Grid-forming converter and control method thereof

Abstract

A grid-forming converter (10) is provided. The grid-forming converter (10) is coupled between a power network (20) and an electrolyzer (30) for transferring electrical power from the power network (20) to the electrolyzer (30). The grid-forming converter (10) comprises a converter controller (11) configured to receive power network-related parameters for grid-forming control in the power network (20); determine at least one adjustment value for at least one of the power network-related parameters, wherein the at least one of the power network-related parameters comprises at least one of frequency droop related control parameter and inertia related control parameter; receive at least one electrolyzer-related parameter for controlling operation of the electrolyzer (30); determine a DC power reference value for powering the electrolyzer based on the at least one electrolyzer-related parameter; determine a reference active power based on the DC power reference value and the at least one adjustment value; determine a reference reactive power based on at least one of the power network-related parameters, wherein the at least one of the power network-related parameter comprises a power network voltage related parameter; and operate the grid-forming converter (10) to provide coordinated control for both the operation of the electrolyzer (30) and the provision of grid-forming control in the power network (20) by providing gate signals to power devices in the grid-forming converter based on the determined reference active power and reference reactive power.

Description

GRID-FORMING CONVERTER AND CONTROL METHOD THEREOFTECHNICAL FILED

[0001] The present disclosure relates to a grid-forming converter for transferring power from a power network to a hydrogen electrolyzer and a method for controlling the grid-forming converter.BACKGROUND

[0002] Given the need to transition to a net-zero emissions energy system, hydrogen technology is emerging as a key catalyst in this transformation. Hydrogen is set to play a dual role in the future energy landscape, serving both as an energy carrier and as a crucial component of energy storage solutions. Consequently, significant efforts are underway to advance the fundamental technologies for efficient hydrogen production. Hydrogen Electrolyzers are used for the generation of hydrogen and are desirably powered by renewable sources. These renewable power sources are often part of Medium Voltage (MV) and High Voltage (HV) power systems and power the hydrogen electrolyzers through power electronics converters. To improve the operation of hydrogen electrolyzers and other loads connected with renewable power sources, it is desirable to operate the power converters with a grid-forming control as grid strength and inertia may dynamically change in time. Therefore, it is desirable to provide a grid-forming (GFM) converter to control hydrogen production when integrating hydrogen electrolyzers into the power grid that enables robust operation and ancillary service provision.SUMMARY

[0003] According to an embodiment of the present disclosure, a grid-forming converter is provided. The grid-forming converter is coupled between a power network and an electrolyzer to transfer electrical power from the power network to the electrolyzer. The grid-forming converter comprises a converter controller configured to receive power network -related parameters for grid-forming control in the power network; determine at least one adjustment value for at least one of the power network-related parameter, wherein the at least one of the power network-related parameters comprises at least one of frequency droop related controlparameter and inertia control / emulation related control parameter; receive at least one electrolyzer-related parameter for controlling the operation of the electrolyzer; determine a DC power reference value for powering the electrolyzer based on the at least one electrolyzer- related parameter; determine a reference active power based on the DC power reference value and the at least one adjustment value; determine a reference reactive power based on at least one of the power network-related parameters, wherein the at least one of the power network- related parameter comprises a power network voltage related parameter; and operate the gridforming converter to provide coordinated control for both the operation of the electrolyzer and the provision of grid-forming control in the power network by providing gate signals to power devices in the grid-forming converter based on the determined reference active power and reference reactive power.

[0004] According to another embodiment of the present disclosure, a method for controlling a grid-forming converter is provided. The grid-forming converter is coupled between a power network and an electrolyzer to transfer electrical power from the power network to the electrolyzer. The method comprises the steps of receiving power network -related parameters for grid-forming control in the power network; determining at least one adjustment value for at least one of the power network-related parameters, wherein the at least one of the power network-related parameters comprises at least one of frequency droop related control parameter and inertia related control parameter; receiving at least one electrolyzer-related parameter for controlling the operation of the electrolyzer; determining a DC power reference value for powering the electrolyzer based on the at least one electrolyzer-related parameter; determining a reference active power based on the DC power reference value and the at least one adjustment value; determining a reference reactive power based on at least one of the power network- related parameters, wherein the at least one of the power network-related parameter comprises a power network voltage related parameter; and operating the grid-forming converter to provide coordinated control for both the operation of the electrolyzer and the provision of gridforming control in the power network by providing gate signals to power devices in the gridforming converter based on the determined reference active power and reference reactive power.

[0005] According to yet another embodiment of the present disclosure, a power supply systemis provided. The power supply system comprises a grid-forming converter coupled between a power network and an electrolyzer. The power supply system is configured to control at least one electrical parameter relating to the power network and power transferred to the electrolyzer by controlling the grid-forming converter based on power network-related parameters received from a power network controller associated with the power network and electrolyzer -related parameters received from an electrolyzer controller associated with the electrolyzer.BRIEF DESCRIPTION OF THE DRAWINGS

[0006] The disclosed aspects will hereinafter be described in connection with the appended drawings, which are provided to illustrate but not to limit the scope of the present disclosure.

[0007] Figure 1 illustrates a GFM converter according to an example of the present disclosure.

[0008] Figure 2 illustrates an embodiment of the GFM controller illustrated in Figure 1.

[0009] Figure 3 is a flowchart of a method for controlling the GFM converter according to an example of the present disclosure.

[0010] Figures 4-7 show embodiments of the main steps of the method illustrated in Figure 3.DETAIEED DESCRIPTION

[0011] The present disclosure proposes a GFM converter for a hydrogen electrolyzer and a control method for controlling the GFM converter, which can provide coordinated control for both the operation of the hydrogen electrolyzer and the provision of grid-forming control in the power network.

[0012] Embodiments of the present disclosure will now be described with reference to the drawings.

[0013] Figure 1 shows a grid-forming (GFM) converter 10 according to an embodiment of the present disclosure. The GFM converter 10 is coupled between a power network 20 and an electrolyzer 30 to transfer power from the power network 20 to the electrolyzer 30.

[0014] The GFM converter 10 can be implemented in various ways, such as a single-stage power converter (e.g., an AC-DC converter), a two-stage power converter (e.g., an AC-DC converter and a DC-DC converter), a modular multilevel converter (MMC), a multi-port converter, or parallel-connected converters. The specific implementation including theconverter topology of the GFM converter 10 depends on the application scenario. The control of the GFM converter is performed by a converter controller 11.

