GB2627434A

GB2627434A - Electrolyzer with variable number of active electrolysis cells

Info

Publication number: GB2627434A
Application number: GB2302082.9A
Authority: GB
Inventors: Manfredi Ugo
Original assignee: Hydrogen Waves Ltd
Current assignee: Hydrogen Waves Ltd
Priority date: 2023-02-14
Filing date: 2023-02-14
Publication date: 2024-08-28
Also published as: GB202302082D0; WO2024170498A3; WO2024170498A2

Abstract

A system, comprising: a plurality of electrolysis cells arranged in a cell stack 102, wherein the electrolysis cells are electrically connected in series and grouped into two or more cell groups (see figure 10, 104). Each cell group has an electrical contact at either end (see figure 10, 105, 107). An electrical circuit is also present with one or more switches (see figure 10, SW1,SW2, SW3), each switch coupled between the electrical contacts of a respective one of the cell groups. Each switch is configured to selectively disconnect the cell group from the cell stack by electrically bypassing the cell group via a lower resistance path. This then varies the number of active electrolysis cells (see figure 10, 104) in the cell stack 102. A controller 120 is configured to determine the number of active electrolysis cells based on a variable amount of direct current (DC) electrical energy supplied to the cell stack by an electrical energy source. The status of one or more switches is controlled based on the determination. The electrical energy source may be a renewable source of energy 114. The electrolyser cell may comprise a unitary cell stack. A battery may be used to store energy 136, controlled via management system 140.

Description

ELECTROLYZER WITH VARIABLE NUMBER OF ACTIVE ELECTROLYSIS CELLS

TECHNICAL FIELD

[0001] The present disclosure relates generally to electrolyzers and, more particularly, to electrolyzers in which the number of active electrolysis cells may be varied.

BACKGROUND

[0002] Electrolyzers use electricity to split water into hydrogen and oxygen in a chemical process called electrolysis. The electricity required for the electrolysis process may come from a renewable source, such as solar or wind. However, using electricity from such renewable sources poses various challenges. For example, the electricity supplied by a renewable source may be intermittent and variable. A need therefore exists for electrolyzers that may operate efficiently under varying input power conditions.

SUMMARY

[0003] An aspect of the present disclosure provides a system, comprising: an electrolyzer having a plurality of electrolysis cells arranged in a cell stack, wherein the electrolysis cells are electrically connected in series and grouped into two or more cell groups, each cell group having an electrical contact at either end; an electrical circuit having one or more switches, each switch coupled between the electrical contacts of a respective one of the cell groups and configured to selectively disconnect the cell group from the cell stack by electrically bypassing the cell group via a lower resistance path, to thereby vary the number of active electrolysis cells in the cell stack; and a controller configured to determine the number of active electrolysis cells based on a variable amount of direct current (DC) electrical energy supplied to the cell stack by an electrical energy source, and to control the one or more switches based on the determination.

[0004] Varying the number of active electrolysis cells in this manner may allow the load of the electrolyzer to be adapted to the variable DC output from an electrical energy source such as a renewable energy source (e.g., a solar photovoltaic system) and/or a battery. In other words, the load of the electrolyzer may be optimised to maximise the amount of electrical energy transferred from the electrical energy source to the electrolyzer for hydrogen production. Furthermore, the variable DC output may be applied 'directly' to the cell stack without the need for intermediate power electronics such as inverters or dc-dc converters.

[0005] The electrolyzer may be implemented physically in different ways. For example, the cell stack may comprise a unitary cell stack having a first electrical contact comprising an anode at one end of the cell stack, a second electrical contact comprising a cathode at the other end of the cell stack, and one or more intermediate electrical contacts.

Alternatively, the cell stack may comprise a first substack and a second substack electrically connected in series, each substack having a first electrical contact comprising an anode at one end of the substack and a second electrical contact comprising a cathode at the other end of the substack. The second substack may have one or more intermediate electrical contacts. Furthermore, the cell stack may comprise a plurality of the second substacks.

[0006] Cell groups may have the same number of electrolysis cells or may have different numbers of electrolysis cells. For example, the different numbers of electrolysis cells may be defined by a geometric sequence expressed as 2(n-1), where n 0. This may allow the electrolysis cells to be disconnected in a 'binary' fashion, which may minimize the number of external switches while still retaining the ability to disconnect/connect any number of the electrolysis cells in the second substacks with 1-cell regulation precision. Other combinations of the number of electrolysis cells may be used. Furthermore, in terms of the total number of electrolysis cells in the cell stack, the number of 'variable' electrolysis cells may comprise about 25% to 10% of the total number of electrolysis cells, while the number of 'fixed' electrolysis cells may comprise about 75% to 90% of the total number of electrolysis cells.

[0007] The system may comprise two or more electrolyzers and respective electrical circuits, and switching circuity to switch the connection between the electrolyzers from a series connection to a parallel connection and vice versa, wherein the controller may be configured to control the switching circuitry to connect the electrolyzers in series or in parallel, determine the number of active electrolysis cells for each of the electrolyzers, and control the respective electrical circuits based on the determination. In this way, the electrolyzers may be managed as a pool of electrolyzer cells, whose series or parallel electrical connection to the DC source may be modified dynamically by the external switches to reach specific goals. For example, the multiple electrolyzers can be connected in parallel to the DC source when the DC source provides a relatively higher current and a relatively lower voltage, while these multiple electrolyzers, or parts of them, may be electrically connected in series when the DC source provides a relatively lower current and a relatively higher voltage. The dynamic allocation of electrolysis cells to a parallel or series electrical connection may enhance hydrogen production and promote a modular approach, as well as improving system reliability and fault tolerance.

[0008] The electrical energy source may comprise one or more of: a renewable energy source or an electric energy storage device.

[0009] The controller may implement a maximum power point tracking (MPPT) algorithm on the renewable energy source in the determination of the number of active electrolysis cells.

[0010] The number of active electrolysis cells may also be varied so that the load of the electrolyzer may be optimised when the amount of electrical energy supplied to the cell stack is regulated. For example, the controller may implement an energy management policy that regulates the amount of DC electrical energy to be supplied to the cell stack. The amount of DC electrical energy may be regulated based on time of day, weather conditions and forecasts, hydrogen production demand, and/or any other factors. The energy management policy may define a first amount of electrical energy that is to be used immediately by the electrolyzer and a second amount of electrical energy that is to be stored or otherwise used. The first and second amounts of electrical energy may be a total amount of electrical energy produced by the electrical energy source. The controller may automatically determine the energy management policy or may receive the energy management policy.

[0011] The controller may determine the number of active electrolysis cells further based on one or more of: an operating condition of the electrical energy source; an operating condition of the cell stack; or an operating condition of the electric energy storage device.

[0012] For example, the operating condition of the electrical energy source may comprise one or more of: an irradiation level on a solar PV array, a temperature of the solar PV array, a voltage and current generated by a wind turbine or other type of electrical generator. The operating condition of the cell stack may comprise a temperature of the cell stack. The operating condition of the electric energy storage device may comprises one or more of: a Status of Charge (SoC) of the electric energy storage device, or a voltage of the electric energy storage device.

[0013] The controller may be configured to partition the electrical energy between the electrolyzer stack and another electrical energy utilization system, controlling the relative amount of electrical energy that goes to each through an appropriate determination of the number of active electrolysis cells. For example, the electrical energy utilization system may comprise an energy storage battery or a grid connected inverter.

[0014] The controller may be configured to regulate the rate of usage of the energy stored in the battery and going into the electrolyzer, through an appropriate determination of the number of active electrolysis cells.

[0015] The system may further comprise multiple voltage sources configured to supply a maintenance voltage to a disconnected cell group or to the whole cell stack when not in use. The controller may be further configured to control the electrical circuit to electrically insulate one or more of the cells, thus allowing to apply a maintenance voltage to the inactive cells or cell groups, while the rest of the stack is electrically connected and operative. This may reduce gas crossover problems and extend electrolysis cell lifetime. The controller may be further configured to control the electrical circuit to put in short circuit the unused cell or cells, in order to protect them against polarity inversion when detached from stack operation.

[0016] The system may further comprise a battery to store electrical energy for use by the cell stack, to smooth the variation in the electrical energy supplied by the electrical energy source. As such, the variable amount of DC electrical energy may comprise electrical energy supplied by a renewable energy source and electrical energy supplied by the battery. The controller may regulate the rate of usage of the electrical energy stored in the battery by modifying the number of active cells in the electrolyzer stack. When implementing an energy management policy, as described above, the controller may also regulate the relative amount of electrical renewable energy that goes towards the cell stack, and the remaining amount that will be stored in the batteries for later use.

[0017] Another aspect of the present disclosure provides an electrolyzer, comprising: a plurality of electrolysis cells arranged in a cell stack, wherein the electrolysis cells are electrically connected in series and grouped into two or more cell groups, each cell group having an externally accessible electrical contact at either end to allow the cell group to be electrically disconnected from the cell stack during operation of the electrolyzer, to thereby vary the number of active electrolysis cells in the cell stack.

[0018] Another aspect of the present disclosure provides a controller, comprising: a processor and memory, the memory storing instructions that, when executed by a processor, cause the processor to: determine a number of active electrolysis cells to be used from among a plurality of electrolysis cells arranged in a cell stack of an electrolyzer, wherein the electrolysis cells are electrically connected in series and grouped into two or more cell groups, each cell group having an externally accessible electrical contact at either end, wherein the determination is based on a variable amount of direct current (DC) electrical energy supplied to the cell stack by an electrical energy source and/or by a policy defined by a user, and control one or more switches, each of which is coupled between the electrical contacts of a respective one of the cell groups, to selectively disconnect one or more of the cell groups from the cell stack based on the determination.

[0019] The processor may be further configured to control one or more voltage sources and associated electrical switches to apply a maintenance voltage to a cell group or cell groups that are not in use. The processor may be further configured to apply a short circuit to a cell group or cell groups that are not in use.

[0020] The processor may be further configured to: determine a first number of active electrolysis cells to be used from among a first plurality of electrolysis cells arranged in a first cell stack of a first electrolyzer, wherein the first plurality of electrolysis cells are electrically connected in series and grouped into two or more cell groups, each cell group having an externally accessible electrical contact at either end, and control one or more first switches to selectively disconnect one or more of the cell groups from the first cell stack based on the determination, determine a second number of active electrolysis cells to be used from among a second plurality of electrolysis cells arranged in a second cell stack of a second electrolyzer, wherein the second plurality of electrolysis cells are electrically connected in series and grouped into two or more cell groups, each cell group having an externally accessible electrical contact at either end, and control one or more second switches to selectively disconnect one or more of the cell groups from the second cell stack based on the determination, and control switching circuitry to connect the first and second electrolyzers in series or in parallel.

