairs / sets / groups thereof. In some embodiments, this may be by including and / or adjusting valves or orifice plates within the system, for example if / when a stack is replaced or when it need to be shut down to remove it from operation.

[0033] In some embodiments, steam flow to a cell array undergoing maintenance, or with a failure condition, could be altered (restricted or increased) by using a valve or orifice plate at a steam supply to that cell array.

[0034] 15188541-1 PN853570GB 5

[0035] In some embodiments, either or both a current and a voltage, but more usually a voltage, supplied across a cell unit, a stack or a cell array, may be a controlled parameter.

[0036] The current drawn across a cell unit, a stack or a cell array typically depends on an electrical resistance of the cell unit, the stack or the cell array in question (which in turn may be temperature dependent). In some embodiments, therefore, the temperature of a cell unit, a stack or a cell array may be controlled for controlling the electrical resistance and thus the current. This may be done using heat exchangers, heaters, or the like for controlling the temperature of fluids supplied to or contained within the system - be that a cathode side, an anode side or a common volume of an enclosure surrounding the cell array(s). As such, in some embodiments, the temperature is controlled by varying temperature of input fluids (e.g. steam and optionally air / oxygen / sweep gas in an electrolyser system). However, in some embodiments, as stack operation may be endothermic or exothermic depending on voltage, current or steam utilisation, temperature control may be performed by controlling the voltage, the current or the steam utilisation of the system.

[0037] The present inventors have found that it is possible to provide an optimised (for a given set of optimisation criteria) failure response strategy for a electrochemical cell system by dividing the lifespan of the system into a number of time intervals, and pre-assigning a maintenance operation depending on when (i.e. in which of these time interval) the failure occurs.

[0038] In embodiments there are at least 3 operational time intervals. These may typically be a start of life interval, from the start of operation of the system to a given number of hours (or equivalent) of use, an end of life interval a given number of hours (or equivalent) prior to the expected or designed end of operational life of the array, and one or more mid life intervals between the start and end intervals. However, more than three, or other numbers of intervals are possible.

[0039] The intervals may be calculated or determined for a given system in advance, taking into account multiple factors and characteristics of the system and its particular operating method / regime / application, as described below. Such calculation or analysis may give rise to the appropriate number of intervals required or selected for a given system.

[0040] In some embodiments, a first associated array replacement rule is to not replace a faulty cell unit, stack or cell array, but to instead leave it in the system and to take it out of service. This may be by electrical and / or fluidic isolation. In an electrolyser system, this would reduce the total hydrogen output of the system, so to compensate, in some embodiments, an accompanying first system control strategy comprises a control instruction to drive remaining cell units, stacks or cell arrays harder to compensate for the reduced output. In some embodiments, this can increase production from the remaining cell units, stacks or cell arrays so as to maintain the previously attained production rate, i.e. the same hydrogen output as before the fault was detected. In some embodiments, the remaining cell units, stacks or cell arrays are driven harder by increasing a current passing through them, e.g. by increasing a voltage across them, or increasing their temperature to reduce their electrical resistance.

[0041] 15188541-1 PN853570GB 6

[0042] In some embodiments, the remaining cell units, stacks or cell arrays are driven harder by increasing steam flow to them (either instead of or in addition to changing the current and voltage).

[0043] In some embodiments, a second associated array replacement rule is to replace the faulty cell unit, stack or cell array.

[0044] Whether replacing or keeping the faulty cell unit, stack or cell array, a second system control strategy may be simply to leave the operating conditions the same as before the replacement (e.g. the same temperature and voltage). However, because the replacement unit, stack or array will typically be new, there will be a difference in "age" compared to the other units / stacks / arrays in the system, i.e. it will be at a different number of hours of use, which difference depends on how long the system has been operating when the replacement occurs. As a result of this age difference the new cell unit, stack or cell array will have a lower electrical resistance than the older cell units, stacks or cell arrays (i.e. the ones that are not determined to be faulty, and thus the ones not being replaced). As a result the new cell units, stacks or cell arrays will draw more current for a given drive voltage than the older cell units, stacks or cell arrays. This will cause the new cell unit, stack or cell array to work harder than the older ones, giving a higher steam utilisation than the faulty cell unit, stack or cell array that it replaced. The greater the difference in age, the greater will be the increase in steam utilisation for the replacement cell unit, stack or cell array if operating at the operating conditions that were in use before the replacement.

[0045] It is to be noted, however, that in some embodiments there will be a maximum operating limit of steam utilisation in an electrolyser system for any given cell unit, stack or cell array, and therefore simply continuing to operate a system with a mixture of new and old cell units, stacks or cell arrays at the operating conditions that were in use before the replacement may not be possible or desirable. For example, it can become undesirable to utilise this strategy after a certain amount of time since the beginning of operation of the system, since by then the temperature for the system may have increased too much since the beginning for compensating for the natural degradation of the electrochemistry over time.

[0046] According to a third system control strategy, therefore, for the same second associated array replacement rule (i.e. to replace the faulty cell unit, stack or cell array), there may be a control instruction, after replacement, to set the conditions of the system to a compromise between that which best suits the new stack(s) and that which best suits the existing stacks. In some embodiments this may be a condition that maintains the previous rate of hydrogen production from the system, but more usually it will slightly reduce the hydrogen output rate.

[0047] In this condition, in some embodiments the operational temperature of the system is lower than prior to the replacement. This results in a higher electrical resistance for all cell units, stacks and cell arrays. This in turn protects the system somewhat against the new cell unit, stack or cell array exceeding the steam utilisation limit, but it also means that the existing stacks will be driven at a lower current than before, i.e. less hard, resulting in a the reduction in the rate of hydrogen production for an electrolyser system.

[0048] 15188541-1 PN853570GB 7

[0049] In some embodiments, in order to counter the reduction in total hydrogen production, the steam flow to the system can be increased. In some embodiments, this is done with some steam balancing between stacks within the system to keep the steam flow to the new cell unit, stack or cell array lower than the steam flow to the old cell units, stacks or cell arrays. In some embodiments, this would be by the use of calibrated orifice plates or valves connected to stack inlets or manifolds therefor.