[0015] In an example, the converter controller 11 can be implemented with at least two control units (i.e., at least two modules). One control unit (e.g., a local controller module), shown in Figure 1 as the first control unit 11A is specifically designed to provide gate signals to power devices in the GFM converter 10 is implemented according to the specific converter topology and way of implementation of the GFM converter. Thus, with the gate signals, the converter controller 11 controls the transfer of power from the power network 20 to the hydrogen electrolyzer 30. The other control unit (i.e., the second control unit 11B) provides power setpoints to the first control unit 11A to cause the first control unit 11A operate the power devices to deliver power according to the received setpoint. The second control unit 11B is configured to provide coordinated control for the operation of the hydrogen electrolyzer 30 and the provision of grid-forming control in the power network 20. The second control unit 11B comprises control models for at least one of the operation of the hydrogen electrolyzer 30 and the provision of the GFM control, and provides coordinated control with electrical parameter measurements made at the input and output of the GFM converter 10 or in coordination with other controllers (e.g. a power network controller and an electrolyzer controller).

[0016] The parameters measured at the input of the GFM converter 10 and received by the converter controller 11 are referred to as power network-related parameters and relate to gridforming control. These parameters also include estimated values derived from the measured values. For example, some parameters that are not available via measurements can be obtained through an estimation process in the converter controller. Similarly, the parameters measured at the output of the GFM converter 10 and received by the converter controller 11 are referred to as electrolyzer-related parameters and relate to controlling the operation of the hydrogen electrolyzer 30. These parameters also include estimated values derived from the measured values. For example, some parameters that are not available via measurements can be obtained through an estimation process in the converter controller. In the embodiment depicted in Figure 2, the converter controller 11 can receive one or more power network-related parameters and electrolyzer-related parameters from other controllers communicatively connected with the converter controller 11. The other controllers can be associated with the power network 20,renewable energy sources, electrolyzer units, or / and with loads and other resources (energy storage systems) connected in the power network 20. In yet another embodiment, some of the power network-related parameters and electrolyzer-related parameters can be obtained from user inputs or through a pre-setting in the converter controller 11.

[0017] The power network 20 is an AC power grid in an embodiment and in another embodiment is a microgrid with renewable energy sources. In yet another embodiment, the power network 20 can include an AC power grid, DC power grid, and a microgrid in various combinations and connect one or more power sources for powering the hydrogen electrolyzer 30. The power network 20 integrates one or more renewable energy sources, such as photovoltaic cells and wind turbine generators to power the hydrogen electrolyzer 30 along with an AC power grid. Specifically, the power network 20 can coordinate and optimize the electric energy from different sources to the loads (e.g. electrolyzers) connected to the power network 20 with a power network controller 21.

[0018] The power network controller 21 can obtain measured values of electrical parameters of the power network 20 through a measurement system / equipment connected to the power network 20, and can also obtain information relating to the availability and dynamic capability of a renewable power source to power the hydrogen electrolyzer 30. The measurement system in the power network 20 includes, for example, instrument transformers, intelligent electronic devices (lEDs), and phasor measurement units (PMU) for measuring the following exemplary electrical parameters: voltage amplitude (instantaneous / RMS), current amplitude (instantaneous / RMS), phase difference, voltage / current phase (instantaneous / RMS), frequency, active power, reactive power, and harmonics. Referring to description in earlier paragraphs, the GFM converter 10 can also directly gather measured electrical parameters from a voltage or / and current sensor or / and equipment (e.g. instrument transformers) for GFM control. These measured electrical parameters can also be obtained by the GFM converter 10 from the power network controller 21 through a communication channel and used for grid-forming control of the power network 20. The measured electrical parameters used for GFM control are referred to as power network-related parameters.

[0019] The power network controller 21 can store reference electrical parameters of the power network 20, or receive these reference electrical parameters from other controllers, such as acentral control platform associated with the power grid. The reference electrical parameters of the power network 20 can include reference frequency, reference voltage, reference current, reference power factor, grid impedance, phase angle, and harmonic content.

[0020] The power network controller 21 can receive power-related information. For example, the power network controller 21 can receive power demand data, such as a curve indicating the change in power demand at various time periods throughout the day, which can help generate user power demand modes at different time periods. The power network controller 21 can also receive information related to electricity prices, for example, a fluctuation curve of electricity prices at various times of the day, which provides a basis for optimizing power usage based on changes in electricity prices. Based on such information, the power network controller 21 can control the power network 20.

[0021] Power network 20 having renewable energy sources can find dynamic change in the behavior of the grid according to the availability and utilization of the renewable energy sources, energy storage systems, rotating machine generators or / and loads, and static compensation devices connected in the power network 20. The behavior of the grid can be regulated by controlling active power consumption, providing or absorbing reactive power in the power network 20, and providing (improving) inertia in the power network 20. Such regulation can also be provided by the GFM converter 10 used for powering the hydrogen electrolyzers 30. The GFM converter 10 can support regulating electrical parameters such as voltage, frequency, frequency droop, power quality with balanced active and reactive power generation and consumption, and by providing inertia in the power network 20. Such electrical parameters that affect grid behavior are referred to as the power network-related parameters in the invention.

[0022] The hydrogen electrolyzer 30 (referred to as the electrolyzer in some places) uses electricity to decompose water into hydrogen and oxygen during the electrolysis process. The electrolyzer produces hydrogen through the electrolytic process. The electrolyzer 30 can be configured to include multiple electrolysis cells, which can be flexibly connected in series, parallel, or a combination of series and parallel to accommodate different requirements and optimize efficiency. To enhance the reliability and fault tolerance of the system, each electrolysis cell is coupled with a bypass switch. Once an electrolysis cell fails, it can be quicklybypassed to ensure that the normal operation of the remaining electrolysis cells is not affected.

[0023] An electrolyzer controller 31, which is associated with the electrolyzer 30, manages the operation of the electrolyzer 30. The electrolyzer controller 31 can be integrated with the electrolyzer 30 to form an electrolyzer with a local controller (i.e., control functions are built into the electrolyzer). The electrolyzer controller 31 can also be an independent device that is communicatively connected to the electrolyzer 30 to manage its operation including the operation of the auxiliary equipment (e.g. pumps, cooler, etc.).