BRIEF DESCRIPTION OF DRAWINGS

[0021] The present disclosure will now be described more fully with reference to the accompanying drawings, in which like numbers refer to like elements throughout, and in which: [0022] FIG. 1 is a graph of current-voltage curves for a PEM electrolyzer at different temperatures.

[0023] FIG. 2 is a table showing data for the current-voltage curve at 50 °C in FIG. 1. [0024] FIG. 3 is a table showing a subset of the data from FIG. 2.

[0025] FIG. 4 is a graph of current-voltage curves for a PV plant under varying irradiance.

[0026] FIG. 5 is a graph of current-voltage curves for a PV plant under varying temperature.

[0027] FIG. 6 is a graph of current-voltage relationships of a PV plant and two cell stacks having different numbers of electrolysis cells.

[0028] FIG. 7 schematically illustrates an electrolyzer.

[0029] FIG. 8 schematically illustrates how bipolar plates provide an electric contact between cathodes and anodes in an electrolyzer cell stack.

[0030] FIG. 9 schematically illustrates electrolyzers with different numbers of electrolysis cells.

[0031] FIGS. 10 to 12 schematically illustrate electrolyzers in which the number of active electrolysis cells may be varied.

[0032] FIG. 13 schematically illustrates a binary switching scheme to vary the number of active electrolysis cells.

[0033] FIG. 14 schematically illustrates operation of the binary switching scheme illustrated in FIG. 13.

[0034] FIGS. 15 to 17 schematically illustrate how multiple electrolyzers may be electrically connected in parallel or series.

[0035] FIG. 18 schematically illustrates a system including an electrolyzer in which the number of active electrolysis cells may be varied, connected to a renewable energy source.

[0036] FIG. 19 schematically illustrates a system including an electrolyzer in which the number of active electrolysis cells may be varied, connected to a battery.

[0037] FIG. 20 schematically illustrates hydrogen crossover in an electrolysis cell. [0038] FIG. 21 is a graph of voltage-current relationship in an electrolysis cell.

[0039] FIG. 22 schematically illustrates the electrolyzer of FIG. 10 with maintenance voltage functionality.

[0040] FIG. 23 schematically illustrates a controller.

[0041] FIGS. 24 are 25 are flowcharts showing operations of the controller of FIG. 23.

DETAILED DESCRIPTION

[0042] Throughout the specification and claims, the following terms take at least the meanings explicitly associated herein, unless the context clearly dictates otherwise. The meanings identified below are not intended to limit the terms, but merely provide illustrative examples for the terms. The meaning of "a," "an," and "the" includes plural reference. As used herein, the term "comprising" means including but not limited to and should be interpreted in the manner it is typically used in the patent context. Use of broader terms such as comprises, includes, and having should be understood to provide support for narrower terms such as consisting of, consisting essentially of, and comprised substantially of. As used herein, the word "example" is used herein to mean "serving as an example, instance, or illustration." Any implementation described herein as "example" is not necessarily to be construed as preferred or advantageous over other implementations.

[0043] Overview [0044] The present disclosure provides an electrolyzer in which the number of active electrolysis cells may be varied. Some principles for determining the number of active electrolysis cells are discussed with reference to FIGS. 1 to 6. It will be appreciated that not all the principles need come into play. Furthermore, although these principles are discussed in relation to a proton exchange membrane (PEM) electrolyzer and a PV source of electricity, they may apply more generally to other types electrolyzers including, but not limited to, alkaline water electrolysis (AWE), anion exchange membrane (AEM), Solid Oxid Electrolyzer (SOE), as well as other types of power sources, and other types of DC power.

[0045] FIG. 1 is a graph showing experimentally obtained current-voltage curves for a PEM electrolyzer working in the 0 A to 40 A current range for different temperatures of water. The working cell voltage is in the 1.2 V to 2.4 V range. Assuming an effective electrolyzer area of 40 cm2, the current density goes from 0.05 A/cm2 for a current of 2 A to 1 A/cm2 for a current of 40 A. In this example, the electrolyzer is designed to operate below 60 °C and with optimal efficiency for low current densities in the range of 0.1 A/cm2 to 0.9 A/cm2. The production of hydrogen is proportional to the current flowing in the cell. The graph shows there is zero current, and hence no production of hydrogen, below the thermodynamic limit UREV = 1.229 V, while there is a minimal production of hydrogen below the thermo-neutral limit UTN = 1.481 V. In the latter case, the cell will use some heat to dissociate water, cooling down if the extra heat is not supplied to the system.

[0046] FIG. 2 is a table showing the electrolyzer cell working voltage for each current as well as the corresponding Lower Heating Value (LHV) and Higher Heating Value (HHV) efficiencies, n LHV and.,HHV, ,HHV, for the current-voltage curve at 50 °C shown in FIG. 1. The highlighted area, corresponding to a capacity factor in the 20% to 85% range, shows an HHV efficiency of the electrolyzer in the 77% to 91% range. The corresponding cell voltage range is 1.62 V to 1.93 V. Taking two exemplary working points of the PEM electrolyzer, for example (10 A, 1.64 V) and (28 A, 1.823 V), and assuming a PV array that produces (10 A, 32.8 V) at maximum power point (MPP) conditions, it can be seen that for the first working point (10 A, 1.64 V) an electrolyzer stack comprising 20 active electrolysis cells (20 x 1.64 V = 32.8V) may provide an optimal load, whereas for the second working point (28 A, 1.823 V) an electrolyzer stack comprising 18 active electrolysis cells may provide an optimal load (18 x 1.823V = 31.8V). Accordingly, different numbers of active electrolysis cells may provide different optimal loads, at a given electrolyzer operating temperature, for two different operating power points of the PV array.

[0047] FIG. 3 is a table showing that an electrolysis cell may work at different currents, e.g., from 1 A to 40 A, with a different cell voltage associated with each current. At a given instance of time, a renewable source of electricity, such as a PV panel, may have a specific combination of current and voltage defining an MPP. To take an example, for an MPP current of 10A, an operating point of 1.64V for one electrolysis cell may be the most appropriate. Accordingly, these MPP voltage and current will become "ideal" for the PV to electrolyzer coupling if, by series connecting the proper number N of electrolysis cells, each of them will operate at the MPP current of 10A, while the MPP voltage is equal to the sum of individual cell voltages: Vmpp = N x 1.64V.

[0048] FIGS. 4 and 5 show I-V curves for a 60-cell PV panel under different solar irradiation conditions and at different temperatures, respectively. As shown in FIG. 4, current produced by the PV panel is directly proportional to irradiance, i.e., increasing with irradiance. The voltage increases as well, but logarithmically, therefore comparatively the voltage increase is less. As shown in FIG. 5, a temperature increase may significantly lower the voltage of the PV panel, for the given 1000 W/m2 irradiation, while the current produced by a PV panel increases only slightly with increasing temperature: hence, the total power produced by a PV panel decreases significantly with increasing temperature.

[0049] FIG. 6 is a graph showing some of the principles at work in combination in a system comprising an electrolyzer with direct current coupling to a PV array. In the graph, the solid and dashed lines are photovoltaic I-V curves at different levels of irradiance G. Each solid line (photovoltaic I-V curve) has two corresponding dashed lines (corresponding photovoltaic kV curves). These represent the effect of varying the number of PV panels in series (denoted by the arrow labelled No) and the effect of varying the number of PV panels in parallel (denoted by the arrow labelled No). The regular and bold dotted lines in the graph are electrolyzer I-V curves for two electrolyzer stacks (denoted EC1 and EC2 respectively) at different temperatures of 20 °C to 80 °C (10 °C intervals). The electrolyzer stacks EC1 and EC2 have different numbers of electrolysis cells. The effect of having different numbers of electrolysis cells is represented by the arrow labelled No. In the graph, 25!cc is the short circuit current, and Voc is the open circuit voltage. The graph of FIG. 6 demonstrates the following: [0050] (1) For any given maximum power point Vmpp of the photovoltaic I-V curves, an appropriate number Nc of electrolyzer cells may be required. For example, the electrolyzer EC1 has too few cells and cannot optimise usage of PV power at a given value of irradiation G. [0051] (2) The number of electrolyzer cells that is optimal at higher irradiation levels may not necessarily be optimal at lower irradiation levels, i.e., while the number of electrolyzer cells in the electrolyzer EC2 may be suitable at an irradiance of 1000 W/m2 (as denoted by the good overlap of curves in the region labelled A in the graph of FIG. 6; it can be observed that region A surrounds the Vmpp (1000), which is the MPP of the PV plant at 1000 W/m2 irradiation), the number of electrolyzer cells in the electrolyzer EC2 may be less suitable at an irradiance of 100 W/m2 (as denoted by the poor overlap of curves in the region labelled B in the graph of FIG. 6; such region surrounds the Vmpp (100), which is the MPP of the PV plant at 100 W/m2 irradiation). Thus, more electrolyzer cells may be required when irradiation is low. This may seem counter-intuitive, but with low irradiation levels the cells of the electrolyzer should be operated at low current density, i.e., at lower voltage per cell, thus the need to have more cells in series.

[0052] (3) The temperature variation in the electrolyzer stack may have a significant effect on the search for the Vmpp of the photovoltaic I-V curves, and different numbers of active electrolyzer cells may be used to match a given MPP if the electrolyzer stack temperature is different.

[0053] (4) The number Ns of PV panels in series may be used to establish the indicative working point of the electrolyzer stack; in particular, the number Nc of electrolysis cells and the number Ns of PV panels in series, may be chosen in advance to be compatible and generate a good overlap in the I-V curves of electrolyzer and PV plant, in most operating conditions of irradiation and temperature.

[0054] It is therefore contemplated that the number of active electrolysis cells may be determined according to one or more of the following: [0055] (1) The number of active electrolysis cells may be determined based on a variable amount of electrical energy supplied to the cell stack by an electrical energy source. For example, the number of active electrolysis cells may be varied to adapt the electrolyzer to a given work point (e.g., IMPPT, Vmpp-r) of a PV source of electricity. By "adapt" it is meant that the load for the electrolyzer may be optimised to maximise the amount of electric energy transferred from the PV source for hydrogen production. Put another way, the number of active electrolysis cells may be varied so that the operating voltage of the electrolyzer may be matched to the incoming voltage (e.g., total MPP voltage of the PV electricity source). For example, the number, N, of active electrolysis cells may be selected so that N multiplied by a single cell voltage approximates the total voltage (Vmpp) provided by the PV electricity source.