[0050] In some situations, a new stack current could exceed an upper working limit of the stack (e.g. if it is running at a temperature higher than it would usually run, thus reducing its electrical resistance). This may be if the temperature or voltage for the system has increased too much since the beginning for compensating for the natural degradation of the electrochemistry over time, and therefore it may not be possible or desirable to use this third system control strategy after a certain period of time since the beginning. According to some embodiments, therefore there can be a fourth system control strategy, where upon replacing the faulty cell unit, stack or cell array in accordance with the second associated array replacement rule, the system is run at the same operating conditions it would have been operating at before the replacement, but for the replacement cell unit, stack or cell array, which replacement cell unit, stack or cell array instead is electrically controlled to be at a different voltage and / or current to the other cell units, stacks or cell arrays.

[0051] In this fourth system control strategy, the new cell unit, stack or cell array is provided with a lower current, which prevents the new cell unit, stack or cell array from reaching or exceeding a safe steam utilisation limit. In this arrangement, a trade-off can be that the new cell unit, stack or cell array will be running slightly endothermically, whereas usually it is preferred that cell units, stacks or cell arrays run at a thermoneutral configuration. As a consequence the new cell unit, stack or cell array may suffer from a higher degradation rate than if operated thermoneutrally.

[0052] In some embodiments, the electrochemical cell arrays comprise electrochemical cell units of solid oxide or molten carbonate cells. In the case of solid oxide cells, they may be metal-supported solid oxide cells.

[0053] In some embodiments, the electrochemical cell units may be electrolyser cell units, and they may be based on a solid oxide electrolyte and so are solid oxide electrolyser cells (SOEC). The electrolyser cell units may be metal-supported electrolyser cells (e.g., MS-SOEC), which aids stability of said cell units.

[0054] In some embodiments the electrochemical cell units operate at a target operational temperature in excess of 400 °C (e.g., 400-800 °C, optionally 450-700 °C). The temperatures referred to may be a temperature of the stack(s) - for example, the temperature of the product or off-gas (from the fuel or oxygen volumes of the stack(s)) may be used as the temperature of the stack(s) or the temperature of the stack(s) derived therefrom.

[0055] In some embodiments, each electrochemical cell array may comprise a plurality of stacks of electrochemical cell units arranged in sets of at least 2 stacks, for example in pairs, or sets of 4 stacks or sets of 6 stacks, optionally sets of at least 10 stacks, and optionally sets of at least 20 stacks. For example, if supplied in pairs that are arranged side by side, the stacks may be arranged

[0056] 15188541-1 PN853570GB 8 as a plurality of sets of stacks that are arranged two wide and in groups of 1, 2, 3, 5 or 10 rows, each group defining an electrochemical cell array. In some embodiments, the electrochemical cell arrays may instead be three, four or more than four wide. For example an electrochemical cell array may feature a four wide, six row, set of stacks, and thus totalling 24 stacks. Instead there may be 2, 3, 4, 5 or 7 (or more) rows in each group.

[0057] In some embodiments, an enclosure contains at least three groups, and thus at least three electrochemical cell arrays.

[0058] In some embodiments, a control system is provided that is configured to regulate (e.g. using back pressure regulation) a pressure of at least one or both of a fuel volume and an oxygen volume of all stacks within the electrochemical cell arrays, and optionally also to regulate pressure in a common volume surrounding the electrochemical cell arrays, for example independently or via fluidic communication between the common volume and one of the fuel and oxygen volumes. The regulated pressure is typically at a pressure greater than ambient pressure.

[0059] In some embodiments, the pressure of at least one or both of the fuel volume and the oxygen volume (and optionally the common volume) is at least 0.5barg, optionally at least lbarg, optionally at least 1.5barg, optionally between 1.5barg and 3barg, optionally at most 10 barg, optionally at most 5barg. 1 barg is 1 bar above ambient, i.e. typically around 2 bar. When pressurised to a pressure higher than ambient pressure, this elevated pressure can be used to balance fluid pressure within the fuel and oxygen volumes of the stacks, thus reducing the stresses within the internal walls of the cell units / stacks. This is beneficial in particular since electrochemically active materials within the cell units of the stacks can be relatively brittle compared to the metal / steel support layer in the case of a MS-SOEC, and the balanced pressures between the stack and the common volume can additionally de-stress the exterior walls of the stacks.

[0060] According to an aspect there is provided an electrochemical cell system comprising a controller configured to perform the methods discussed herein.

[0061] According to an aspect there is provided a controller configured to perform the methods discussed herein.

[0062] According to an aspect there is provided a computer program comprising instructions which, when the program is executed by a computer, cause the computer to carry out the steps of the method according to the aspects above. In particular, when the program is executed by a computer, cause the computer to: perform at least periodic detections for a failure condition amongst the plurality of electrochemical cell arrays; upon detecting a failure condition in an electrochemical cell array, determine in which of said operational time intervals the electrochemical cell array having the failure condition is in; and for the electrochemical cell array having the failure condition, follow the associated array replacement rule and system control strategy for that operational time interval.

[0063] According to an aspect there is provided a non-transitory computer-readable medium with instructions stored thereon, that when executed by a processor, perform (or cause the processor to perform) the steps of the method of according to the aspects above. In particular, when

[0064] 15188541-1 PN853570GB 9 executed by a processor, perform (or cause the processor to perform): performing at least periodic detections for a failure condition amongst the plurality of electrochemical cell arrays; upon detecting a failure condition in an electrochemical cell array, determining in which of said operational time intervals the electrochemical cell array having the failure condition is in; and for the electrochemical cell array having the failure condition, following the associated array replacement rule and system control strategy for that operational time interval.

[0065] Particular and preferred aspects of the invention are set out in the accompanying independent claims. Combinations of features from the dependent claims may be combined with features of the independent claims as desired and appropriate and not merely as explicitly set out in the claims. The term "comprising" as used herein to specify the inclusion of components also includes examples in which no further components are present.