[0024] A measurement system (not shown) is provided at the electrolyzer 30 for measuring the current, voltage, and temperature of the electrolyzer 30, as well as water flow rates at an inlet and outlet of a cooling system of the electrolyzer 30, and these measured values are used by the electrolyzer controller 31 to control the operation of the electrolyzer 30 and provide safety / protection to the various units / equipment use for hydrogen production. In addition, the measurement system can also measure the pH value, density, and conductivity of the electrolyte during the electrolysis process. These measured values can be obtained by the electrolyzer controller 31 from the measurement system. In an embodiment, the electrolyzer controller 31 can calculate the performance state of the electrolyzer 30 based on one or more of these measured values, such as the internal resistance, aging degree, health state, and reaction time of the electrolyzer 30. In another embodiment, converter controller 11 can obtain one or more of these measured values from the electrolyzer controller 31 and calculate the performance state of the electrolyzer 30.

[0025] The electrolyzer controller 31 stores the following parameters related to the electrolyzer 30: 1) the type of the electrolyzer 30; 2) physical characteristics of the electrolyzer 30, including a voltage-current characteristic curve of the electrolyzer 30 at a nominal temperature and a temperature change curve relative to operating hours under normal operating conditions; and 3) protection limits that restrict the operation of the electrolyzer 30. These parameters are referred to as electrolyzer-related parameters in the invention. In an embodiment, the converter controller 11 can obtain one or more of these electrolyzer-related parameters from the electrolyzer controller 31.

[0026] The protection limits can include a current limit parameter. The current limit parameter includes: 1) the long-term maximum operating current, which is the maximum current for theelectrolyzer 30 to normally operate over a long period of time; 2) the short-term maximum operating current which is greater than the long-term maximum operating current and less than or equal to the maximum critical current and serves as a buffer especially for the grid-forming control, and the electrolyzer 30 can operate at this current for a short period of time; 3) the long-term minimum operating current, which is the minimum current that the electrolyzer 30 can operate at for a long period of time; and 4) the short-term minimum operating current, which is less than the long-term minimum operating current and greater than or equal to the minimum critical current and serves as a buffer especially for the grid-forming control, and the electrolyzer 30 can operate at this current for a short period of time.

[0027] The protection limits can include a voltage limit parameter. The voltage limit parameter includes: 1) the long-term maximum operating voltage, which is the maximum voltage that the electrolyzer 30 can operate at for a long period of time; 2) the short-term maximum operating voltage which is greater than the long-term maximum operating voltage and less than or equal to the maximum critical voltage and serves as a buffer especially for the grid-forming control, and the electrolyzer 30 can operate at this voltage for a short period of time; 3) the long-term minimum operating voltage, which is the minimum voltage that the electrolyzer 30 can operate at for a long period of time; and 4) the short-term minimum voltage which is less than the longterm minimum operating voltage and greater than or equal to the minimum critical voltage and serves as a buffer especially for the grid-forming control, and the electrolyzer 30 can operate at this voltage for a short period of time.

[0028] The protection limits can include a power change limit. The power change limit includes a ramp-up rate limit for constraining the increasing rate of power supplied to the electrolyzer and a ramp-down rate limit for constraining the decreasing rate of power supplied to the electrolyzer. In this way, the power supplied to the electrolyzer cannot change too fast, to protect the electrolyzer. The ramp-up rate limit and ramp-down rate limit can be the same, that is, the maximum increasing rate is equal to the maximum decreasing rate in terms of power change. The ramp-up rate limit and ramp-down rate limit can also be different, that is, it can increase at a faster rate and decrease at a slower rate in terms of power change, or vice versa.

[0029] Values of the protection limits are predetermined according to the type and physical characteristics of the electrolyzer 30 and stored in the electrolyzer controller 31. Theseprotection limits can be set for the operation of the electrolyzer 30 through protection circuitry that can ensure the operation of the electrolyzer 30 within the protection limits. In this invention, such limits are advantageously considered in the configuration of the converter controller 11. For example, the control models used in the second control unit 1 IB consider such limits while determining power setpoints for the operation of the GFM converter 10 to power the electrolyzer 30. These limits may be dynamically changed according to the performance of the electrolyzer 30 or to meet a short-term requirement relating to hydrogen production or gridforming control, and the setpoints for operating the converter controller can be adjusted accordingly.

[0030] In an embodiment, a minimal control configuration e.g. for control of auxiliary devices of the electrolyzer 30 can be provided in the electrolyzer controller 31, and the converter controller 11 can determine the performance and health of the electrolyzer 30 with measurements made at the output of the GFM converter 10 and correlating the measured values with the output (e.g. hydrogen production rate) of the electrolyzer 30 received from the electrolyzer controller 31. In an example of an integrated converter-electrolyzer plant, the value of the output of the electrolyzer plant can also be provided as feedback to the converter controller 11 for adjusting the power transferred to the electrolyzer 30.

[0031] In an example, one or more of the aforementioned protection limits can be time-varying. For example, all the protection limits may change over time due to factors such as temperature, ageing, or the performance state of the electrolyzer 30. Next, an embodiment illustrating such time-varying protection limits will be introduced, using the example of how the protection limits adjust according to the performance state of the electrolyzer 30.

[0032] The converter controller 11 can adjust these protection limits according to the change in the performance state of the electrolyzer 30 based on the received electrolyzer-related parameters. For example, due to the ambient temperature fluctuation violently or long-term operation, the performance state of the electrolyzer 30 has changed beyond a threshold level, requiring a reset of the protection limits initially set for its operation. In this case, the converter controller 11 will adjust one or more of these protection limits in the control models configured in the converter controller 11 for determining setpoints for the operation of the GFM converter 10. In one embodiment, when at least one of the following situations occurs, the electrolyzercontroller 31 or the converter controller 11 determines that the change in the performance state of the electrolyzer 30 is beyond the threshold level: 1) the aging rate of the electrolyzer 30 becomes greater than an aging rate threshold; 2) a health index of the electrolyzer 30 indicates that the health of the electrolyzer 30 has decreased to a health drop threshold; and 3) the internal resistance of the electrolyzer 30 increases to exceed an internal resistance threshold.

[0033] It is noted that in the above description, the threshold level and multiple thresholds (e.g., the long-term maximum and minimum operating currents and voltages, the short-term maximum and minimum operating currents and voltages) are involved, which are predetermined and stored in the electrolyzer controller 31 and communicated to the converter controller 11 or are available directly with the converter controller 11 as a presetting or determined over time based on the received electrolyzer -related parameters, and the converter controller 11 can adjust them according to specific application scenarios.