[0056] (2) The number of active electrolysis cells may be determined based on an operating condition of the electrical energy source such as the irradiance level.

[0057] (3) The number of active electrolysis cells may be determined based on an operating condition of the electrolyzer such as a temperature of the cell stack. Herein, the term "operating condition" may refer generally to any of various conditions that can be measured or calculated. The operating condition may relate to the current or real-time environmental or operational state of the apparatus or system that is being monitored. Thus, the operating condition may include environmental conditions that may affect the operation of the apparatus or system that is being monitored.

[0058] (4) A maximum power point tracking (MPPT) algorithm may be employed to determine the number of active electrolysis cells. The number, N, of active electrolysis cells is an integer number, so the MPPT algorithm may be considered a "discrete" MPPT algorithm in the sense that only a discrete number of electrolysis cells may be selected to match the total voltage (Vmpp). Thus, there may not be an exact match to the total voltage (Vmpp). However, as the number of active electrolysis cells increases, i.e., as the operating voltage of the electrolyzer increases, the (IEL, VEL) point selected by a "discrete" MPPT algorithm may become closer or equal to the (IEL, VEL) point selected by a "continuous" MPPT algorithm. In other words, with increasing voltages, for example 100 V and beyond, the "discretisation" effect may vanish and the "discrete" MPPT optimisation may produce the same optimal results of a "continuous" MPPT algorithm. For example, testing has shown that with a low PV operating voltage of about 35 V to 40 V and 23 electrolysis cells, 99.4% of the PV MPP energy may be delivered to the electrolyzer. This percentage may be expected to exceed 99.5% when the working voltage exceeds 100 V. [0059] Broadly speaking, systems described herein include switching electronics configured to switch a dynamically variable number of electrolysis cells in/out from an electric circuit defined by an electrolyzer, depending on one or more parameters such as the DC-source, the electrolyzer, a battery (if present), and the policies that a user wants to enforce on the system. A first example of what systems described herein may achieve is to track the MPP (Maximum Power Point) of a PV field in any realistic condition of PV irradiation, PV temperature, and stack temperature, without introducing appreciable conversion losses. A second example of what systems described herein may achieve is to dynamically configure the electrolyzer to share a pre-defined quota of the energy coming from the PV field, while another quota is sent to a battery-based storage system. In this scenario, systems described herein may partition the renewable energy into two fractions as desired by a user. A third example of what systems described herein may achieve is to maintain constant the amount of energy sent to the electrolyzer, while the excess energy is stored in the battery for later use.

[0060] Exemplary electrolyzer configurations [0061] With reference to FIG. 7, the components of an electrolyzer stack typically include a membrane electrode assembly (MEA), a bipolar plate, compression plates, and current collectors. The MEA forms the core of an electrolysis cell. There may be multiple MEAs in a stack. For example, the electrolyzer stack shown in FIG. 7 includes two MEAs and may therefore be referred to as a 2-cell electrolyzer stack. Generally, the MEA includes a porous or ionomer membrane, catalysts, porous layers, gas and water flow ducts, and both electrodes (anode and cathode) that do not have external connections. Adjacent MEAs are electrically coupled in the stack by the bipolar plate, so that the anode of one MEA is in in electrical contact with the cathode of the adjacent MEA (or vice versa). The bipolar plate also has water and gas flow ducts to allow a multi-cell stack to operate as a unitary electrolyzer stack. The stack is terminated by the metal compression plates, which are used to apply pressure to the internal elements. Compression may be achieved by tightening multiple threaded bars (not shown) that extend along the length of the stack. The stack may have metal plates, called current collectors, having electrical contacts to external circuitry to provide power to the cells. Typically, an electrolyzer stack has only two current collectors. Current collectors are monopolar plates, usually thicker than the bipolar plates, due to the need to transfer all the stack current to the external electrical contacts, while the bipolar plate has the task of transmitting current from one cell to an adjacent cell. Between the various elements, gaskets are interposed (not shown in figure), ensuring gas/water tightness and electrical insulation where required. Gaskets may be made of rubber, plastic, elastomers, etc. It is noted that polymer electrolyte membrane (PEM), anion-exchange membrane (AEM) and alkaline water electrolyzer (AWE) stacks may have similar configurations.

[0062] FIG. 8 schematically shows how conventional bipolar plates provide an electric contact between a cathode (-) of one MEA and an anode (+) of an adjacent MEA.

Generally, bipolar plates are not accessible from outside the cell stack for power delivery purposes. Rather, each end of the electrolyzer stack has a monopolar plate that is a single-electrode contact. For example, the electrolyzer stack shown in FIG. 8 has an end plate (anode) on left side and an end plate (cathode) on the right side. The anode and cathode at the ends of the stack should not be confused with the anodes and cathodes of the M EAs.

Monopolar plates are thick enough to deliver the whole current sent to the electrolyser. The current flowing in the whole electrolyzer is the same current Ice" flowing in each single cell connected in series. The voltage Uslack of the electrolyzer stack is the sum of individual voltage Ucau of all cells.

[0063] FIG. 9 schematically shows typical electrolyzer stacks having different numbers of electrolysis cells. The 1-cell electrolyzer stack has one MEA, an anode at one end of the stack, and a cathode at the other end of the stack. It has no bipolar plate. The 2-cell and 4-cell electrolyzer stacks have two and four MEAs 104, respectively, as well as one anode and one cathode, i.e., two current collectors, placed near the compression plates.

This means that the number of active electrolysis cells is fixed, and all the electrolysis cells are active, with the same current flowing through all of them.

[0064] Incidentally, it is noted that an actual electrolyzer is more complex than shown in the previous figures and may include additional elements such as a porous transport layer (PTL) and a gas diffusion layer (GDL) where H2 and 02 gases may diffuse after separation.

Furthermore, the bipolar plates and the end plates may have the additional function to guide fluid or gases inside the stack and therefore also often referred to as flow field plates to underline this specific function. For convenience and to avoid unnecessary complexity, the PTL and GDL may be considered part of the MEA shown in the previous figures.

[0065] Electrolyzers described herein may provide additional (internal) current collectors and allow an external control on the number of active cells. Thus, electrolyzers described herein may be of a 'dynamic size' in which some of the cells are active, and some of the cells are not active by bypassing them using external switches.

[0066] FIG. 10 schematically shows an electrolyzer 102 comprising a unitary cell stack having intermediate electrical contacts 103 disposed between a cathode 105 at one end of the stack and an anode 107 at the other end of the stack. As an example, an intermediate electrical contact 103 may be provided on a bipolar plate so that the bipolar plate also acts as a current collector. Existing bipolar plates may be modified to provide this additional functionality. As another example, an intermediate electrical contact 103 may be provided on a 'dedicated' current collector disposed between two bipolar plates.

This may provide a robust path for the current towards an external electric contact. Such dedicated' current collectors may be different from the current collectors located near the end plates, because they can ensure the continuity of the liquid/gas flow channels (piping) inside the stack. The intermediate electric contacts 103 and the anode 107 are connected to a circuit 112 comprising one or more switches. The cathode 105 is connected to switch SWO. Thus, the intermediate electric contacts 103, the cathode 105, and the anode 107 are also referred to herein simply as "electrical contacts" that may be connected to an external circuit. Broadly speaking, the term "circuit" refers to a combination of a number of electrical devices and conductors that, when interconnected to form a conducting path, fulfil some desired function. The term "switch" broadly encompasses any device for making, breaking, or changing a connection in an electric circuit. When a switch opens or closes, it breaks or makes, respectively a connection. Examples of switches include diode and transistor structures, electromechanical switches (e.g., relays), etc. [0067] As shown, the circuit 112 comprises three switches SW1, SW2, SW3 connected in parallel to the three intermediate electrical contacts 103, i.e., one switch is provided for each intermediate electrical contact 103. A controller (not shown) may selectively drive the switches SW1, SW2, SW3 to electrically disconnect or connect one or more of the electrolysis cells 104 from or to the cell stack (i.e., from or to the active electrolyzer circuit), thus realising in a straightforward way a 'variable length' electrolyzer stack. For example, if switch SW1 is "closed", the current will circulate through the lowest resistance path of the closed switch SW1, thus bypassing the electrolysis cells 104 that are electrically connected in parallel to the switch SW1. By selectively closing/opening the switches SW1, SW2, SW3, a variable number of active electrolysis cells 104 may be provided. This variable number of active electrolysis cells may be used to adapt the operating characteristics of the electrolyzer 102 to renewable energy source 114, which may be a PV field generator or any other DC source whose voltage is compatible with the stack regulation interval. The variable (switchable) electrolysis cells may act as a variable load that can compensate for PV array (I, V) variation depending on temperature and irradiation, and also for different stack temperatures. Here, SO is the general switch used to activate/deactivate the entire electrolyzer stack 102.

[0068] FIG. 11 schematically shows an electrolyzer 102 comprising a plurality of electrolysis cells including a first substack 108 electrically connected in series to a second substack 110. As used herein, the term "substack" generally refers to a portion of a cell stack (i.e., a group of electrolysis cells) having an anode at one end of the substack and a cathode at the other end of the substack. It will be appreciated that the terms "substack" and "cell group" may, in some cases, be used interchangeably. In other cases, a "substack" may include "a plurality of cell groups." For example, the first substack 108 has no intermediate electrical contacts so that the first substack 108 is equivalent to a cell group.

The second substack 110 includes two intermediate electrical contacts 103, such as described above with reference to FIG. 10, so that the second substack 110 is equivalent to a plurality of cell groups (three cell groups). Furthermore, the first substack 108 may be referred to as a "fixed" substack since it provides a fixed number of active electrolysis cells 104 to the cell stack 102. The second substack 110 may be referred to as a "variable" substack since it provides a variable number of active electrolysis cells 104 to the cell stack 102.