[0066] An enabling disclosure of the present invention, to one of ordinary skill in the art, is provided herein. Reference now will be made in detail to examples of the invention, one or more examples of which are set forth below. Each example is provided by way of explanation of the invention, and not limitation of the invention.

[0067] Brief Description of the Drawings

[0068] Features of the present invention will now be described in further detail, by way of various embodiments, and just by way of example, with reference to the accompanying drawings (which drawings are not to scale), in which:

[0069] Figs. 1 and 2 are simplified schematic views of an electrolyser system;

[0070] Fig. 3 is a simplified control device for controlling an electrolyser system;

[0071] Figs. 4 and 5 are side elevation and top plan, cut-away schematic views of an electrochemical cell system;

[0072] Fig. 6 is a time graph showing possible time intervals for four possible options for system control after detection of a faulty cell array, cell unit or stack;

[0073] Fig. 7 is a flow diagram setting out an embodiment of a control process; and

[0074] Fig. 8 is a flow diagram setting out an embodiment of a set of four options for the array replacement rules and the system control strategy, one for each of four pre-set operational time intervals during the projected life cycle of the system.

[0075] Detailed Description

[0076] In the following figures and description, like reference numerals will be used for like elements in different figures.

[0077] 15188541-1 PN853570GB 10

[0078] In the following description, the electrochemical cell systems will be described as if they are electrolyser systems (which could be a reversible fuel cell operating in a regenerative mode), for ease of reference. However, it is to be appreciated that the electrochemical cell systems could also be configured as a fuel cell system operating in a power delivery mode.

[0079] Fig. 1 is a simplified schematic of an electrolyser system 100 including a plurality of electrolyser stacks 10 in an enclosure 105. The enclosure 105 may be a pressure vessel 105 configured to withstand a pressure difference between its interior and exterior.

[0080] An array 110 of four stacks 10 is shown in Fig. 1, but it will be understood that any number of stacks 10 may be present, including fewer and greater than four, for example 1, 2, 10, 12 and so forth.

[0081] Each stack 10 comprises a stack of electrolyser cell units. A typical stack may have 100 to 500 electrolyser cell units.

[0082] The electrolyser cell units each comprise a first fluid volume (for a first fluid - typically fuel for the electrolysis process - e.g. at least one of steam, carbon dioxide and nitrogen dioxide) and a second fluid volume (for a second fluid - typically oxygen as a product of the electrolysis process), which fluid volumes are fluidically separated from one another such that the first and second fluids therein cannot mix.

[0083] Each stack 10 has a first fluid inlet 115 and a first fluid outlet 116, each of which is in fluidic communication with the first fluid volumes of each electrolyser cell unit 10. Supply of the first fluid to the first fluid volume is via the first fluid inlet 115 and exhaust from the first fluid volume is by the first fluid outlet 116. The first fluid is provided to the first fluid volume of the stacks 10 from a first fluid source 143 via the respective first fluid inlets 115 of the stacks 10.

[0084] In the example of Fig. 1, the first fluid is fully manifolded within the system. That is, there is a manifold enclosing the fluid within the system and providing fluidic communication between the first fluid source 143, external to the enclosure 105, and the first fluid inlets 115 of the stacks 10. Likewise, there is a manifold providing fluidic communication between the first fluid outlets 116 of the stacks 10 and a first fluid volume off-gas collection 163, external to the enclosure 105.

[0085] Each stack 10 may have a second fluid inlet 122 and a second fluid outlet 123, which, where present, is in fluidic communication with the second fluid volume of each cell unit 10. Exhaust from the second fluid volume is by the second fluid outlet 123, and an optional supply of a second fluid to the second fluid volume is via the optional second fluid inlet 122. The optional second fluid is provided, in this example, as a sweep flow gas to the second fluid volume of the stacks 10 from a second fluid source 144 via respective second fluid inlets 122 to assist in sweeping product (e.g. oxygen) from the cell units.

[0086] In the example of Fig. 1, the second fluid is open manifolded within the system. That is, there is a vessel inlet 120 to the enclosure 105 for delivery of the second fluid to the interior of the

[0087] 15188541-1 PN853570GB 11 enclosure 105 from the second fluid source 144, but there is no branched manifold to deliver the second fluid to respective second fluid inlets 122 of each stack 110.

[0088] Exhaust from the second fluid volume of the stacks 110 is also open manifolded. That is, there is a vessel exhaust 121 from the enclosure 105 for exhaust of second fluid volume off-gas from the interior of the enclosure 105 to the second fluid off-gas collection 164 - in this example the offgas or product, mixed with the second fluid, but there is no branched manifold between the respective second fluid outlets 123 of each stack 110 and the vessel exhaust 121.

[0089] In some cases the second fluid may be partially manifolded within the enclosure - i.e. there may exist a manifold between the vessel inlet 120 and the respective second fluid inlets 122 of the stacks 110, or there may exist a manifold between the respective second fluid outlets 123 of the stacks 110 and the vessel exhaust 121.

[0090] In some cases both the inlet and exhaust of the second fluid may be manifolded, as depicted in Fig. 2, to ensure that any second fluid (herein in this example a sweep gas) that is supplied to the second fluid volume is effective at sweeping through the cell units, such that the product / off-gas released into the second fluid volume can then be removed therefrom. This will occur particularly during start up, shut down and standby modes.

[0091] In Fig. 2 the electrolyser system 101, which is otherwise similar to the electrolyser system 100 of Fig. 1, includes inlet and exhaust manifolds for the second fluid which communicate with the second fluid source 144 and second fluid collection 164, respectively. Further, the electrolyser system 101 has a vessel inlet 120 and a vessel exhaust 121 for a vessel fluid supplied to a common volume of the enclosure 105 that contains the stacks 10, from a common volume supply 165.

[0092] Optionally the vessel fluid can be supplied via a heater 152. Heaters may optionally also be provided for the first and second fluids supplied from the first and second fluid sources 143, 144.