[0034] It is noted that these protection limits, as key critical protection values, are relatively stable in their settings and only undergo necessary regulation when there are significant changes in the performance state of the electrolyzer 30. This regulation mechanism aims to ensure that the electrolyzer 30 can maintain a stable and safe operating state. Moreover, the step size of regulation, that is, the increase or decrease amount of each regulation and the maximum regulation amount, have been predetermined to ensure that all regulations are made within the predetermined regulation range, thereby ensuring the accuracy and effectiveness of the regulation.

[0035] The converter controller 11 or a plant level controller (not shown) provides supervisory control by providing setpoints to the electrolyzer controller 31 and converter controller 11 receives a reference hydrogen production rate, that is, a target hydrogen production rate. The reference hydrogen production rate can be expressed as the hydrogen production rate per unit time or as the reference hydrogen production amount over a period of time. The reference hydrogen production rate can be determined based on the actual needs of hydrogen users, the requirements of hydrogen storage capacity, or a combination of hydrogen production needs and storage requirements, ensuring that the electrolyzer operates in accordance with actual application scenarios and needs.

[0036] The converter controller 11 can be communicatively connected to the power networkcontroller 21 and the electrolyzer controller 31 to receive power network -related parameters from the power network controller 21 and electrolyzer-related parameters from the electrolyzer controller 31. The converter controller 11 determines the active power setpoint (Pset) and the reactive power setpoint (Qset) based on the received power network-related parameters and electrolyzer-related parameters and using the control models configured in the second control module 1 IB of the converter controller 11. Based on the determined active power setpoint (Pset) and the reactive power setpoint (Qset), the first control module 11A determines gate signals (i.e., gate control signals), for controlling the GFM converter 10. This control mechanism not only ensures that the physical characteristics and protection limits of the electrolyzer 30 are fully considered when supplying power to the electrolyzer 30, but also provides grid-forming control to the power network 20. In other words, by controlling the GFM converter 10, coordinated control and optimization of both the power supplied to the electrolyzer 30 for hydrogen production and grid-forming control for the power network 20 are achieved.

[0037] In an embodiment, the first controller unit 11A and the second controller unit 11B of the converter controller 11 can be integrated with the GFM converter 10. For example, by integrating the converter controller 11 within the GFM converter 10. The first controller unit 11 A and / or the second controller unit 1 IB of the converter controller 11 can also be provided independently from the GFM converter 10. These two configurations can be adapted to different application scenarios. In another embodiment, the second controller unit 11B can be integrated into the electrolyzer controller 31. In yet another embodiment, the second controller unit 11B can be integrated into the power network controller 21, especially in a microgrid controller when the power network 20 is a microgrid. In still another embodiment, the second controller unit 11B can be integrated in a hydrogen plant controller as a part of a central controller providing setpoints to various local controllers (e.g. the first controller unit 11A of the converter 10 and the electrolyzer controller 21). In still another embodiment, the second control unit 11B can be provided as an advanced controller configured in a cloud server, an edge server, or a central control platform.

[0038] Figure 3 is a flowchart of a method 200 for controlling a GFM converter 10 according to an embodiment of the disclosure. The operations of the method 200 do not necessarily need to be performed in the order described. Instead, some operations can be carried out in a differentsequence or simultaneously, and additional operations may be included, or some may be omitted.

[0039] With reference to Figure 3, at block 302, the converter controller 11 receives electrolyzer-related parameters, for example, from one or more means such as the electrolyzer controller 21, direct measurements made at the output of the converter, or user settings. The electrolyzer-related parameters include: measurement parameters that indicate the operating state of the electrolyzer 30 (e.g., the voltage and current of the electrolyzer 30 and the temperature of the cooling water), calculation parameters that indicate the performance state of the electrolyzer 30 (e.g., the internal resistance, rate and extent of aging, health state, efficiency, reaction time of the electrolyzer 30), protection limits of the electrolyzer 30 (e.g., the current limit parameters and voltage limit parameters of the electrolyzer 30), the type of electrolyzer 30 (e.g., ALK / AEC, PEM / PEMEC, SOEC), and physical characteristics of the electrolyzer 30 (e.g., the voltage-current curve of the electrolyzer 30). It is noted that the above introduction of examples for these parameters is also applicable here.

[0040] At block 304, the converter controller 11 receives power network-related parameters from the power network controller 21. The power network-related parameters include measured electrical parameters of the power network 20 (e.g., voltage and frequency measured in the power network), and reference electrical parameters of the power network 20 (e.g., parameters representing power demand and parameters representing electricity prices). It is noted that the above introduction of examples for these parameters is also applicable here. In an example, a part of the power network-related parameters can be measured at the input of the converter 10 and another part of the power network -related parameters can be received from the power network controller 21 or provided by a user.

[0041] At block 306, the converter controller 11 determines reference power Pdc,ref for powering the electrolyzer 30 based on the electrolyzer-related parameters. For example, the converter controller 11 determines the reference power Pdc, ref based on the reference hydrogen production rate and protection limits of the electrolyzer 30.

[0042] Next, an embodiment of block 206 will be described with reference to Figure 4.

[0043] Referring to Figure 4, at block 3061, the converter controller 11 calculates a deviation AH2 between the reference hydrogen production rate H2, ref and the actual hydrogen productionrate H2 of the electrolyzer 30 and inputs the deviation to a first PID controller PID_1. The first PID controller processes the input deviation AH2 according to the preset proportional gain Kp, integral gain Ki, and derivative gain Ka, and outputs a reference DC current Ide, ref of the electrolyzer 30. The reference DC current is used to guide the operation of the electrolyzer 30 to ensure that the hydrogen production rate can be maintained at or as close to the reference hydrogen production rate as possible.

[0044] At block 3062, the converter controller 11 uses the current limit parameter of the electrolyzer 30 to constrain the reference DC current Ide, ref to obtain a constrained DC reference current Ide, ref-iim. Referring to block 3062, Il and 14 represent the short-term minimum and maximum operating currents, while 12 and 13 represent the long-term minimum and maximum operating currents. 12 and 13 define the long-term operating current range of the electrolyzer 30. Additionally, Il and 12 define a first short-term operating current range, and 13 and 14 define a second short-term operating current range for the electrolyzer 30. The first and second shortterm operating current ranges are specifically designed to provide grid-forming control, allowing the electrolyzer 30 to operate for a relatively short period of time. Therefore, according to some examples, the range of the reference DC current Ide, ref can include both the long-term operating current range and the first and second short-term operating current ranges.