[0069] The intermediate electrical contacts 103 of the second substack 110 may be positioned to define multiple cell groups that follow a "binary rule" approach, to reduce to a minimum the number of power switches (contactors) required. This will be described in more detail later, but as an example the cell groups making up the second substack 110 may contain 1, 2, 4, 8 and 16 cells respectively, i.e., a total of 5 cell groups and 6 external contacts. This means that the second substack 110 can be configured dynamically to have any active cell size in the [0..31] cell range. This configuration has the advantage that the second substack 110 may be provided as an 'add-on' device (with all the necessary electrical contacts) to an existing electrolyzer stack having a fixed number of cells.

[0070] FIG. 12 schematically shows an electrolyzer 102 comprising a first (fixed) substack 108 and a plurality of second (variable) substacks 110. The substacks 108, 110 are electrically connected in series. The circuit 112 is connected to the cathodes 105 and anodes 107 of the second substacks 110 (as well as the anode 107 of the first substack 108). As such, the cathodes 105 and anodes 107 of the substacks 110 may serve as electrical contacts. However, any of the substacks 110 may also include additional intermediate electrical contacts as described above with reference to FIGS. 10 and 11. The number of substacks 110 may depend on requirements such as the ability to adjust the number of active cells with fine granularity (e.g., 1-cell precision). Having more substacks may increase cost and complexity in terms of the electrical, water and gas piping. On the other hand, having more substacks may also add redundancy since a whole substack could be bypassed if for example there is damage to one of its cells.

[0071] The percentage of electrolysis cells 104 in the substacks 110 may comprise between about 15% to 20% of the total number of electrolysis cells 104 in the electrolyzer 102. The precision of regulation in the substacks 110 is higher for increasing operating voltage. As an example, and as noted above, for a working voltage of 100 V, corresponding to about 60 electrolysis cells 104, the "discrete" MPPT algorithm may be around 99.5% efficient, whereas for a voltage of 400 V this may increase to more than 99.7%. The PV array voltage and the electrolyzer working voltage and regulation range may be determined in advance to increase compatibility.

[0072] It will be appreciated that the number of electrolysis cells depicted in FIGS. 10-12 is only exemplary, and that the number of electrolysis cells that may be electrically disconnected from the cell stack may be one or any greater number.

[0073] It will be further appreciated that the electrolyzer 102 may be an acidic or alkaline electrolyzer of any type, provided it is composed of multiple electrolysis cells (plates), and a fraction of these have a directly accessible electric contact to the internal electrode.

[0074] Binary grouping [0075] Referring now to FIGS. 13 and 14, the cells 104 may be grouped so that they may be switched on/off in a "binary" fashion. Such binary grouping may allow maintaining a 1-cell regulation precision, while reducing to the minimum the number of switches required. As shown in FIG. 13, a variable cell electrolyzer 102 comprises a first substack 108 with two electrical contacts (anode and cathode) at the input and output respectively (only one is shown), and a plurality of second substacks 110 each having an electrical contact (anode and cathode) at either end. As before, the first and second substacks 108, 110 may be referred to as fixed and variable substacks for convenience. In this example, the variable substacks 110 comprise four switchable groups of electrolysis cells, having one, two, four, and eight cells, respectively. Circuit 112 may comprise four switches S1, S2, S4, S8 corresponding to four switchable groups of cells.

[0076] In operation, if all the bypass switches S1, S2, S4, S8 are closed the variable substacks 110 are completely bypassed by the current. A group of cells may be inserted in the current path by putting the corresponding switch in the open position. With different open/closed combinations of the S1, S2, S4, S8 switches any combination of between zero to fifteen cells may be obtained (i.e., 24= 16 combinations) in this example. (In general terms, there may be 2N combinations for N switches.) Some of the different combinations of switch states of the circuit 112 are shown in FIG. 14. The label "EC" indicates the corresponding number of active cells. For example, in the state corresponding to EC=1 (one active cell from the variable substacks 110), switch S1 is open, while switches S2, S4 and S8 are closed. This causes the current from the output electrical contact 118 the bypass the 2-, 4-, and 8-cell groups, so that only the 1-cell group is active. Similarly, in the switch state corresponding to EC=5 (five active cells from the variable substacks 110), switches S1 and S4 are open, while switches S2 and S8 are closed. This causes the current from the output electrical contact 118 the bypass the 2-and 8-cell groups, so that only the 1-and 4-cell groups (substacks) are active. As another example, in the switch state corresponding to EC=14 (fourteen active cells from the variable groups 110), switches S2, S4 and S8 are open, while switch S1 is closed. This causes the current from the output electrical contact 118 the bypass the 1-cell group, so that the 2-, 4-and 8-cell groups (substacks) are active. Thus, when a bypass switch is closed, the corresponding cell group (substack) is bypassed (i.e., excluded from the electrolyzer circuit), because the current will flow through the closed low resistance switch (a short circuit with sub milli-ohm resistance) instead of traversing the electrolyzer cell group (substack).

[0077] It will be appreciated that other cell grouping combinations may be used to reach the desired total number of switchable cells and to reduce the number of switches. For example, the group with the largest number of cells, i.e., the 2(1 -1) group, can be made smaller than calculated by the binary rule, if less total switchable cells are required in the electrolyzer. For example instead of having four groups with 1, 2, 4, and 8 cells respectively, there may be provided four groups with 1, 2, 4, and 4 cells respectively or 1, 2, 4, N cells (N=1..7) respectively, if this number of cells is sufficient for electrolyzer voltage regulation purposes.

[0078] Multiple electrolvzers in series or parallel [0079] The electrolyzer cell switching architecture described above may be further improved if, instead of a single electrolyzer of the rated power, two (or more) electrolyzers are provided, which may be electrically connected in parallel or series by the controller, depending on the operating conditions of the whole system.

[0080] Consider a PV panel with Voc(1000) = 50 V at STC (standard test condition) of 25 °C and 1000 W/m2 normal irradiation. Voc has a logarithmic positive dependency from irradiation (i.e., increases with increased irradiation), indicatively there is a loss of 10% of Voc voltage going from 1000 W/m2 to 100 W/m2 irradiation, so Voc(100)=45V at 25°C. The Vmpp is always lower than Voc for a given irradiation, because the MPPT system must maximize the power P=V x I, and this means moving away from Voc where loc=0 by definition: Vmpp(1000)=41.0V that is 18% lower than Voc, while Vmpp(100) could be as low as 37V. The voltage of the PV panel for temperatures different from 25°C, can be calculated with the "Temperature Coefficient of Voc", with typical values in the -0.25 %/°C to -0.30 %PC. If the PV panel operates from -10°C (low sun, cold winter) to +80°C (high irradiation, hot summer), the electrolyzer cell switching system may be able to compensate for the voltage variations vs STC, corresponding to (-35 °C) negative variation in cold winter and (+55°C) variation in hot summer: these variations correspond to adding 3.5V or subtracting 5.5V from Voc. Thus, the following principles may be relevant for the direct-coupling with electrolyzers: [0081] (1) Current is proportional to solar irradiation, while voltage increase only logarithmically.

[0082] (2) Temperature rise has a significant effect on PV voltage, lowering it linearly with temperature, up to 85-90 °C which is the max temperature allowed for a PV panel.

[0083] (3) Temperature rise increases the current linearly, but the coefficient is smaller than that of voltage, so the power P=V x I decreases with temperature.

[0084] A typical PEM electrolyzer operates at 1.6V/cell when the load factor is low (10% of rated power), 2V/cell at 100% load, and intermediate voltages at intermediate loads. This means that to match a constant input voltage, a higher number of electrolysis cell is needed when input current is low, and a lower cell number when current is high. There is another effect to consider with PV fields, namely temperature: when solar irradiation is high, this is typically connected with high ambient temperature, and anyway the panel itself will heat tens of degrees beyond ambient temperature. A high PV temperature reduces the PV voltage and unavoidably leads to a non-optimal utilization of electrolyzer when coupled to PV plants, because the effective power (in a fraction of the cells) is lower than rated electrolyzer power (using all the cells, which is impossible). Thus, when temperature is high, the PV voltage is lower than in cold conditions. But high irradiation leads always to a high PV panel temperature (except exceptional condition when high irradiation is coupled to cold strong winds cooling the panel), so high irradiation will mostly always be coupled with a partial utilization of the cell stack.

[0085] As an example, assuming a real PV-electrolyzer system has at least two PV panels connected in series, for a total Voc=100V. The issue of scarce stack utilization with a PV source may be unavoidable with a single stack, but may improve significantly with two stacks, i.e., a modular electrolyzer, where the electronic switching system has the option to use cells from both stacks in series when solar power is low, while both stacks are used in parallel when solar irradiation is high.

[0086] The systems described above may be adapted to drive multiple electrolyzers connected to a single renewable power source, and dynamically modify the electric topology in order to maximize system performance and minimize cost. In particular, having two or more electrolyzer stacks that can be connected in series or parallel, depending on the operating conditions, may provide one or more of the following advantages.

[0087] A typical electrolyzer stack will run at lower voltage per cell when current is low, or in other words, when the load factor of electrolyzer is low. In scarce load conditions, the stack efficiency will be maximum, because working voltage of the cell will be near the 1.481 UTN thermo-neutral voltage. Wth increasing current, the cell will operate at higher voltage and lower efficiency. Some indicative values of voltage vs current characteristics of a typical electrolyzer are shown in the table below. It can be observed that maximum efficiency of the stack does not necessarily map to the highest efficiency of the whole system, where parasitic loads like energy to run the cooling system, could move the optimal point away from the minimum current. A typical electrolyzer can be run, without damages, up to 110% or 120% of the rated load, but clearly with further increased voltage per cell, and further deterioration of the efficiency.

Load factor: Vice!! 100% 2.0 50% 1.8 10% 1.6

Table 1

[0088] Consider a solar PV string, driving an electrolyzer system, made of 500W half-cell mono-crystalline PV panels, each one characterized by the following typical characteristics: Voc=50V, Vmpp=41V, lmpp= 12.2A. With 20 of such panels connected in a 2s1Op configuration (2 panel in series, 10 couples in parallel), a 10kW plant operating at nominal voltage of Voc=100V and Vmpp=82V, and capable of up to 122A current, is obtained. There are at least three alternative configurations for attaching an electrolyzer to such a PV string, namely: I/cell 120A 60A 12A [0089] (1) One electrolyzer rated at 120A current and run by conventional power supply electronics.

[0090] (2) One electrolyzer rated at 120A current and run by cell switching electronics described herein.

[0091] (3) Multiple smaller electrolyzers, run by cell switching electronics described herein. For example, two smaller electrolyzers, each one with 60A max current rating.