[0093] The vessel fluid may thereafter be exhausted from the common volume to a common volume collection 166 through the vessel exhaust 121. As such, in this embodiment the common volume of the enclosure 105 is in fluid communication with neither the first nor the second fluid volumes.

[0094] When the stacks are operated at elevated pressures, the vessel fluid in the common volume may be regulated to balance the pressure in the common volume with the pressure in the first and / or second fluid volumes. Similarly, the pressures may be balanced between the first and second fluid volumes.

[0095] The vessel fluid may be a relatively inert gas to avoid corrosion of the components within the enclosure, and to minimized the possibility of a reaction between the vessel fluid and any leakage of product (e.g. hydrogen, carbon monoxide, nitrogen or oxygen) or first and second fluids from the manifolded passages elsewhere within the enclosure. For example, the vessel fluid may be air or nitrogen.

[0096] 15188541-1 PN853570GB 12

[0097] It will be appreciated that it is sometimes preferable for the second fluid to be open manifolded, as in Fig. 1, or partially manifolded as discussed above, such that there exists fluidic communication between the second fluid volume of the stacks and the interior of the enclosure 105 for pressure equalisation therebetween - in turn simplifying control strategies and pipework required for the respective fluids.

[0098] It will be understood that similar partial or fully manifolded arrangements to those shown schematically in Figs. 1 and 2 are known in the art, albeit not with the stacks being provided in an enclosure.

[0099] In the examples of Figs. 1 and 2, the first fluid volume is typically a fuel volume and, during steady operation, is for fuel - usually steam and / or carbon dioxide, but possibly other compounds for electrolysing, such as nitrogen dioxide. In cases where the electrolyser cell units in the stacks 10 comprise an oxygen ion conducting electrolyte (e.g., solid oxide electrolyte cell units), the first fluid volume off-gas exhausted to the first fluid off-gas collection 163 comprises hydrogen and / or carbon monoxide and / or nitrogen monoxide (dependent on the fuel used). This off-gas is generated in the electrolysis reaction, and is mixed with unspent fuel in the off-gas. The second fluid volume off-gas exhausted to the second fluid off-gas collection 164 instead comprises oxygen that is generated in the electrolysis reaction. The second fluid off-gas may be pure oxygen or oxygen enriched sweep gas if mixed with a sweep gas.

[0100] The second fluid volume off-gas is preferably controlled through selective use of a sweep gas such that it may comprise at least 50% by weight of oxygen generated by the cell units. In some embodiments, the second fluid volume off-gas 164 may be substantially pure oxygen, if purely the product of the electrolysis process, although it might more normally be oxygen enriched air or nitrogen if a sweep gas (air or nitrogen) is also being used. Ideally, however, the oxygen is at least 90% (by weight) pure oxygen when the system 20 is operating at a steady state operation, as the present invention is ideally operated without an externally sourced sweep gas. In such a configuration the stacks are just supplied the first fluid, i.e. one fluid to the first fluid input 115 for each stack 10.

[0101] Other than the fluid inlet(s) and outlets, the stack / electrolyser / vessel will have inputs (terminals, not shown) for power (for applying a current across the electrolyser cell units in the stacks 110).

[0102] The electrolyser system 100 typically operates at an elevated temperature, for example 400-700 °C for cell units based on a solid oxide electrolyte. It will be appreciated that heat in the first fluid off-gas and the second fluid off-gas will typically be exchanged with (transferred to) the first fluid and, if present, second fluid prior to their delivery to the stacks 10, typically prior to their entry into the enclosure 205 using one or more heat exchangers (but typically two or more heat exchangers such that heat is recovered from the first fluid volume off-gas and the second fluid volume off-gas). In steady state operation said heat exchange may be sufficient to maintain an operating temperature of the electrolyser system 100 when practiced alongside electrical temperature control by varying a power level applied to the electrolyser cell units. Heaters 150 and / or 151 (e.g., an electric heater and / or a trim heater) may be provided in an input stream of the first and / or second fluid, respectively, to provide additional heat to said fluids and for

[0103] 15188541-1 PN853570GB 13 providing additional heat control flexibility within the system. Said heaters may be sized for steady state requirements only, or may also be used for other operational modes, e.g. warm-up or standby.

[0104] In steady state operation, the first and second fluid volume off-gases will usually be at a similar temperature to the operational temperature of the electrolyser cell units. However, a specific delta from the input temperature will depend upon the amount of electrical power supplied to the electrolyser system / stacks / cell units, and the internal electrical resistance of the cell units.

[0105] The electrical power is generally supplied to the stacks with a constant current. The stack is thus operated in galvanostatic conditions. The electrical resistance of the stack thus controls the voltage applied across the stack, and there is thus a variable power draw from each stack and cell unit as the electrical resistance changes. Alternatively the power supplied to the stacks is controlled potentiostatically.

[0106] Operational efficiency can be best improved by reducing the amount of external heat supplied to the system via its fluid temperature control system - i.e. via the heaters. Where that external heat is provided for free - for example as a waste product of another industrial process, then that external heat can be usefully used without cost - i.e. it provides added financial efficiencies. However, if that external heat has an associated cost, then operational efficiencies would be better improved instead by reducing the need for such external heating. At steady state this is achieved by using both galvanostatic conditions within the stacks, i.e. a constant electrical current (constant amps), and by adopting thermoneutral voltages across the stack to avoid heat wastage in the stacks, as at a thermoneutral condition the electrolyser is in an adiabatic state, i.e. it is balanced energetically, which effectively means no heat is consumed or released.

[0107] In the prior art, it is known to use either constant voltages or constant currents on a stack, and then to control the current or the voltage, respectively, to maintain the stack at a substantially thermoneutral condition. This then allows the stack to avoid overcooling or overheating, as when at under-voltage (for a galvanostatic stack - i.e. constant current), the stack shows endothermic characteristics, and it thus cools the fluid (and likewise the operational temperature of the stack), whereby the fluid output temperature is lower than the fluid input temperature, whereas when at over-voltage (for a galvanostatic stack - i.e. constant current), the stack shows exothermic characteristics, and it thus heats the fluid (and likewise the operational temperature of the stack), whereby the fluid output temperature is higher than the fluid input temperature.