[0045] In an embodiment, the converter controller 11 determines whether the reference DC current Ide, ref is greater than the short-term maximum operating current 14 or less than the shortterm minimum operating current II. If the judgment result is negative (i.e., the reference DC current Ide, ref is less than or equal to the current 14 and greater than or equal to the current II), then the value of the reference DC current Ide, ref is not changed. In this case, the value of Ide, ref- iim is equal to the value of Ide, ref. If the judgment result is affirmative, the converter controller 11 uses the short-term maximum or minimum operating current to constrain Ide, ref. Specifically, when the reference DC current Ide, ref is greater than the short-term maximum operating current 14, the converter controller 11 uses the short-term maximum operating current 14 to constrain Ide, ref, resulting in the value of Ide, ref-iim being equal to 14. When the reference DC current Ide, ref is less than the short-term minimum operating current II, the converter controller 11 uses the short-term minimum operating current II to constrain Ide, ref, resulting in the value of Ide, ref-iim being equal to II.

[0046] At block 3063, the converter controller 11 calculates a deviation Aide between the constrained DC reference current Ide, ref-iim and the actual current Ide of the electrolyzer 30 and inputs the deviation to a second PID controller PID_2. The second PID controller processes the input deviation Aide according to the preset proportional gain Kp, integral gain Ki, and derivative gain Kd, and outputs a reference DC voltage Vdc,ref of the electrolyzer 30. The reference DC voltage is used to guide the operation of the electrolyzer 30 to ensure that the actual current of the electrolyzer 30 can be maintained at or as close to the target current (i.e., the constrained DC reference current Ide, ref-iim) as possible.

[0047] At block 3064, the converter controller 11 uses the voltage limit parameter of the electrolyzer 30 to constrain the reference DC voltage Vdc, ref to obtain a constrained DC reference voltage Vdc, ref-iim. Referring to block 3064, VI and V4 represent the short-term minimum and maximum operating voltages, while V2 and V3 represent the long-term minimum and maximum operating voltages. V2 and V3 define the long-term operating voltage range of the electrolyzer 30. Additionally, V 1 and V2 define a first short-term operating voltage range, and V3 and V4 define a second short-term operating voltage range for the electrolyzer 30. The first and second short-term operating voltage ranges are specifically designed to provide grid-forming control, allowing the electrolyzer 30 to operate for a relatively short period of time. Therefore, according to some examples of the present disclosure, the range of the reference DC voltage Vdc, ref can include both the long-term operating voltage range and the first and second short-term operating voltage ranges.

[0048] In an embodiment, the converter controller 11 determines whether the reference DC voltage Vdc, ref is greater than the short-term maximum operating voltage V4 or less than the short-term minimum operating voltage VI . If the judgment result is negative (i.e., the reference DC voltage Vdc, ref is less than or equal to the voltage V4 and greater than or equal to the voltage VI), then the value of the reference DC voltage Vdc, ref is not changed. In this case, the value of Vdc, ref-iim is equal to the value of Ide, ref. If the judgment result is affirmative, the converter controller 11 uses the short-term maximum or minimum critical voltage to constrain Vdc, ref. Specifically, when the reference DC voltage Vdc, ref is greater than the short-term maximum operating voltage V4, the converter controller 11 uses the short-term maximum operating voltage V4 to constrain Vdc, ref, resulting in the value of Vdc, ref-iim being equal to the voltage V4.When the reference DC voltage Vdc, ref is less than the short-term minimum operating voltage VI, the converter controller 11 uses the short-term minimum operating voltage VI to constrain Vdc, ref, resulting in the value of Ide, ref-iim being equal to the voltage V 1.

[0049] At block 3065, the converter controller 11 calculates a deviation A Vdc between the constrained DC reference voltage Vdc, ref-iim and the actual voltage Vdc of the electrolyzer 30 and inputs the deviation to a third PID controller PID_3. The third PID controller processes the input deviation according to the preset proportional gain Kp, integral gain Ki, and derivative gain Kd, and outputs a reference power Pdc,ref for powering the electrolyzer 30. The reference power Pdc.ref is used to guide the operation of the electrolyzer 30 to ensure that the actual voltage of the electrolyzer 30 can be maintained at or as close to the target voltage, i.e., the constrained DC reference voltage Vdc, ref-iim.

[0050] It is noted that, in the examples described above, the PID controllers are used as an exemplary type of controller and can be replaced by other controllers that have the same functionality, such as those based on AI / ML or data-driven approaches.

[0051] Referring back to Figure 3, at block 308, the converter controller 11 determines a first adjustment parameter Pdroop for providing frequency droop control to the power network 20 based on parameters related to the frequency of the power network 30.

[0052] Next, an embodiment of block 308 will be described with reference to Figure 4.

[0053] Referring to Figure 5, the converter controller 11 calculates a frequency deviation Aco between a reference frequency coref and an actual frequency co of the power network 20. Then, the converter controller 11 inputs the frequency deviation Aco into a droop control gain module (Kdroop, G>). In this module, the mapping relationship between the frequency deviation and the active power adjustment is adjusted by a droop control gain, and the first adjustment parameter Pdroop which indicates this mapping relationship is obtained. The first adjustment parameter is used to dynamically adjust the active power output from the power network 20 to ensure that the frequency of the power network 20 returns to or stays as close as possible to the reference frequency, thereby achieving frequency droop control of the power network 20.

[0054] In addition to the control method based on the frequency deviation described with reference to Figure 5, other control methods can also be used to implement frequency droop control to achieve frequency stability in the power network 20. In one embodiment, a virtualimpedance can be introduced into a control algorithm to provide frequency droop control by adjusting the value of the virtual impedance. This control method is particularly suitable for microgrids and distributed power generation systems. In another embodiment where multiple distributed power sources operate in parallel in the power network, a distributed coordinationbased droop control method can be used to achieve global frequency stability. In this control method, the output active power is adjusted through communication and coordination between distributed power sources to maintain the frequency stability of the power network 20. In yet another embodiment, intelligent algorithms such as neural networks or fuzzy logic can be used to predict the frequency variation trend of the power network 20 and adjust the droop control gain accordingly. This control method can improve the robustness and control accuracy of frequency control in the power network 20.