[0092] Initially, the 'lengths' of the electrolyzer (i.e., the required number of cells) of both the larger and the multiple smaller electrolyzers of these three configurations is verified. The operating voltage of a stack depends on both the solar irradiation and the PV panel temperature. There are other dependencies, such as the stack temperature, but these are not considered here to keep the example simple, and anyway they do not significantly

modify the final conclusion.

[0093] Table 2 below shows the number of cells required to exactly match the PV field Vmpp in different operating conditions from -20 °C to +90°C of PV panel temperature, at different solar irradiation from 1000 W/m2 (100% load), 500 W/m2 (50% load) and 100 W/m2 (10% load). The table is built using all the parameters of a typical PV panel, including the temperature dependence of Voc, the temperature dependence of Ise, the Voc dependence on solar irradiation, the electrolyzer cell operating voltage for three different loads. As noted above, the temperature of the stack is not considered to keep the model simple but in practice may be considered. It is observed that the number of cells shown in the table is computed and is a fractional number which may be rounded to the nearest integer value. 46.2

53:84: 53 7 45:fl 2.50 44.4 43.9 42.7 49.82 46.75 48.48 45.49 41.6 47.8 41.0 44.9 7.95 46AV: 39. as

PV Temp °C #cell, 100% #cell, 50% #cell, 10% 51.83

Table 2

[0094] The design temperature interval for the PV panels has been chosen as a subset of the table values in the [-10.. +80] °C range. It can be observed that the highest number of cells, that is 53, is required when load is 10% and the PV panel temperature is very cold at -10°C, a situation corresponding to deep cold winter. Low load is synonym of low solar irradiation, and low current circulating in the cells.

[0095] Conversely, the lowest number of cells needed to match the MPP of PV string, that is 35 cells, correspond to the maximum current and the highest temperature of the PV panel of 80°C.

[0096] Finally, it can be observed that certain combination of load and temperature in the table, are extremely rare or of short duration: a high solar irradiation of 1000 W/m2 will heat the PV panel of many tens of degrees over ambient temperature, so it is impossible to have, for more than a few minutes, a PV panel at -10°C and max output. The same is true on the right side of the table: if irradiation is 100W/m2, the PV panel will rapidly cool to the ambient temperature and cannot stay at 80°C for long.

[0097] FIG. 15 compares a traditional fixed-plate electrolyzer driven by a DC-DC converter, with an electrolyzer 102 in which direct connection to electrolyzer cell groups is provided by means of a switching circuit 112. The alternatives 1) and 2) are analyzed.

[0098] As shown at the top of FIG. 15, if the single electrolyzer is coupled to a traditional power supply electronics, like a controlled rectifier eventually coupled with a buck or boost converter, the fixed number of plates will be around 45. This number will ensure a decent coupling efficiency in average operating condition of PV field, but will introduce appreciable losses when operating at full irradiation (100%), where fewer cells would be better suited, as well as when operating at low irradiation (10%), where more cells would be required

(see Table 2 above).

[0099] If the single electrolyzer is coupled with switching electronics described herein, no losses are introduced, but the stack is slightly longer than with traditional electronics, with 52 switchable cells instead of 45 fixed cells. This may ensure an 'ideal' coupling with a wide range of operating conditions and addresses the [-10..+80]°C range of PV panel temperatures.

[00100] Recall the observation made with reference to Table 2, that certain combinations of temperature and load are improbable. The present disclosure contemplates the use of two smaller 60A stacks, instead of a single larger 120A one, which may allow a serial connection of the two stacks to be used in conditions of low load and low temperature, while a parallel connection may be used in conditions of high solar irradiation and high PV temperature. The required length of the stack and the number of cells required are evaluated below to determine whether there is an advantage in doing so.

[00101] FIG. 16 shows how the cell switching architecture described herein may be applied to two smaller stacks (40 cell, 60A each) instead of a single stack (52 cells, 120 A). Assuming the system is at 10% load and PV temperature is 10°C, it can be seen from Table 2, that the correct number of cells is 50. The controller 120 can easily configure this cell number by connecting in series the two electrolyzer stacks. A modification for multiple electrolyzer stacks is to add an auxiliary switch SW-AUX to control the insertion/exclusion of the largest group of cells (the 27 cell group) from each stack 102. For solar irradiation below 50%, indicatively, the system processor will typically use the "serial connection" of the two stacks 102. With plenty of cells to insert/exclude from current path, the serial connection can track the MPP of solar PV even in case of very low loads, such as 5%, or extremely cold temperatures. The reduced membrane area of the 60A stacks 102 may limit crossover problems and maintain hydrogen production active even with low irradiation levels.

[00102] When irradiation goes beyond 50%, and current exceeds the 60A limit, the switching electronics can reconfigure the electrical connections and run the two smaller stacks 102 in parallel, thus doubling the managed current to 120A. In high irradiation conditions, the PV panels will heat up, so it is virtually impossible that they will have low or negative temperature, and this will restrict considerably the number of cells to switch on/off.

Also, the current will be consistently in the 60-120A range, thus also restricting the voltage range of electrolyzer to a small range (from 1.8V to 2.0V). All this will make a shorter stack having 40 plates sufficient to track the MPP of solar PV in high irradiation and temperature conditions. As another example, FIG. 17 shows a configuration with two stacks connected in parallel, both configured with 38 of 40 active cells, corresponding to 100% irradiation and 50°C temperature.

[00103] It is important to note that FIGS. 16 and 17 show the series and parallel connections of the electrolyzers 102 separately for reasons of clarity. The systems may include additional circuitry (e.g., nodes, connections, switches, etc.) not shown in these figures that allow the electrolyzers 102 shown in each of these figures to be selectively connected in series or parallel, i.e., switching circuity to switch the connection between the electrolyzers 102 from a series connection to a parallel connection and vice versa.

[00104] In summary, the advantages of a parallel/serial configuration of two smaller stacks using the switching architecture described herein instead of a single larger stack may include one of more of the following: [00105] (1) Smaller cell stacks (e.g., two 60A stacks) may need fewer plates than a single stack (e.g., 40 compared to 53 for a single stack).

[00106] (2) There may be no significant cost increase for using multiple smaller cell stacks with half-current plates compared to the cost of a single stack.

[00107] (3) Multiple stacks may increase redundancy, allowing a hydrogen plant to operate even in case of a failure in one stack, and making easier the maintenance on one stack.

[00108] (4) The series-configuration, with plenty of cells to switch on/off, may allow lossless tracking of the MPP even with very low solar irradiation and cold temperatures.

[00109] (5) The reduced membrane area of smaller cell stacks may allow operating the PV plant even at low irradiation, for example 5% of rated system power, without significant hydrogen crossover problems.

[00110] (6) The switching architecture may manage two or more stacks in series/parallel combinations, allowing an incremental upgrade in plant capacity, and promoting the development of low-cost modular stacks.

[00111] (7) Stacks with different power and varying number of plates may be integrated in the EC pool safeguarding previous investments.

[00112] (8) With multiple stacks running in series/parallel and controlled by the same controller, it is possible to scale the size of a plant to very large powers (>10MW) with ease, and without managing too high currents in each switch.

[00113] It is to be understood that the electrolyzers that are connected to the DC energy source can be two or more, with preference for even numbers that can go in parallel.

[00114] FIG. 18 shows an exemplary system comprising an electrolyzer 102 and control electronics 119. The control electronics 119 may comprise a controller 120 that may be configured to define and maintain an optimal working point of the electrolyzer 102. In particular, the controller 120 may dynamically adjust the load of the electrolyzer 102, i.e., the working voltage, current, and power of the electrolyzer 102, to achieve different aims such as: (i) track the MPP of power source 114 in order to maximize the extracted energy; (ii) define the fraction of renewable energy that is directly sent to the electrolyzer 102, and the fraction that is sent to a storage system (not shown) for later use; (iii) regulate the rate of power extraction from the storage system in order to reach a target number of hours in the use of the electrolyzer 102. The second and third aims are described later with reference to FIG. 19. Broadly speaking, the controller 120 may achieve these aims by selectively commanding switches SW1, SW2, SW3 of circuit 112 to activate/deactivate the cells (plates) in the corresponding cell groups 110 of the electrolyzer 102.

[00115] The controller 120 may receive, measure, calculate, or otherwise obtain data relating to operation of the electrolyzer 102 and/or the PV power source 114. For example, the controller 120 may measure the current/voltage from the PV power source 114 (as indicated in the figure by "Vpv" and "lpv"), may obtain or compute MPP value(s) (as indicated in the figure by "Vmpp" and "Impp"), and/or may obtain input and/or output water temperatures of the electrolyzer from temperature sensors (as indicated in the figure generally by "TEin" and "TEout" and specifically by "TEii" and "TE21"). Based on one or more of these values, the controller 120 may determine the number of active electrolysis cells of the electrolyzer 102. The controller 120 may output one or more signals (commands) to circuit 112 to activate/deactivate variable cell groups 110 of the electrolyzer 102 using switches SW1, SW2, SW3 (e.g., described above with reference to FIGS. 10 to 12), to regulate the working point of the electrolyzer 102 in terms of voltage, current, and power.

[00116] As illustrated in FIG. 18, circuit 112 comprises electromechanical switches SW1, SW2, SW3 operated by magnetic coils 124 through actuator 122. However, any suitable switch technology may be employed including, for example, solid state switches such as MOSFETs, IGBTs, and other power semiconductors.

[00117] The Mpp voltage may depend on the irradiation condition and PV panel temperature (as indicated in the figure by "IRR" and "Tpv"). Their values may be sensed in different ways, for example with a connection to the PV power source 114, or with a reference PV power source 128 and reference MPPT electronic load 126 (as illustrated in FIG. 18). Alternatively, they may be emulated using an irradiation sensor and an ambient temperature sensor. However, a connection to the PV power source 114 may be advantageous since it may take account of environmental conditions such as wind conditions which may affect the temperature of PV panels of the PV power source 114.

[00118] The controller 120 may perform a "perturb and observe" type of optimisation, by modifying the number of active electrolyzer cells in small increments in order to maximise the power absorbed by the solar field. The controller 120 may also be configured to perform auxiliary functions such as obtain forecasts relating to energy production of the PV array 114 to optimize energy management policies, send work data, performance data, and/or alarm data to external systems, and so on.