[0108] At steady state, the present invention may use a galvanostatic condition for the stack, but fluctuates as necessary between thermoneutral conditions, over-voltage conditions and undervoltage conditions, in response to input fluid temperature control that aims to maintain a fluid input temperature equal to a fluid output temperature. This is done since the present inventors realised that it is relatively straightforward to measure temperature at an inlet and an outlet of the stack and thus to instead control the thermoneutral voltage automatically. In other words, at steady state the temperature delta between the fluid output temperature from the stack and the fluid input temperature for the stack is minimised, aiming for a zero delta. This then allows

[0109] 15188541-1 PN853570GB 14 extended use of a galvanostatic condition for the stack, alongside a thermoneutral voltage condition, even while the stack, or one or more of the cell units therein, degrades.

[0110] Galvanostatic and thermoneutral (and galvanostatic thermoneutral) conditions such as those discussed above apply in steady state when the electrolyser system is at an operational temperature. Other steady state conditions or operating strategies, such as potentiostatic, endothermic or exothermic may also be used depending upon characteristics of the cell units and / or of input and exhaust fluids available or desired, respectively.

[0111] The abovementioned steady state operation is one mode of operation of an electrolyser system. Other modes of operation, including warm up, standby, and shutdown may be used with transitions therebetween. The electrolyser system may transition to a standby mode from a steady state (also referred to as a product-generating mode) or a warm up mode. As used herein, reference to supply / supplying or provide / providing to a volume also involves exhaust from that volume unless the context requires otherwise.

[0112] Referring next to Fig. 3, there is shown a control device 400 for controlling an electrolyser cell stack of an electrolyser in an electrolyser system. The control device 400 comprises an input device 402 for receiving input from sensors 404 so as to determine at least one of: a stack operating voltage across the electrolyser stack(s), an inlet temperature at a fluid inlet, an outlet temperature at a fluid outlet and a common volume temperature. The control device 400 thus comprises a voltage monitoring system 406 for determining a stack operating voltage across the electrolyser stack(s), an inlet temperature monitoring and / or control system 408 for determining an inlet temperature at the fluid inlet and an outlet temperature monitoring and / or control system 410 for determining an outlet temperature at the at least one fluid outlet of the electrolyser stack(s). These systems may utilize sensors and data transmission devices or wiring. The control device receives sensor data relating to each of these measurements. The control device may control the system by controlling valves for supply to and exhaust from each fluid volume and power supplies for non-fluid heating. A suitably programmed processor 412 and associated memory 414 is provided for processing such inputs.

[0113] The control device's inlet temperature monitoring and / or control system 408 may comprise an output device for controlling the identity and temperature of a fluid entering the electrolyser stack(s) at a fluid inlet. An electrical control system 416 is also provided for controlling a current supply to (and in some cases from) the electrolyser stack. As mentioned above, during normal operation this is adapted to provide a constant current to the electrolyser cell stack, but the current / voltage may be reduced or turned off to automatically shut down the stack or increased to warm-up the electrolyser stack(s).

[0114] The control device 400 may also control non-fluid heat sources (heaters and such like described above), recirculation loops, and temporary warm up configurations as described above.

[0115] In such a way, the control device 400 is adapted to control an electrolyser stack(s) and the electrolyser system 100 - for example that shown in any of Figures 4 to 10 as appended hereto.

[0116] 15188541-1 PN853570GB 15

[0117] Referring next to Figs. 4 and 5, an example of an electrochemical cell system in the form of an electrolyser system 200 is shown. In this embodiment, the electrolyser system 200 comprises an electrolyser assembly 12 within an enclosure 14, which enclosure 14 has an inner thermal insulation layer 16 on its inside wall. The thermal insulation layer 16 covers an inner sidewall and inner end wall of the enclosure 14.

[0118] The enclosure 14 has an opening at its end that opposes the inner end wall, which opening is closed by a removable lid 18. The removable lid 18 is also thermally insulated by an insulation layer 20.

[0119] Connecting between stacks 10, there is also provided a manifold 32 for providing fluidic connection to the stacks 10, via the removable lid 18, to the outside of the enclosure 14. In this example, the manifold 32 is shown to connect to the removable lid 18. Connections beyond that connection to external fluid subsystems can be as known in the art.

[0120] As shown in figure 5, the manifold 32 also connects to the stacks 10 via links 36. Such links 36 can be any form of fluidic and / or electrical connection as well known in the art. The links are typically flexible connections to allow relative movement to occur between the manifolds and the stacks, as may occur due to the variable temperatures within the system 200, and the different materials used within the system 200.

[0121] It will be understood manifolds may communicate with more than one stack, if, for example, there are four stacks provided across the width of the enclosure (e.g., a 4 by 4 arrangement, rather than the 2 by 4 arrangement shown in figures 4 and 5).

[0122] Referring to Figure 5, it can be seen that the stacks 10 are arranged in pairs and in lines along the inside of the enclosure 14. This forms an array - 2 by 4 - of stacks within the common volume 24. Other embodiments may have fewer or more rows and columns of stacks.

[0123] Each stack 10 is connected by a link 36 to a manifold 32. In this embodiment, sensors 404 are provided for each link 36. These sensors 404, as per Figure 3, can be for detecting any one or more of temperatures, pressures or flow rates of the fluids passing therethrough, or of the manifolds or links themselves, or any other such parameter of the system 200 that may want to be monitored.

[0124] Whereas only a single sensor 404 is shown for each link 36, there may be separate sensors 404 for each detected parameter, and they may be provided elsewhere along the manifolds 32, along the links 36 or relative to the stacks 10 and / or the cell units therein, either or both on the inlet side and the outlet side.

[0125] Each sensor 404 may also be accompanied by a control flow mechanism, such as valves or orifice plates.

[0126] 15188541-1 PN853570GB 16

[0127] The sensors 404 and the flow control mechanisms may be connected to a control device 400, for example as per Figure 3, to allow sensing and subsequent detection of faulty cell units, stacks or cell arrays, and to control the flow through the respective links 36 (or manifolds 32 or stacks 10).