[0055] Referring back to Figure 3, at block 310, the converter controller 11 determines a second adjustment parameter Pinertia for providing inertia to the power network 20 based on parameters related to the inertia of the power network 20. The inertia of the power network 20 is reflected in its ability to resist power disturbances and respond quickly to frequency changes. In one embodiment, an inertia emulation control system based on PLL (Phase-locked loop) can be used to calculate a mapping relationship between power fluctuations and active power adjustments in the power network 20 (also other types of inertia emulation based methods can be considered such as virtual synchronous machine, virtual synchronous generator, virtual oscillator-based methods, among others), thereby determining the second adjustment parameter to provide the required inertia support for the power network 20. In other embodiments, inertia support can also be provided to the power network 20 through virtual synchronous generator technology and distributed power coordination control. These methods are applicable to different grid scenarios and demands.

[0056] At block 312, the converter controller 11 determines a reference active power Pref based on the reference DC power, the first and second adjustment parameters for providing frequency droop control and inertia. In general, the converter controller 11 determines an optimal value of the reference active power based on three factors: the power supply of the electrolyzer, the frequency droop support, and the inertia support of the power network.

[0057] Next, an example of block 312 will be described with reference to Figure 6

[0058] Referring to Figure 6, the converter controller 11 configures proportional coefficients for the three factors mentioned above, namely: a first proportional coefficient Cl for regulating the reference DC power; a second proportional coefficient C2 for regulating the first adjustment parameter; and a third proportional coefficient C3 for regulating the second adjustment parameter. The value of each of Cl, C2, and C3 is between 0 and 1, that is, greater than or equal to zero and less than or equal to one

[0059] In an embodiment, the converter controller 11 sets each of the proportional coefficients to 1. This setting is based on the assumption that all factors have equal weight on the system in the initial stage, and this approach can provide a relatively neutral control foundation. This means providing a relatively balanced starting point for control, and facilitating subsequent fine-tuning based on actual needs.

[0060] In another embodiment, the converter controller 11 creates an optimization model to determine the optimal combination of proportional coefficients C1-C3 to minimize system losses or maximize revenue. The optimization model can be implemented as an objective function, aiming to minimize losses or maximize revenue. The proportional coefficients Cl, C2, and C3 are variables. Constraints include fluctuations in electricity prices, changes in the availability of green power from the network controller 21, the impact of electrolyzer aging on its performance from the electrolyzer controller 31, and the interactions between these factors. The converter controller 11 uses online or offline mathematical optimization algorithms such as genetic algorithms, particle swarm optimization, and gradient descent to obtain the optimal combination of values for Cl, C2, and C3 as the solution to the objective function. In this embodiment, the constraints can be adjusted accordingly based on different information obtained from the network controller 21 and from the electrolyzer controller 31.

[0061] Continuing to refer to Figure 6, the converter controller 11 can use the power change limit to correct the determined reference active power and use the corrected reference active power for subsequent processing. For example, the converter controller 11 can calculate the power change rate (ramp-up rate or ramp-down rate) according to the determined reference active power and the current active power, and judge whether the calculated power change rate exceeds the power change limit (i.e., a corresponding ramp-up or ramp-down rate limit). If the judgment result is negative, the reference active power is directly used without any correction;if the judgment result is positive, the power change limit (i.e., the ramp-up rate limit or rampdown rate limit) is used to correct the determined reference active power, resulting in the corrected reference active power

[0062] Referring back to Figure 3, at block 314, the converter controller 11 calculates a reference reactive power Qref of the power network 20 based on parameters measured in the power network 20, such as power factor, voltage, and current, to accommodate changes in its reactive power demand.

[0063] Next, an embodiment of block 314 will be described with reference to Figure 7.

[0064] Referring to Figure 7, the converter controller 11 calculates the deviation Avac.mag between the measured voltage Vac,mag and the reference voltage Vac,mag,ref of the power network 20 and inputs this deviation into a voltage droop control module Kdroop.v. In this module, the deviation is used to calculate the demand for voltage regulation, thereby obtaining and outputting a reactive power compensation parameter Qdroop. Next, based on Qdroop and a reference value of reactive power Qac, ref (which can be included in the power network-related parameters), the converter controller 11 obtains the reference reactive power Qref that combines the output of the droop control with the reference value Qac, ref.

[0065] In addition to the control method described above with reference to Figure 7, other control methods can also be used to achieve reactive power compensation. These methods aim to maintain the voltage stability of the power network 20 through different strategies and technologies. For example, voltage droop control based on power factor control can be used (Also automatic voltage regulator (AVR) type controllers can be used) . This method ensures proper compensation of reactive power by monitoring and adjusting the power factor to an appropriate range, thereby maintaining voltage stability. Additionally, intelligent control algorithms can be applied to voltage droop control. Algorithms such as PID control, fuzzy control, and neural network control can utilize the real-time state and historical data of the power network to calculate the reference reactive power. The advantages of intelligent control algorithms lie in their adaptability and accuracy, allowing for a quick response to changes in the power network and achieving stable voltage control.

[0066] Referring back to Figure 3, at block 316, the converter controller 11 determines an active power setpoint and a reactive power setpoint based on the reference active power andreference reactive power to limit the apparent power. For example, the converter controller 11 can reduce the reference active power and reference reactive power by the same proportion to ensure that the square root of the sum of their squares does not exceed a predetermined limit value, thereby limiting the apparent power.

[0067] At block 318, the converter controller 11 determines a control signal to control the converter 10, such as PWM signals used to control the power switches of the converter 10, based on the active power setpoint and the reactive power setpoint. It is noted that block 218 can be implemented in various ways, and the present disclosure does not limit the specific implementation of block 318

[0068] In an example, blocks 302-316 can be implemented by the first control unit 11A and block 318 can be implemented by the second control unit 1 IB.

[0069] According to examples of the present disclosure, a power supply system is provided. The power supply system includes the grid-forming converter 10 coupled between a power network 20 and an electrolyzer 30. The power supply system controls at least one electrical parameter relating to the power network 20 and power transferred to the electrolyzer 30 by controlling the grid-forming converter based on power network-related parameters received from a power network controller 21 associated with the power network 20 and electrolyzer- related parameters received from an electrolyzer controller 31 associated with the electrolyzer 30.

[0070] In an embodiment, the power supply system receives information about electrical power generated by the at least one renewable energy source from the power network controller 21 and adjusts a reference hydrogen production rate based on the electrical power generated by the at least one renewable energy source.

[0071] In an embodiment, the power supply system receives information on fluctuations in electricity prices and changes in the availability of green power from the network controller, receives information on the impact of electrolyzer aging on its performance from the electrolyzer controller 31 ; and determines an optimal combination of the first, second, and third proportional coefficients to minimize system losses or maximize revenue based on the received information from the network controller 21 and from the electrolyzer controller 31.