[00119] By way of example, and not limitation, for hydrogen production in the 240 kW-20 MW scale, the operating voltage of the electrolyzer 102 may be in the 120-1000 VDC range, the switched current may be in the 1,000-10,000 ampere range, and the number N of switches may be 4 to 7, depending on the operating voltage of the stack, as shown in the following table. A number of switches N=7, will allow to insert/remove from 0 to 127 plates from the stack, corresponding to a voltage regulation interval of 256V, more than sufficient to even regulate a 1000 VDC stack. Smaller stacks will have N=4 or N=5 switches. This is shown in Table 3 below.

Current (A) Voltage (V) V/cell kW/cell # cells Power kW # switches'. V reg % reg 1000 120 2.0 2.0 60 240 4 32 27% 2000 120 2.0 4.0 60 480 4 32 27% 2000 250 2.0 4.0 125 1000 5 64 26% 6000 250 2.0 12.0 125 3000 5 64 26% 8000 400 2.0 16.0 200 6400 6 128 32% 10000 400 2.0 20.0 200 8000 6 128 32% 10000 500 2.0 20.0 250 10000 6 128 26% 10000 800 2.0 20.0 400 16000 7 256 32% 10000 1000 2.0 20.0 500 20000 7 256 26%

Table 3

[00120] This means that, even for very large plants, the switches SW[i..1,1] may be operated at relatively low voltage differential, a favourable condition to reduce arcing problems that could deteriorate the contacts. For reasons of clarity, FIG. 18 does not show a switch SO as in FIGS. 10 to 12, though this may be provided as a general switch in the form of a high voltage contactor. However, it is expected that switch SO will be operated seldomly, for example when putting the entire electrolyzer 102 offline. On the other hand, switches SW[i NI corresponding to variable cell groups 110 may be operated more frequently and may therefore be implemented using solid state switches based on MOSFETs.

[00121] It will be appreciated that the system shown in FIG. 18 may employ other types of power sources such as wind power. For example, the working point of a wind turbine may be defined by wind speed Vw, blade pitch angle 8 and rotor speed D. The controller may determine the number of active electrolysis cells accordingly.

[00122] With reference now to FIG. 19, systems described herein may support the use of a battery unit 136 to store electric energy produced by power source 114 for later use by the electrolyzer 102. In the illustrated system, the task of tracking the MPP of power source 114 to maximize the extracted energy may be delegated to an MPPT charge controller 134, which may also ensure that the battery unit 136 is charged at the right voltage. Meanwhile, the controller 120 may regulate the load of the electrolyzer 102. It may also regulate the amount of renewable energy that is diverted to the battery unit 136 and/or schedule the use of energy stored in the battery unit 136 by the electrolyzer 102. For example, during the day the controller 120 may determine the amount of energy to be used immediately by the electrolyzer 102 and the amount of energy to be diverted to the battery unit 136, and during the night may regulate the discharge rate of the battery unit 136 to the electrolyzer 102. Thus, the battery unit 136 may be used to maintain a controlled and uniform current density in the electrolyzer 102, for example to reduce or eliminate periods in which the electrolyzer 102 is offline and /or to reducing the dynamic variation of current when online. Operating the electrolyzer 102 with controlled, nearly constant current may significantly extend its lifetime since wearing effects associated with electrolyzer use in strongly dynamic conditions may be reduced or eliminated. It will be understood, of course, that the functions of the MPPT charge controller 134 and the controller 120 may be carried out by the same controller.

[00123] The battery unit 136 may include a battery management system 140 configured to continuously check the battery cells (1... N), maintaining them in a correct voltage/current operating range. The battery management system 140 may also implement different safety mechanism to protect the battery unit 136 against improper user, failure of external equipment, and extend its life. An active balancer 138 may be employed to level the state of charge (SoC) of each cell (1... N). The battery unit 136 may also include a Coulomb meter 142 and shunt 144. The Coulomb meter 142 may determine parameters of the battery unit 136 such as the voltage and current (as indicated in the figure by "VBAT" and "'BAT"). Any suitable battery technology may be used such as lithium LFP batteries.

[00124] The system of FIG. 19 may employ wind, grid, and/or PV power. For a combination of wind and solar power, the voltage of the wind power may be adapted to the current battery level with an AC-DC converter (if the wind turbine produces AC) or a DC-DC converter (if the wind turbine produces DC). For example, an AC producing wind turbine may be introduced to the system by adding a 3-phase rectifier, a levelling circuit (capacity, inductance), and multiple DC-DC high voltage chargers in parallel, adapting the levelized DC current to the battery voltage. This is the same architecture as may be used for a PV power source and is both simple and scalable.

[00125] Some advantages of having electrical storage may include the following. The electrolyzer 102 may be smaller because the battery unit 136 may provide the ability to operate for 24 hours per day with a PV power source that typically harvests energy only between 6 to 12 hours per day, depending on season and country. Furthermore, the battery unit 136 may compensate for variation in the PV power, so the electrolyzer 102 may be powered at nearly constant current. Also, it may be straightforward to integrate multiple power sources.

[00126] Referring now to FIG. 20, PEM electrolyzers may be subject to a problem called "hydrogen crossover" (also called permeation or diffusion), i.e., the undesired diffusion of the H2 gas from the anode to the cathode through the membrane. This has been ascribed as one of the main causes of deterioration of perfluorinated ionomer membranes (Nafion), normally employed in PEMWE and PEMFCs. Hydrogen crossover increases when temperature, pressure and humidity of the cell rise. Hydrogen crossover may result in lower conversion efficiency, may require the addition of catalysts that will force a recombination of permeated H2 with local 02 to produce water, may result in the production of highly corrosive peroxide radicals that cause the degradation of both the PEM and the catalyst layer, or may result in hydrogen permeation rates that exceed 10-20 mA cm-2 after long term operation (compared to a normal hydrogen permeation rate through a very thin PEM is typically lower than 1 mA cm-2). The last point may be of relevance because aged thin (e.g., 57 pm) Nafion membrane cannot be used at low current densities in a PEM stack, indicatively below 0.5 A/cm2, without originating a large hydrogen crossover problem. When current density is lower than 500 mA/cm2, the permeate H2 may correspond to a current around 10 mA/cm2, which is a sizable amount (1/50) of the 02 generated by the given 500 mA/cm2 current density. In this situation the H2 percentage could reach 2% of the H2-02 mixture, not far away from the 4% limit.

[00127] Molecular H2 diffusion in Nafion membranes may happen both in dry and hydrated PEM, but the hydration makes the process more effective. Nafion is the ionomer (a polymer made of ions) used in PEM cells to separate the electrodes while allowing protonic (H+) flow. The H2 diffusion problem is independent from the differential pressure in the H2/02 sides, while is inversely proportional to the membrane thickness: thinner hydrated aged PEM membranes are in the worst situation, with more molecular hydrogen diffusing to the wrong side. On the other hand, the thinner Nafion membrane are also the ones with lower overpotentials (better voltage performance), so with PEM there is a difficult trade off: to improve the cell performance it is desirable to use thin membrane, but this will make the H2 diffusion problem unmanageable, and risk throwing away a lot of PV power in periods of low irradiation. Other membrane types such as Zirfon membranes and Anion Exchange Membranes (AEMs) typically have 20-40 times less H2 diffusion than Nafion membranes.

However, even in these membranes the hydrogen crossover problem may exist.

[00128] If electric power is removed from the electrolyzer, there is no production at all of H2 and 02, while the permeation mechanism is still active. The situation may become critical at the anodic compartment, where the existing 02 gas is progressively diluted by H2 gas diffused from cathode compartment. In absence of any electrical field, there no way for these H2 molecule to return to cathode, so the percentage of H2 in 02 will gradually increase and eventually go beyond the 4% explosion limit. The abundance of H2 in 02 environment, and the favourable pH condition, can lead to peroxide formation and chemical aggression to the PEM membrane or the whole M EA. Another condition leading to corrosion and oxidation is the abrupt change in potential that comes for the abrupt interruption of DC voltage to the electrolyzer.

[00129] Systems described herein may introduce a voltage source configured to apply a conditioning or polarization voltage to single cells, group of cells, or even entire electrolyzer stack, when they are not in use, or after/before they are or will be in use. This possibility may be useful not only for PEM electrolyzers but also AWE and AEM electrolyzers. For example, the polarization voltage may be a constant maintenance voltage. For example, a constant maintenance voltage in the range of 1.0 V to 1.5 V per electrolysis cell (i.e., around the UREV = 1.229 V water dissociation reversible voltage) may be applied over a prolonged time period. This may reduce corrosion and passivation phenomena, control the hydrogen diffusion, extend the lifetime of the electrolyzer cells, and limit the pH changes in the cell. By applying a small polarization voltage, in the 1.0 V to 1.5 V range, it may be possible to counteract the diffusion and return some of the H+ protons to the cathode side. The purpose of the polarization voltage applied to cells not in use, is to maintain a good purity of oxygen and hydrogen gas generated, delay membrane aging and electrolyzer components corrosion, oxidation and passivation, extend the electrolyzer useful life with acceptable performance. The presence of the voltage polarization mechanism may be less relevant in AWE and AEM electrolyzers, that are more resistant to hydrogen diffusion; however, AWE and AEM electrolyzer are much more sensitive than PEM electrolyzer to corrosion phenomena, due to their non-noble metal electrode and catalyst, so at the end all electrolyzer types can take advantage of the proposed voltage polarization mechanism for inactive cells. Another effect of the polarization voltage mechanism may be to limit the pH changes in the cathode/anode compartment when electrical current is removed. These changes certainly happen in PEM cells, leading to partial loss of catalyst performance, peroxide and other strongly oxydizing radical formation, attack to the Nafion (PFSA, Per-Fluoro-Sulfonic Acid) proton exchange membrane, corrosion or passivation of the conductive metal parts. The control of pH condition by application of a conditioning (polarization) voltage, can control these degradation phenomenon, extending electrolyzer lifetime and performance.

[00130] As shown in FIG. 21, when the voltage per cell applied for maintenance is small, for example in the 1.2 V to 1.5 V per cell (i.e., around the water electrolysis reversible voltage UREV of 1.229 V per cell), the corresponding current density will be small (below 1 mA/cm2). That is, there may be impact on the production of hydrogen/oxygen and minimal impact on the use of the externally supplied DC electrical energy.