[0128] In these embodiments the enclosure is shown in a horizontal configuration so that the opening of the enclosure 14 is to a side of the enclosure 14 and the electrolyser assembly is inserted in and out of the enclosure 14 in a horizontal direction. It is possible, however, for the enclosure 14 and the electrolyser assembly to be instead configured for vertical insertion of the electrolyser assembly in and out of the enclosure 14.

[0129] The above examples of an electrochemical cell system, i.e. an electrolyser system 200, or a similar fuel cell system, can be operated utilizing the method of the present invention for specific control of the operation of the system in the event of a deterioration or failure of any one or more of a cell unit, a stack 10 or a cell array (comprising such stacks 10 or cell units) within the system 200. With the present invention, through applying predetermined array replacement rules and system control strategies, maintenance of the system 200 can be made quicker and simpler.

[0130] The method of the present invention involves operating an electrochemical cell system with a plurality of electrochemical cells, each having one or more electrochemical cell units or stacks 10 thereof, and performing at least periodic detections for a failure condition amongst the cell arrays within the system. The electrochemical cell arrays each have an expected array lifetime, and that time period is divided into a set of predefined operational time intervals, each said operational time interval having an associated array replacement rule and an associated system control strategy.

[0131] Upon detecting a failure condition in an electrochemical cell array (such as a failure of a cell unit or a stack 10, or of the cell array as a whole), the method involves determining in which of said operational time intervals the electrochemical cell array (or the stack or the cell unit) having the failure condition is in, and following the associated array replacement rule and system control strategy for that operational time interval for the failed electrochemical cell array.

[0132] The associated array replacement rule may be to isolate (electrically and / or fluidically) or disable the cell array, the stack or the cell unit (or multiples thereof - e.g. pairs of stacks) having the failure condition, or it may be to replace the cell array, the stack or the cell unit (or multiples thereof - e.g. pairs of stacks) having the failure condition.

[0133] As for the associated system control strategy, that can typically be one of four options, as discussed below. Other options can also be envisioned too.

[0134] In a usual electrolyser system, the system will typically operate with a substantially common temperature and pressure amongst all stacks within the system, but which conditions change over time. For example, the temperature may change. These conditions can be controlled using a control device 400, and a common temperature and pressure can be achieved through operating the cell arrays in a common volume of an enclosure.

[0135] 15188541-1 PN853570GB 17

[0136] In such systems, the temperature is typically gradually changed over time since during normal operations the properties of the electrochemistry within each stack 10 will tend to slowly change. In particular, the electrical resistance of the stack tends to increase over time. The change in temperature serves to compensate for this. Therefore, by the control device 400 slowly increasing an operational temperature of the cell arrays (usually either intermittently or regularly throughout a working life of the system), this can maintain other criteria or parameters at a constant level. For example, the increase in temperature will reduce the electrical resistance of the stacks 10 back to the earlier level of electrical resistance, and thus the voltage and current can be kept constant. This in turn leads to a consistent hydrogen production rate from the electrolyser system (i.e. when it is converting steam into hydrogen and oxygen).

[0137] However, upon replacement of a cell unit, a stack 10 or a cell array within the system 200, there is a need to shut down the entire electrochemical cell system, and then to cool and disconnect electrically and fluidically the electrochemical cell array, the cell unit(s) or the stack(s) to be replaced, and then there is the process of reconnecting and ramping up the electrochemical cell system back up to normal operating temperature. These take time, creating undesirable downtime, and hence when possible it is desired to choose the first array replacement rule of not changing the faulty component, and to instead allow the system to continue to operate with the faulty component within it. Indeed, the present inventors realised that it can be more cost effective, or better in terms of the overall performance of the electrochemical cell system (by virtue of avoiding or minimising downtime), to not replace an electrochemical cell array, or the cell unit(s) or stack(s) thereof, merely because of a detected failure or deterioration condition, especially if the electrochemical cell array with the failure condition, or enough of the other electrochemical cell arrays within the system, are still sufficiently operational to provide an acceptable performance output. After all, performance of an electrochemical cell system is typically considered as a whole, in terms of the total operational output of the system, be that power from a fuel cell system (e.g. electrical power output) or off-gas from an electrolyser system (e.g. the rate of hydrogen production).

[0138] It has been found that a drop in performance of individual electrochemical cell arrays, or individual stacks or cell units within such electrochemical cell arrays, can thus, in some circumstances, be considered acceptable. This will typically depend on how far through the life of the array(s) failure occurs.

[0139] Nevertheless, optimizing the performance is still desirable, and the present inventors have realised that multiple factors can be considered in deciding how to act so as to optimise the overall lifetime performance of the electrochemical cell system, especially as some electrochemical cell systems - for example metal supported solid oxide electrolyser cell systems, may have a long service life, such as 40,000 hours or more, and multiple thousands of electrolyser cells may be provided in such an electrochemical cell system, so shutting down the system upon any failure can be highly disruptive.

[0140] In some embodiments, a system control strategy may be how to control operating parameters for the electrochemical cell system after detection of a failure condition. Such parameters may

[0141] 15188541-1 PN853570GB 18 include any one or more of input steam flow, voltage through the electrochemical cell arrays, current across the electrochemical cell arrays, or target operating temperature of any one or more of the fluids or internal volumes within the electrochemical cell system internal temperature of the SAM. Other factors that may be taken into consideration include, in the decision of whether to replace, the time and financial cost, and the effort required and length of downtime taken, along with the resulting loss or reduction of power or off-gas (e.g. hydrogen) production, while the electrochemical cell system is offline. Additional factors that may be taken into consideration when designing the options for the associated system control strategy, once the decision on whether to replace has been taken, can include target steam utilization, i.e. the percentage of water to be converted to steam, for either individual stack of whole electrochemical cell system, as dictated by the available heat of the system, or the target steam flow, i.e. the rate of steam provided to the electrochemical cell system. Typically a electrochemical cell system is designed to provide (as close as reasonably possible to) even distribution of steam to all stacks in an electrochemical cell system and this is controlled by manifolding, but the present invention, in some embodiments may provide as a system control strategy a potential option to vary steam supply between stacks, cell arrays or pairs of stacks by adjusting / including different sized orifice plates or valves within the system, for use if or when a stack is replaced. Additionally or alternatively, steam flow to the electrochemical cell system when it is undergoing maintenance could be altered by using an orifice plate or valve at the steam supply to that electrochemical cell system.