[0072] According to examples of the present disclosure, the proposed grid-forming controlarchitecture has various advantages. For example, it enables frequency / voltage droop control and inertia support in combination with the electrolyzer operation, all of which are provided through a single control operation realized in the integrated control of a converter controller. The integrated control method of the invention combines various grid-related measured electrical parameters and electrolyzer protection limit parameters together to produce setpoints to operate the converter controller. The control method of the invention can be applied to different converter designs for controlling the DC voltage to the electrolyzer (DC control), PQ limiter for controlling active and reactive power, and low -level converter control for providing gate signals to the converter. Here, in the invention a high-level control architecture is provided, therefore, it is compatible with different converter topologies and arrangements of multiple electrolyzers. For example, electrolyzers can be powered through a DC-DC converter or through an AC-DC converter. Further, an optimization layer can be provided as an additional layer of control, and thereby combining optimization with the proposed control solution to balance hydrogen provision with ancillary grid service provision, depending on the most profitable service through adaptive modification of control gains in the controller for the various control parameters (e.g. frequency control, voltage control, etc.). The control and protection limiters for the DC voltage or / and current to the electrolyzer, that is, the hydrogen generation control and the DC voltage control, can be merged into a unified nonlinear constrained power controller. Moreover, bypass mechanisms can also be controlled for the protection limiters and combined with the control method to disable the limiters provided for regular operation of electrolyzer during a start-up, shut-down and emergency operation situations, such as during failure of electrolyzer.

[0073] In some examples, all the aforementioned controllers have the flexibility to be either replaced by, or integrated with, controllers that leverage stochastic AI / ML technologies and other control techniques based on data-driven methodologies. By doing so, intelligence based on the specific conditions or states of the electrolyzer and / or power network can be built into the converter controller.

[0074] In some examples, one or more fuel cells can be combined with or included in the power system. For example, when a fuel cell is connected to the power grid through the GFM converter with the aforementioned control strategies, the electrical energy generated by the fuelcell can be injected into the power grid to provide active power support. During peak power demand periods, the fuel cell can quickly start up and increase power output, thereby alleviating the pressure on the power grid. Moreover, the fuel cell can also be combined with the GFM converter to provide reactive power support for the grid by adjusting the output voltage and current phase.

[0075] In addition, the fuel cell can be combined with the electrolyzer to improve energy consumption efficiency. For example, during periods of low power demand, excess electricity can be used to produce hydrogen through the electrolyzer for storage. During peak power demand, the stored hydrogen is converted into electrical energy through the fuel cell and supplied to the power grid. This interactive mode can effectively balance the load on the power grid and improve its flexibility and stability.

[0076] In addition, the fuel cell can also be combined with renewable energy sources, such as solar and wind energy, to form a complementary energy system. When renewable energy generation is insufficient, the fuel cell can provide additional power support; when there is an excess of renewable energy generation, the fuel cell can reduce its output or stop working, thereby providing a more reliable and stable power supply to the grid.

[0077] While the present disclosure has been illustrated in the appended drawings and the foregoing description, such illustration is to be considered illustrative or exemplifying and not restrictive; the present invention is not limited to the disclosed embodiments. Other variations to the disclosed embodiments can be understood and effected by those skilled in the art in practicing the claimed invention, from a study of the drawings, the disclosure, and the appended claims. In the appended claims, the word "comprising" does not exclude other elements or steps, and the indefinite article "a" or "an" does not exclude a plurality. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage. Any reference signs in the claims should not be construed as limiting the scope.

Claims

WHAT IS CLAIMED IS:

1. A grid-forming converter (10) coupled between a power network (20) and an electrolyzer (30) for transferring electrical power from the power network (20) to the electrolyzer (30), the grid-forming converter (10) comprising a converter controller (11) configured to: receive power network-related parameters for grid-forming control in the power network (20); determine at least one adjustment value for at least one of the power network -related parameter, wherein the at least one of the power network-related parameters comprises at least one of frequency droop related control parameter and inertia related control parameter; receive at least one electrolyzer-related parameter for controlling the operation of the electrolyzer (30); determine a DC power reference value for powering the electrolyzer based on the at least one electrolyzer-related parameter; determine a reference active power based on the DC power reference value and the at least one adjustment value; determine a reference reactive power based on at least one of the power network -related parameters, wherein the at least one of the power network -related parameters comprises a power network voltage related parameter; and operate the grid-forming converter (10) to provide coordinated control for both the operation of the electrolyzer (30) and the provision of grid-forming control in the power network (20) by providing gate signals to power devices in the grid-forming converter based on the determined reference active power and reference reactive power.

2. The grid-forming converter (10) of claim 1, wherein the electrolyzer-related parameters comprise one or more of:- measurements comprising measured current, voltage, and temperature of the electrolyzer- the performance state of the electrolyzer (30) comprising at least one of an aging degree, health state, and internal resistance of the electrolyzer (30);- a protection limit comprising a current limit, a voltage limit, and a maximum allowable rate of change in power of the electrolyzer (30);- a reference hydrogen production rate and / or volume;- the type of the electrolyzer; and- physical characteristics of the electrolyzer (30) comprising at least one of a voltagecurrent curve under the nominal temperature, a voltage offset according to the temperature of the electrolyzer (30), and a response time of the electrolyzer (30).

3. The grid-forming converter (10) of claim 2, wherein the converter controller (11) is configured to adjust the protection limit by a predetermined amount if a change in the performance state of the electrolyzer (30) is greater than a threshold level.

4. The grid-forming converter (10) of claim 1, wherein the power network -related parameters comprise one or more of:- measurements of the power network (20) comprising at least one of voltage amplitude, current amplitude, frequency, active power, reactive power, and harmonics;- information on power demands; and- information on electricity prices.

5. The grid-forming converter (10) of claim 1, wherein the converter controller (11) is configured to: control the frequency in the power network (20) by performing frequency droop control and providing inertia in the power network (20) using an inertia emulation control loop; and control the voltage in the power network (20) by performing voltage droop control.

6. The grid-forming converter (10) of claim 1, wherein the converter controller (11) is configured to determine the reference DC power by:determining a reference DC current based on a reference hydrogen production rate or volume and a current limit parameter of the electrolyzer (30); determining a reference DC voltage based on the reference DC current and a voltage limit parameter of the electrolyzer (30); and determining the reference DC power based on the reference DC voltage.