[00131] FIG. 22 shows how the circuit 112 of FIG. 10 may be modified to provide a protection mechanism against the aforementioned issues. In addition to the switches SW1-SVV3 to de/activate cell groups 1 to 3, the system provides N + 1 = 4 auxiliary switches SW1., SW122, SVV23. and SVV30a to apply a maintenance voltage to each cell group. The switch SWO acts as a contact breaker for the whole stack. It is noted that, for the auxiliary switches SW1a, SW12a, SVV23. and SW3oa, "open" means inactive and "closed" means active. This is opposite to the switches SW1-SW3, where "closed" means inactive (bypass), i.e., when a switch SW1-SW3 is closed it will bypass (making inactive) the corresponding group N of cells, because the current will flow through the low-resistance path of the closed switch instead of traversing the cell.

[00132] As shown in the figure, two cell groups are in "maintenance mode", where V1 maintenance voltage is applied to cell group 1, while V3 voltage is applied to group 3. Both group 0 and group 2 cells are active, with electrolyzer current flowing through them. When a cell group is active, it does not require maintenance. A maintenance voltage Vo can be applied also to group 0 of cells (the fixed part of the stack) when the whole stack is not in use. The circuits 112 of FIGS. 11 and 12 may be modified in a similar manner.

[00133] Thus, in summary, auxiliary switches may be introduced for each cell group to electrically insulate the group from the rest of the stack. Multiple DC sources, operating at different voltages, may be used to "maintain" the correct voltage in cell groups of different size. A maintenance voltage may be applied to the whole electrolyzer when not in use to extend its lifetime. The maintenance voltage may be a polarization voltage in the neighbourhood of 1.23V per cell (the reversible voltage), used to reduce the corrosion/passivation phenomena in the cell when it is not active: in this case, an electrolyzer cell can transform into a battery cell, with a complete polarity inversion where the cathode becomes the anode and vice-versa. Together with polarity inversion, internal electrochemical conditions in the cell change abruptly, with corrosion and catalyst passivation that are more relevant for alkaline electrolysers with non-precious metal catalysts.

[00134] It is noted that the simple act of short-circuiting a cell or group of cells when deactivated, may offer some protection against polarity inversion and subsequent corrosion. This may happen automatically where unused cells are bypassed with a short-circuit from the corresponding relay/contactor. The aforementioned methods may be combined, for example where a polarization voltage is applied for a brief period to stabilize the cell conditions, while long-term inactivity protection is ensured by the permanent short-circuiting of cells.

[00135] The use of an appropriate conditioning voltage may therefore have one or more of the following advantages. It may suppress or limit changes in the acidity/basicity condition of the cell components when going from an active to an inactive usage condition, and vice-versa. This may avoid or attenuate unwanted effects like corrosion, oxidation, formation of aggressive chemical like hydrogen peroxide, that could damage the cell components and/or could reduce catalyst activity (passivation), eventually leading to a reduced lifetime or reduced efficiency of the electrolyser. The conditioning voltage may address both the electrolyzer cell lifetime duration issue, and strongly reduce the gas crossover problem in unused cells or entire electrolyser. The gas crossover problem may be more relevant for electrolyzer working at high pressure, both absolute and differential, and thinner membranes. The conditioning voltage may reduce the damages to the electrolyzer components deriving from dynamic usage conditions, happening with renewable energy sources like PV. The conditioning voltage may reduce the damage to cells components deriving by switching them on/off line.

[00136] FIG. 23 is a schematic block diagram of a controller 120. The controller 120 may include a processor 150, memory 152, a communications interface 154, an input interface 158, and an output interface 160. The controller 120 may be configured to execute the operations described below in connection with FIGS. 24 and 25. Although components of the controller 120 are described in some cases using functional language, it should be understood that the particular implementations include the use of particular hardware. The processor 150 may be in communication with the memory 152 via a bus for passing information among components of the controller 120. The memory 152 may be non-transitory and may include, for example, one or more volatile and/or non-volatile memories.

In other words, for example, the memory may be an electronic storage device (e.g., a non-transitory computer readable storage medium). The memory 152 may be configured by a programming device 156 to store information, data, content, applications, instructions, or the like, for enabling the controller 120 to carry out various functions in accordance with examples described herein.

[00137] The processor 150 may be embodied in a number of different ways and may, for example, include one or more processing devices configured to perform independently. Additionally, or alternatively, the processor may include one or more processors configured in tandem via a bus to enable independent execution of instructions, pipelining, and/or multithreading. The use of the term "processor" may be understood to include a single core processor, a multi-core processor, multiple processors internal to the controller 120, and/or remote or "cloud" processors. In an example, the processor 150 may be configured to execute instructions stored in the memory 152 or otherwise accessible to the processor 150. Alternatively, or additionally, the processor 150 may be configured to execute hard-coded functionality. As such, whether configured by hardware or by a combination of hardware with software, the processor 150 may represent an entity (e.g., physically embodied as an integrated circuit or other electronic device) capable of performing operations according to the present disclosure while configured accordingly. Alternatively, as another example, when the processor 150 is embodied as an executor of software instructions, the instructions may specifically configure the processor 150 to perform the algorithms and/or operations described herein when the instructions are executed.

[00138] The input and output interfaces 158, 160 may be any device or circuit embodied in either hardware or a combination of hardware and software, through which the processor 150 may receive information from external devices and transmit information to external devices. The inputs 157 may be values relating to the operation of an electrical energy source and/or an electrolyzer and/or an electric storage device, and the outputs 161 may be signals (commands) to operate one or more switches and/or a voltage source, as described above.

[00139] The communications interface 154 may be any device or circuit embodied in either hardware or a combination of hardware and software that is configured to receive and/or transmit data from/to a network and/or any other device, circuit, or module in communication with the controller 120. In this regard, the communications interface 154 may include, for example, a network interface for enabling communications with a wired or wireless communication network. For example, the communications interface 154 may include one or more network interface cards, antenna(s), buses, switches, routers, modems, and supporting hardware and/or software, or any other device suitable for enabling communications via a network. Additionally, or alternatively, the communications interface 154 may include a circuit for interacting with the antenna(s) to cause transmission of signals via the antenna(s) or to handle receipt of signals received via the antenna(s).

These signals may be transmitted by the controller 120 using any of a number of wireless personal area network (PAN) technologies, such as Bluetooth® v1.0 through v3.0, Bluetooth Low Energy (BLE), infrared wireless (e.g., IrDA), ultra-wideband (UWB), induction wireless transmission, 3G, 4G, 5G, or the like. In addition, it should be understood that these signals may be transmitted using Wi-Fi, Near Field Communications (NFC), Worldwide Interoperability for Microwave Access (WiMAX) or other proximity-based communications protocols. The communications interface 154 may be used by the controller 120 to receive forecast data and to transmit performance data, operative data, and/or alarm data such as described above.

[00140] In addition, computer program instructions and/or other types of code may be loaded onto a computer, processor or other programmable device to produce a machine, such that the computer, processor or other programmable device that execute the code on the machine create the means for implementing the various functions, including those described in connection with the components of controller 120. The computer program instructions may be stored on at least one non-transitory computer-readable storage medium (e.g., computer software stored on a hardware device). Exemplary non-transitory computer-readable media may include, but are not limited to, one or more types of hardware memory, non-transitory tangible media (for example, one or more magnetic storage disks, one or more optical disks, one or more USB flash drives), computer system memory or random access memory (such as, DRAM, SRAM, EDO RAM), and the like.

[00141] FIG. 24 illustrates a flowchart containing a series of operations for improved operation of an electrolyzer. The operations illustrated in FIG. 24 may, for example, be performed by, with the assistance of, and/or under the control of the controller 120, as described above. In this regard, performance of the operations may invoke one or more of processor 150, memory 152, input interface 158, output interface 160, or communications interface 154.

[00142] As shown at block 10, the processor 150 may determine the number of active electrolysis cells to be used, as described above. This may include, for example, measuring the current/voltage from an electrical energy source, obtaining or computing MPP value(s), and/or may obtaining input and/or output water temperatures of the electrolyzer from temperature sensors. At block 20, the processor 150 may control one or more switches to electrically disconnect one or more switches, as described above. For example, the processor 150 may generate and transmit a signal to operate a relay of an external circuit, a signal to operate a transistor of an external circuit, or the like. For example, with reference to FIG. 18, the signal may be output to actuator 122 to activate the relays 124 to selectively open/close the switches SW1, SW2, SW3. The operations may include a feedback loop 30 by which the controller 120 repeatedly performs at block 10 and 20.

[00143] FIG. 25 is another flowchart containing a series of operations for improved operation of an electrolyzer. The processor 150 may repeat the operations at given periods of time, e.g., every 10 seconds. At block 170, relevant values are read from the electrical energy source(s) (e.g., voltage, current, power, temperature, irradiation, wind speed, turbine rotation speed, ...), from the electrolyzer (e.g., temperatures) and/or the battery (e.g., State of Charge, Voltage). These input values, together with updated forecasts (e.g., solar irradiation forecast, weather forecast for next days) and/or updated user policies, will become the input to the processor 150 (block 172).

[00144] In case of a single electrolyzer stack, the flowchart proceeds to the left-hand branch, where the processor 150 may determine the exact number of electrolyzer cells to use, for example, to match the MPP of a PV string (block 174). Next, at block 176, the processor 150 may activate/deactivate one or more switches to select the exact number of cells to conned in series. For example, the processor 150 may generate and transmit a signal to operate a relay or solid state device. In particular, with reference to FIG. 18, the signal may be output to actuator 122 to activate the relays 124 to selectively open/close the switches SW1, SW2, SW3. A polarization voltage may optionally be applied at block 178 to the excluded (non-active) cells.

[00145] In case of multiple electrolyzer stacks, the flowchart proceeds to the right-hand branch, where the processor 150 may determine if, in current operating condition, a series or parallel connection of the stacks is required (block 182). At blocks 184 and 186, the internal configuration (number of active cells) may be carried out for each stack. The process for each stack may be like the processes carried out at blocks 174 and 176 in the left-hand branch. At block 188, an auxiliary set of switches may be used to realize the series or parallel connection of the stacks depending on current operating conditions. Optionally, at block 190, the maintenance voltages may be activated on disconnected cells.

[00146] After the operations for a branch are executed, the processor 150 may wait for a predefined period of time, e.g., 10 seconds, and the procedure may start again as shown by feedback loops 180 and 192. Of course, other predefined periods of time may be implemented.

[00147] The operations of FIG. 25 described for a processor 150 that may be used in both a system having one stack and a system having multiple stacks that can be connected in series or parallel and configured individually. It will be appreciated, however, that the processor 150 need not be configured to carry out the operations for a multiple electrolyzer stack if it is being used in a system having a single electrolyzer stack (and vice versa), i.e., the method may include only the operations for one branch depending on the configuration of the system.

