Water electrolytic tank system and method for controlling water electrolytic tank system

Abstract

The present invention appropriately connects a water electrolysis system to a power grid. Accordingly, this water electrolytic tank system is provided with a water electrolytic tank for producing hydrogen by DC power supplied from a power converter, and with an effective power adjustment unit (B) that: defines, as an input effective power, an effective power received from an AC grid by the power converter; defines, as a reference effective power, the input effective power when the frequency deviation of a grid frequency is "0"; defines, as an effective power correction amount (del_Pfil), the result obtained by subtracting the reference effective power from the input effective power; defines, as a first gain, the gain of the fluctuation amplitude of the effective power correction amount with respect to the fluctuation amplitude of the frequency deviation in the case where the fluctuation cycle of the frequency deviation is a prescribed first cycle; defines, as a second gain, the gain of the fluctuation amplitude of the effective power correction amount with respect to the fluctuation amplitude of the frequency deviation in the case where the fluctuation cycle of the frequency deviation is a prescribed second cycle which is shorter than the first cycle; and sets the effective power correction amount such that the second gain is less than the first gain.

Description

Water electrolysis cell system and method for controlling the water electrolysis cell system

[0001] The present invention relates to a water electrolysis cell system and a method for controlling a water electrolysis cell system.

[0002] Changes in energy usage toward carbon neutrality are now becoming a global trend, with transitions underway on both the power generation and energy consumption sides. Industrial decarbonization is being driven primarily by the electrification of heat sources. However, due to the amount of heat required and transportation demand, electrification does not necessarily solve all of the problems, and expectations are rising for decarbonized fuels. Because hydrogen does not emit carbon dioxide during the combustion process, there are high hopes for the development of production and utilization technologies. Meanwhile, progress in decarbonizing power sources is leading to the increasing disconnection of aging thermal power plants from grids, which is reducing the grid's ability to maintain frequency, including inertia, and ultimately reducing grid stability.

[0003] In addition to the inertial supply using conventional rotary generators and stability improvement services using governor-free functions, in recent years, the high control performance of semiconductor power converters has been utilized to provide frequency stabilization functions from storage batteries and renewable energy sources. Water electrolysis systems also have the potential to control the power received from the AC grid, as they are equipped with power converters between the AC grid and the water electrolyzer. In addition, as the capacity of water electrolysis systems continues to increase, their contribution to power grid stabilization is expected to grow even more in the future.

[0004] As an example, the abstract of Patent Document 1 listed below states: "Described is a method for operating an electrolyzer (10) having a converter (12) connected on its AC voltage side to an AC voltage grid (15) via a decoupling inductor (13) and drawing AC active power from the AC voltage grid (15), and an electrolyzer (11) connected on its DC voltage side to the converter (12) and operated at a power of 50% to 100% of the nominal power of the electrolyzer (11) at a grid frequency that corresponds to the nominal frequency of the AC voltage grid (15) and is constant over time, wherein the converter (12) is operated with an applied voltage such that the AC active power drawn from the AC voltage grid (15) is changed directly in response to changes and / or the rate of change of the grid frequency of the AC voltage grid (15). Also described is an electrolyzer and a method for providing instantaneous reserve power to the AC voltage grid."

[0005] Special Publication No. 2022-515821

[0006] However, in the above-mentioned technology, there is a demand for a more appropriate connection of a water electrolysis system to an AC system, taking into consideration the lifespan of the water electrolyzer. The present invention has been made in view of the above-mentioned circumstances, and aims to provide a water electrolysis cell system that can be appropriately connected to an AC system, and a method for controlling the water electrolysis cell system.

[0007] In order to achieve the above object, a water electrolytic cell system of the present invention comprises: a self-excited power converter that converts AC power received from an AC system into DC power; a water electrolytic cell that produces hydrogen using the DC power supplied from the power converter; a frequency detection unit that detects a system frequency that is the frequency of the AC system; an active power adjustment unit that defines input active power as the active power received from the AC system by the power converter; a frequency deviation as a result of subtracting a predetermined rated frequency from the system frequency; a reference active power as the input active power when the frequency deviation is "0"; a result of subtracting the reference active power from the input active power as an active power correction amount; a first gain as a gain of the fluctuation amplitude of the active power correction amount relative to the fluctuation amplitude of the frequency deviation when the fluctuation period of the frequency deviation is a predetermined first period; and a second gain as a gain of the fluctuation amplitude of the active power correction amount relative to the fluctuation amplitude of the frequency deviation when the fluctuation period of the frequency deviation is a predetermined second period that is shorter than the first period; and sets the active power correction amount so that the second gain is smaller than the first gain.

[0008] According to the present invention, the water electrolysis system can be appropriately connected to an AC system.