[0142] In some systems, the current and / or voltage to the stacks, the cell units or the cell arrays are controlled, for example by controlling the voltage, whereupon the current drawn depends upon the electrical resistance of the stack(s) in question. Electrical resistance can also be temperature dependent. Therefore, another parameter that can be controlled is the target operational (or fluid) temperatures. The operating temperature of the electrochemical cell system is typically controlled by varying a target temperature of input fluids (e.g. steam and optionally air / oxygen in an electrolyser system), or output temperatures by controlling the stack between endothermic and exothermic voltages - stack operation may be endothermic or exothermic depending on the supplied voltage or current or by the system's steam utilization.

[0143] Through these control decisions it has been found that it is possible to provide an optimised (for a given set of optimisation criteria) failure response strategy for a stack, or for a pair of stacks, or for a given cell array.

[0144] Fig. 7 shows a flow chart illustrating the typical control steps in performing this process.

[0145] Pre-assigned response strategies (i.e. the array replacement rule and the system control strategy employed) will typically include at least the following four options, which as illustrated in the illustrations of Figs. 6 and 8, can be optimal for given ages (i.e. given operational time intervals within an expected life cycle) of the system:

[0146] Option 1 - this option is usually applicable towards the end of the operational lifetime of the system. For example, it might be appropriate for between 30,000 hours of use and the end of the

[0147] 15188541-1 PN853570GB 19 normal life cycle, i.e. commonly at around 40,000 hours of use, making this option appropriate for the final quarter of expected life cycle.

[0148] This first option involves not replacing the failed or degraded cell array, cell unit or stack with a new replacement component. Instead the faulty or degraded component is taken out of service. This may be by providing electrical (or fluidic) isolation for the faulty cell array, cell unit or stack.

[0149] This option can also have a sub-option as this array replacement rule alone would reduce the total hydrogen output of the electrolyser system. For example, it may be combined with a system control strategy of driving the remaining stacks "harder", i.e. increasing the output hydrogen production rate of each stack. This could be, for example, to produce the same overall system hydrogen output as before, e.g. by increasing the current through the remaining stacks (through a voltage or temperature increase (or both). In some embodiments, the strategy may also include increasing the steam flow to the remaining stacks.

[0150] Option 2 - this option is usually applicable upon the occasion of an early life failure of a component. For example, this may occur within (typically) the first 100 or 200 or 500 hours of use.

[0151] This second option involves, for the array replacement rule, a decision to replace the failed or faulty / degraded component(s), and for the system control strategy, a decision to leave the operating conditions the same as before the replacement. Because of the age difference between the new component(s) and the old components, the new components will typically have a lower electrical resistance than the old components (as electrical resistance increases over time). Thus the new component(s) will draw more current for a given drive voltage. This will cause the new component(s) to work "harder", giving a higher steam utilisation. This can work at the beginning of the life cycle. However, the greater the difference in age of the components (old versus new), the greater will be the increase in steam utilization for the replacement component versus the old. However, there is often a maximum safe steam utilisation for electrolyser cell arrays and the stacks therein, and therefore it is inadvisable or potentially not possible or desirable to utilise this strategy (i.e. this second option) after the change in electrical resistance becomes too great. On other words, after certain period of time - for example after 100 hours of use since the beginning of operation of the SAM, a different option such as option 3 might become more preferable.

[0152] Option 3 - this option is usually applicable for the first half of the projected lifetime of a cell array or stack thereof. For example, this may occur for up to the first 20,000 hours of use.

[0153] For the array replacement rule - this option involves a decision again to replace the failed or degraded component(s), but for the system control strategy, the conditions of the system are instead set to a compromise between that which best suits the new component(s) and that which best suits the existing components. This will typically involve an operating temperature lower than the temperature before replacement, and thus a higher electrical resistance for the existing components than before the replacement of the faulty component(s). This protects somewhat against the new component(s) exceeding their steam utilisation limit, but means that the existing

[0154] 15188541-1 PN853570GB 20 components will have to be run at a lower current (i.e. less "hard"). This results in a total reduction in hydrogen production from the electrolyser system.

[0155] In order to counter the reduction in total hydrogen production, the steam flow to the system can be increased, with some steam balancing between stacks being introduced, for example to keep the steam flow to the new stacks lower than the steam flow to the old stacks. This would typically be by the use of calibrated orifice plates or valves at the respective stack inlets or the manifolds therefor.

[0156] In such an arrangement, the new stack current could exceed an upper working limit, especially if the existing stacks are too old, as they will have been running at a temperature higher than the new stacks would accommodate, and to reduce the temperature too much (for accommodating the new stacks) would reduce the overall system efficiency too much. Therefore, this third option may not be possible or desirable to use after a certain amount of time, such as after the halfway point in the projected system's lifetime.

[0157] Option 4 - the fourth option is then, in this example, the final option and is theoretically suitable throughout the life of the system, but it may not be as efficient as some of the other options at certain time extremes - such as at or near the projected end of life of the system, where option 1 may be better, or at the beginning or the first 100 or 200 or 500 or so hours of use of the system, where option 2 may be better, or in the second quarter of the projected life of the system, where the third option may be better.

[0158] For the array replacement rule, the fourth option involves a decision again to replace the failed or faulty / degraded component(s), but for the system control strategy, the system is run at the same operating point it would have been at prior to that replacement, but with the new / replacement component(s) being electrically controlled to be operated at a different voltage / current to that of the previously existing components still remaining in the system. The new component(s) can thus be provided with a lower current, which prevents their stacks from reaching or exceeding their safe steam utilisation limit.