[00148] The electrolyzer architecture described herein may be implemented in conjunction with a wind turbine generator or a battery storage (coupled to any type of renewable energy source including a wind turbine generator). A wind turbine can have DC output or AC output. In the case of DC output, the interface to the electrolyzer system can be direct, provided the turbine voltage and the electrolyzer voltage are in a compatible range. In the case of AC output, rectifier and leveling stage are typically provided because the alternator is synchronous to wind, and has variable and different frequency than the 50/60 Hz grid. A DC-AC power converter may be used to transform the levelized DC power into AC.

[00149] The controller 120 may dynamically modify the electrolyzer configuration for different uses. For example, if all the power from the wind turbine is to be used for hydrogen production, the controller 120 may regulate the electrolyzer load without any DC-DC power conversion and associated loss. If the power from the wind turbine is used for hydrogen production and to provide electricity to the grid, the controller 120 may send part of the wind power to the electrolyzer for hydrogen production, while the rest is converted to AC electric power (via DC-AC inverter) and sent to the grid. If an electric storage (battery) is installed, the controller 120 may discharge it at the desired rate towards the electrolyser, to realize the desired energy usage policy.

[00150] The potential savings for wind power application may be lower than with PV, given the absence of the capex saving of the solar PV inverter. However, it is still an advantageous solution because there are no conversion losses, the cost of the switching electronics is lower than the cost of a full power DC-DC converter and is also more reliable, in low wind operation the electrolyzer will operate at high efficiency (which is not achievable with an electrolyzer having a fixed number of cells coupled to a DC-DC converter), and the switching circuit may offer protection against polarity inversion of unused cell or group without limited extra cost.

[00151] As explained through FIG. 15 to 17 and the related accompanying text, the single switchable cell electrolyzer can be substituted with two or more separated (modular) electrolyzers offering the controller 120 the option to run them in series or parallel to match current operating conditions, for example to track the MPP of a directly connected PV string. The use of multiple electrolyzers instead of the single one may have advantages of better modularity and fault tolerance. In connection with this disclosure, however, the use of multiple separated electrolyzer may even be convenient in term of cost, given that they can have a lower number of cells that a single electrolyzer: in practice the controller 120 will run the two electrolyzer (the active number of cells inside each one) in series when the PV plant operates at low power, low current and cold temperature conditions, where a higher number of cells is needed to match the MPP of the PV string, while the controller 120 will modify the electrical connection to a parallel one, when the PV string operates at higher power, higher current and hot temperature, where a lower number of cells is needed to match a different M PP. As seen in FIGS. 15 to 17 and related text, the size of the single electrolyzer used in a flexible series/parallel combination, can even be smaller, that is having less internal cell, than a fixed number of cells electrolyzer used with a traditional high-cost DC-DC or AC-DC power supply. From the economic point of view, apart the other advantages, a couple of half-size electrolyzer, could cost less than a single full-size one, with higher number of cells of larger area. A half-size electrolyzer can also run at very low solar irradiation levels, in the 50 to 100 W/m2, where a single electrolyzer should stop working, due to insufficient current density leading to hydrogen crossover problems.

[00152] Many other implementations and modifications will come to mind to one skilled in the art to which this disclosure pertains having the benefit of the teachings presented in the foregoing descriptions and the associated drawings. Therefore, it is to be understood that this disclosure is not to be limited to the specific examples described herein and that modifications and other implementations are intended to be included within the scope of the appended claims. Moreover, although the foregoing descriptions and the associated drawings describe certain example combinations of elements and/or functions, it should be appreciated that different combinations of elements and/or functions may be provided by alternative implementations without departing from the scope of the appended claims. In this regard, for example, different combinations of elements and/or functions than those explicitly described above are also contemplated as may be set forth in some of the appended claims. Although specific terms are employed herein, they are used in a generic and descriptive sense only and not for purposes of limitation.

Claims

CLAIMS1. A system, comprising: an electrolyzer having a plurality of electrolysis cells arranged in a cell stack, wherein the electrolysis cells are electrically connected in series and grouped into two or more cell groups, each cell group having an electrical contact at either end; an electrical circuit having one or more switches, each switch coupled between the electrical contacts of a respective one of the cell groups and configured to selectively disconnect the cell group from the cell stack by electrically bypassing the cell group via a lower resistance path, to thereby vary the number of active electrolysis cells in the cell stack; and a controller configured to determine the number of active electrolysis cells based on a variable amount of direct current (DC) electrical energy supplied to the cell stack by an electrical energy source, and to control the one or more switches based on the 15 determination.
2. The system according to claim 1, wherein the electrolyzer comprises a unitary cell stack having a first electrical contact comprising an anode at one end of the cell stack, a second electrical contact comprising a cathode at the other end of the cell stack, and one or more intermediate electrical contacts.
3. The system according to claim 1, wherein the electrolyzer comprises a first substack and a second substack electrically connected in series, each substack having a first electrical contact comprising an anode at one end of the substack and a second electrical contact comprising a cathode at the other end of the substack.
4. The system according to claim 3, wherein the second substack has one or more intermediate electrical contacts.
5. The system according to claim 3 or 4, wherein the cell stack comprises a plurality of the second substacks.
6. The system according to any one of the preceding claims, wherein at least two of the cell groups have the same number of electrolysis cells.
7. The system according to any one of the preceding claims, wherein at least two of the cell groups have different numbers of electrolysis cells.
8. The system according to claim 7, wherein the different numbers of electrolysis cells are defined by a geometric sequence expressed as 201-1), where n 0.
9. The system according to claim 1, comprising two or more electrolyzers and respective electrical circuits, and switching circuity to switch the connection between the electrolyzers from a series connection to a parallel connection and vice versa, wherein the controller is configured to control the switching circuitry to connect the electrolyzers in series or in parallel, determine the number of active electrolysis cells for each of the electrolyzers, and control the respective electrical circuits based on the determination.
10. The system according to any one of the preceding claims, wherein the electrical energy source comprises one or more of: a renewable energy source or an electric energy storage device.
11. The system according to claim 10, wherein the controller implements a maximum power point tracking (MPPT) algorithm on the renewable energy source in the determination of the number of active electrolysis cells.
12. The system according to any one of the preceding claims, wherein the controller implements an energy management policy that regulates the amount of DC electrical energy to be supplied to the cell stack.
13. The system according to any one of the preceding claims, wherein the determination is further based on one or more of: an operating condition of the electrical energy source; an operating condition of the cell stack; or an operating condition of the electric energy storage device.
14. The system according to claim 13, wherein the operating condition of the electrical energy source comprises one or more of: an irradiation level on a solar PV array, a temperature of the solar PV array, a voltage and current generated by a wind turbine or other type of electrical generator.
15. The system according to claim 13, wherein the operating condition of the cell stack comprises a temperature of the cell stack.
16. The system according to claim 13, wherein the operating condition of the electric energy storage device comprises one or more of: a Status of Charge (SoC) of the electric energy storage device, or a voltage of the electric energy storage device.
17. The system according to any one of the preceding claims, wherein the controller is configured to partition the electrical energy between the electrolyzer stack and another electrical energy utilization system, controlling the relative amount of electrical energy that goes to each through an appropriate determination of the number of active electrolysis cells.
18. The system according to claim 17, wherein the electrical energy utilization system comprises an energy storage battery or a grid connected inverter.
19. The system according to any one of the preceding claims, wherein the controller is configured to regulate the rate of usage of the energy stored in the battery and going into the electrolyzer, through an appropriate determination of the number of active electrolysis cells.
20. The system according to any one of the preceding claims, further comprising multiple voltage sources configured to supply a maintenance voltage to a disconnected cell group or to the whole cell stack when not in use.
21. The system according to any one of the preceding claims, wherein the controller is further configured to control the electrical circuit to electrically insulate one or more of the cell groups, thus allowing to apply a maintenance voltage to the inactive cells or cell groups, while the rest of the stack is electrically connected and operative.
22. The system according to any one of the preceding claims, wherein the controller is further configured to control the electrical circuit to put in short circuit one or more unused cell groups, in order to protect them against polarity inversion when detached from stack operation.
23. An electrolyzer, comprising: a plurality of electrolysis cells arranged in a cell stack, wherein the electrolysis cells are electrically connected in series and grouped into two or more cell groups, each cell group having an externally accessible electrical contact at either end to allow the cell group to be electrically disconnected from the cell stack during operation of the electrolyzer, to thereby vary the number of active electrolysis cells in the cell stack.
24. A controller, comprising: a processor and memory, the memory storing instructions that, when executed by a processor, cause the processor to: determine a number of active electrolysis cells to be used from among a plurality of electrolysis cells arranged in a cell stack of an electrolyzer, wherein the electrolysis cells are electrically connected in series and grouped into two or more cell groups, each cell group having an externally accessible electrical contact at either end, wherein the determination is based on a variable amount of direct current (DC) electrical energy supplied to the cell stack by an electrical energy source and/or by a policy defined by a user, and control one or more switches, each of which is coupled between the electrical contacts of a respective one of the cell groups, to selectively disconnect one or more of the cell groups from the cell stack based on the determination.
25. The controller according to claim 24, wherein the processor is further configured to control one or more voltage sources and associated electrical switches to apply a maintenance voltage to a cell group or cell groups that are not in use.
26. The controller according to claim 24 or 25, wherein the processor is further configured to apply a short circuit to a cell group or cell groups that are not in use.
27. The controller according to any one of claims 24 to 26, wherein the processor is further configured to: determine a first number of active electrolysis cells to be used from among a first plurality of electrolysis cells arranged in a first cell stack of a first electrolyzer, wherein the first plurality of electrolysis cells are electrically connected in series and grouped into two or more cell groups, each cell group having an externally accessible electrical contact at either end, and control one or more first switches to selectively disconnect one or more of the cell groups from the first cell stack based on the determination, determine a second number of active electrolysis cells to be used from among a second plurality of electrolysis cells arranged in a second cell stack of a second electrolyzer, wherein the second plurality of electrolysis cells are electrically connected in series and grouped into two or more cell groups, each cell group having an externally accessible electrical contact at either end, and control one or more second switches to selectively disconnect one or more of the cell groups from the second cell stack based on the determination, and control switching circuitry to connect the first and second electrolyzers in series or in parallel.