7. The grid-forming converter (10) of claim 6, wherein the current limit parameter comprises a short-term maximum operating current that is greater than a long-term maximum operating current and less than the maximum critical current of the electrolyzer (30), and a short-term minimum operating current that is less than a long-term minimum operating current and greater than the minimum critical current of the electrolyzer (30); and the voltage limit parameter comprises a short-term maximum operating voltage that is greater than a long-term maximum operating voltage and less than the maximum critical voltage of the electrolyzer (30), and a short-term minimum operating voltage that is less than a long-term minimum operating voltage and greater than the minimum critical voltage of the electrolyzer (30).

8. The grid-forming converter (10) of claim 1, wherein the converter controller (11) is configured to obtain an optimal combination of first, second, and third proportional coefficients by solving an objective function, and wherein the first proportional coefficient is for regulating the reference DC power, the second proportional coefficient is for regulating the first adjustment parameter, and the third proportional coefficient is for regulating the second adjustment parameter.

9. The grid-forming converter (10) of claim 1, wherein the power network (20) is coupled with at least one renewable energy source, and the converter controller (11) is configured to: receive information about electrical power generated by the at least one renewable energy source from a power network (21) controller associated with the power network (20); and adjust the operation of the electrolyzer (30) by adjusting a reference hydrogen production rate based on the electrical power generated by the at least one renewable energy source.

10. A method for controlling a grid-forming converter (10) coupled between a power network (20) and an electrolyzer (30) for transferring electrical power from the power network (20) to the electrolyzer (30), the method comprising: receiving power network-related parameters for grid-forming control in the power network (20); determining at least one adjustment value for at least one of the power network -related parameters, wherein the at least one of the power network-related parameters comprises at least one of frequency droop related control parameter and inertia related control parameter; receiving at least one electrolyzer-related parameter for controlling operation of the electrolyzer (30); determining a DC power reference value for powering the electrolyzer based on the at least one electrolyzer-related parameter; determining a reference active power based on the DC power reference value and the at least one adjustment value; determining a reference reactive power based on at least one of the power network-related parameters, wherein the at least one of the power network -related parameters comprises a power network voltage related parameter; and operating the grid-forming converter (10) to provide coordinated control for both the operation of the electrolyzer (30) and the provision of grid-forming control in the power network (20) by providing gate signals to power devices in the grid-forming converter based on the determined reference active power and reference reactive power.

11. The method of claim 10, wherein the electrolyzer-related parameters comprise one or more of:- measurements comprising measured current, voltage, and temperature of the electrolyzer (30);- the performance state of the electrolyzer (30) comprising at least one of an aging degree, health state, and internal resistance of the electrolyzer (30);- a protection limit comprising a current limit, a voltage limit, and a maximum allowable rate of change in power of the electrolyzer (30);- a reference hydrogen production rate and / or volume;- the type of the electrolyzer (30); and- physical characteristics of the electrolyzer (30) comprising at least one of a voltagecurrent curve under the nominal temperature, a voltage offset according to the temperature of the electrolyzer (30), and a response time of the electrolyzer (30).

12. The method of claim 11, wherein the method comprises adjusting the protection limit by a predetermined amount if a change in the performance state of the electrolyzer (30) is greater than a threshold level.

13. The method of claim 10, wherein the power network-related parameters comprise one or more of:- measurements of the power network (20) comprising at least one of voltage amplitude, current amplitude, frequency, active power, reactive power, and harmonics;- information on power demands; and- information on electricity prices.

14. The method of claim 10, wherein determining the reference DC power comprises: determining a reference DC current based on a reference hydrogen production rate or volume and a current limit parameter of the electrolyzer (30); determining a reference DC voltage based on the reference DC current and a voltage limit parameter of the electrolyzer (30); and determining the reference DC power based on the reference DC voltage.

15. The method of claim 14, wherein the current limit parameter comprises a short-term maximum operating current that is greater than a long-term maximum operating current and less than the maximum critical current of the electrolyzer (30), and a short-term minimumoperating current that is less than a long-term minimum operating current and greater than the minimum critical current of the electrolyzer (30); and the voltage limit parameter comprises a short-term maximum operating voltage that is greater than a long-term maximum operating voltage and less than the maximum critical voltage of the electrolyzer (30), and a short-term minimum operating voltage that is less than a long-term minimum operating voltage and greater than the minimum critical voltage of the electrolyzer (30).

16. The method of claim 10, the method comprising: obtaining an optimal combination of first, second, and third proportional coefficients by solving an objective function, wherein the first proportional coefficient is for regulating the reference DC power, the second proportional coefficient is for regulating the first adjustment parameter, and the third proportional coefficient is for regulating the second adjustment parameter.

17. The method of claim 10, the power network comprising at least one renewable energy source, and the method comprising: receiving information about electrical power generated by the at least one renewable energy source from a power network controller (21) associated with the power network (20); and adjusting the operation of the electrolyzer by adjusting a reference hydrogen production rate based on the electrical power generated by the at least one renewable energy source.

18. A power supply system comprising: a grid-forming converter (10) coupled between a power network (20) and an electrolyzer (30), wherein the power supply system is configured to control at least one electrical parameter relating to the power network (20) and power transferred to the electrolyzer (30) by controlling the grid-forming converter based on power network-related parameters received from a powernetwork controller (21) associated with the power network (20) and electrolyzer -related parameters received from an electrolyzer controller (31) associated with the electrolyzer (30).

19. The power supply system of claim 18, wherein the power network (20) is coupled with at least one renewable energy source, and the power supply system is configured to: receive information about electrical power generated by the at least one renewable energy source from the power network controller (21); and adjust a reference hydrogen production rate based on the electrical power generated by the at least one renewable energy source.

20. The power supply system of claim 18, wherein the power supply system is configured to: receive information on fluctuations in electricity prices and changes in the availability of green power from the network controller (21); receive information on the impact of electrolyzer aging on its performance from the electrolyzer controller (31); and determine an optimal combination of first, second, and third proportional coefficients to minimize system losses or maximize revenue based on the received information from the network controller (21) and from the electrolyzer controller (31), wherein the first proportional coefficient is for regulating a reference DC power for powering the electrolyzer (30), the second proportional coefficient is for regulating a first adjustment parameter for providing frequency droop control to the power network (20), and the third proportional coefficient is for regulating a second adjustment parameter for providing inertia to the power network (20).

Images