[0159] The trade-off with the fourth option is that some of the stacks may be running endothermically, and therefore higher degradation is to be expected. This may affect the possible operating window for this strategy if it is to be utilized continuously, but can be operationally beneficial versus the other options during, for example, the third quarter of the projected life of the system, albeit perhaps not at the end of the fourth quarter, when option 1 may become more appropriate instead.

[0160] Typically, any replacement of the failed or degraded component(s) is limited to stacks in the system. However, in some cases, other components may also be replaced, for example heaters 150, 151, 152, pipework or manifolds 32, or other balance of plant components. When such other components are replaced (and indeed, when the system is newly built), they may release contaminants upon first start and in early life (in the first hour, for example, in higher concentration than typically through life). To avoid release of such contaminants to the stack(s) 10, and potential loss of performance of the stacks, a hot flush of the system may be conducted.

[0161] 15188541-1 PN853570GB 21

[0162] During the hot flush, stacks are bypassed or not installed, for example with temporary pipes transferring fluid from system connections which in normal operation connect to the inlet and exhaust manifolds. The hot flush may comprise passing fluid through the system components at elevated temperature, for example over 300°C, preferably over 400°C, preferably less than 600°C. During the hot flush, components exposed to oxygen / air during normal operation may be flushed with air. During the hot flush, components exposed to fuel during normal operation may be flushed with air or fuel. During the hot flush, contaminants such as oils, chromium, or silicon may be released from said other components and removed from the system without passing through and potentially contaminating the stacks. It will be appreciated that the present invention is not limited to only the above examples. Other examples will be readily apparent to one of ordinary skill in the art without departing from the scope of the appended claims.

[0163] These and other features of the present invention have been described above purely by way of example. Modifications in detail may be made to the invention within the scope of the claims.

[0164] 15188541-1

Claims

PN853570GB 22CLAIMS1. A method of operating an electrochemical cell system comprising a plurality of electrochemical cell arrays, each said electrochemical cell array having an expected array lifetime divided into a set of predefined operational time intervals, each said operational time interval having an associated array replacement rule and system control strategy, the method comprising: performing at least periodic detections for a failure condition amongst the plurality of electrochemical cell arrays; upon detecting a failure condition in an electrochemical cell array, determining in which of said operational time intervals the electrochemical cell array having the failure condition is in; and for the electrochemical cell array having the failure condition, following the associated array replacement rule and system control strategy for that operational time interval.

2. The method of claim 1, wherein the electrochemical cell arrays each have independent electrical control.

3. The method of claim 1 or claim 2, wherein each electrochemical cell array comprises two or more stacks of electrochemical cell units.

4. The method of any one of the preceding claims, wherein the electrochemical cell arrays are contained within an enclosure.

5. The method of any one of the preceding claims, wherein the array replacement rule is whether or not to replace or repair a failed cell unit, a failed stack, or a failed pair of stacks, or more than 2 stacks, within the electrochemical cell array having the failure condition.

6. The method of any one of the preceding claims, wherein the system control strategy involves controlling certain parameters of the system.

7. The method of claim 6, wherein the system monitors steam or fuel utilisation within the system and controls them within limits set by the associated system control strategy.

8. The method of any one of the preceding claims, wherein the system monitors and controls steam flow.

9. The method of any one of the preceding claims, wherein the method has an option within its system control strategy to vary steam supply to respective cell arrays, or to different stacks or pairs / sets / groups thereof, and optionally wherein steam supply is varied by including and / or adjusting valves or orifice plates within the system.

10. The method of any one of the preceding claims, wherein either or both a current and a voltage supplied across or to a cell unit, a stack or a cell array, is a controlled parameter.15188541-1PN853570GB 2311. The method of any one of the preceding claims, wherein a first associated array replacement rule is to not replace a faulty cell unit.

12. The method of claim 11, wherein an accompanying first system control strategy comprises a control instruction to drive remaining cell units, stacks or cell arrays harder.

13. The method of claim 12, wherein the remaining cell units, stacks or cell arrays are driven harder by increasing a current passing through them.

14. The method of claim 12 or claim 134, wherein the remaining cell units, stacks or cell arrays are driven harder by increasing steam flow to them.

15. The method of any one of the preceding claims, wherein a second associated array replacement rule is to replace the faulty cell unit, stack or cell array.

16. The method of claim 15, wherein a second system control strategy is to leave the operating conditions the same as before the replacement17. The method of claim 15, wherein a third system control strategy involves a control instruction, after replacement, to set the conditions of the system to a compromise between that which best suits the new stack(s) and that which best suits the existing stacks.

18. The method of claim 17, wherein the compromise is a condition that reduces the hydrogen output rate.

19. The method of claim 17 or claim 18, wherein the compromise is a condition in which the operational temperature of the system is lower than prior to the replacement.

20. The method of any one of the preceding claims, wherein, in order to counter a reduction in total hydrogen production, steam flow to the system is increased.

21. The method of claim 20, wherein steam flow is increased with some steam balancing between stacks within the system to keep the steam flow to any new cell unit, stack or cell array lower than steam flow to any old cell units, stacks or cell arrays.

22. The method of any one of the preceding claims, wherein a fourth system control strategy involves, upon replacing a faulty cell unit, stack or cell array, running the system at the same operating conditions it would have been operating at before the replacement, but for the replacement cell unit, stack or cell array, electrically controlling that or those replacement cell unit, stack or cell array instead at a different voltage and / or current to the other cell units, stacks or cell arrays.

23. The method of any one of the preceding claims, wherein the electrochemical cell arrays comprise electrochemical cell units of solid oxide or molten carbonate cells, and optionally wherein the cells are metal-supported solid oxide cells.

24. The method of any one of the preceding claims, wherein the electrochemical cell units are electrolyser cell units.15188541-1PN853570GB 2425. A controller configured to perform the method of any one of the preceding claims or an electrochemical cell system comprising a controller configured to perform the method of any one of the preceding claims.15188541-1

Images