[0009] FIG. 1 is a block diagram of a water electrolytic cell system according to a first embodiment. FIG. 2 is a circuit diagram of an example of a self-excited converter. FIG. 3 is a block diagram of a waveform control device. FIG. 4 is a block diagram of an AC current control unit. FIG. 5 is a diagram showing an example of the characteristics of an adjustment power supply characteristic calculator. FIG. 6 is a diagram showing an example of calculation in the adjustment power calculator. FIG. 7 is a block diagram of a frequency deviation determiner. FIG. 8 is a block diagram of an active current limit value calculator. FIG. 9 is a block diagram of a water electrolytic cell system according to a second embodiment. FIG. 10 is a block diagram of a water electrolytic cell system according to a third embodiment. FIG. 11 is a block diagram of a waveform control device in the third embodiment. FIG. 12 is a block diagram of a reactive power command calculator in the third embodiment. FIG. 13 is a block diagram of a computer.

[0010] [Overview of the embodiment] Applying the above-mentioned technology of Patent Document 1, it is believed that pseudo-inertia can be supplied to a power grid. Similarly, the embodiment described below controls a power converter as a variable power source to provide instantaneous reserve power in response to changes in the grid phase and voltage at the interconnection point, thereby achieving a pseudo-stabilizing effect similar to that of a rotary generator. However, the supply and demand balance of a power system is constantly fluctuating. The grid frequency is a state quantity that indicates the state of the supply and demand balance, and this frequency includes medium-term fluctuations of several seconds or longer and short-term fluctuations shorter than one second. Therefore, when supplying instantaneous reserve power using the technology of Patent Document 1, both short-term and medium-term fluctuations are mixed in with the power supplied to the water electrolyzer.

[0011] However, frequent changes in the power or current supplied to a water electrolyzer increase mechanical stress on the insulating film and induce structural changes in the electrodes and catalysts, resulting in accelerated deterioration. Therefore, there is a need for control characteristics of a water electrolysis system that can contribute to the stability of the power grid while minimizing the impact on the lifespan of the water electrolyzer, and a structure that realizes such control characteristics. The following embodiment solves the above problem by providing an adjustment power calculation function with a correction function that corrects the received power with a low gain for frequency pulsations in the frequency components of the grid connected to the water electrolyzer system that are shorter than a predetermined fluctuation period. This embodiment suppresses the frequency of changes in the current supplied to the water electrolyzer in response to grid frequency fluctuations, while enabling the water electrolysis system to supply adjustment power to the connected AC grid for frequency fluctuations longer than a predetermined fluctuation period. This makes it possible to avoid an extreme shortening of the lifespan of the water electrolyzer and stabilize the grid at the same time.

[0012] First Embodiment Overall Configuration of Water Electrolyzer System 8 Figure 1 is a block diagram of a water electrolyzer system 8 according to a first embodiment. The water electrolyzer system 8 is connected to an AC system 1 and receives power for water electrolysis from the AC system 1. The water electrolyzer system 8 includes a transformer 2, a power converter 3, a cable 4 (connection circuit), a water electrolyzer 5, and a host controller 200. The transformer 2 is connected between the AC system 1 and the power converter 3. The power converter 3 converts AC power received from the AC system 1 into DC power. The cable 4 transmits DC current supplied from the power converter 3 to the water electrolyzer 5. The water electrolyzer 5 produces hydrogen by hydrolyzing water using the supplied DC current. The host controller 200 performs various controls on the power converter 3.

[0013] The power converter 3 also includes, as main circuit elements, a capacitor 31 for the AC filter, a reactor 32 for the AC filter, a self-excited converter 33, a capacitor 34 for the DC filter (capacitor circuit), an AC voltage sensor 10, current sensors 11u and 11w, a DC voltage sensor 12, a current sensor 13, and a waveform control device 100.

[0014] The AC voltage sensor 10 is connected to the secondary side of the transformer 2 to detect the secondary voltage and output the result as detected AC voltage values ​​Vu, Vv, and Vw. The current sensors 11u and 11w detect the output currents of the U and W phases of the transformer 2 and output the result as detected AC current values ​​Iu and Iw. The DC voltage sensor 12 detects the voltage across the capacitor 34 and outputs the result as detected DC voltage value Vdc. The current sensor 13 detects the DC current flowing from the power converter 3 to the water electrolyzer 5 and outputs the result as detected DC current value Idc. The waveform control device 100 generates a group of gate signals GS, which are drive signals for the self-excited converter 33, based on the detected AC voltage values ​​Vu, Vv, and Vw and the detected AC current values ​​Iu and Iw. As a result, the waveform control device 100 controls the active power and reactive power exchanged between the AC system 1 and the water electrolytic cell system 8 , and supplies the desired DC power to the water electrolytic cell 5 via the cable 4 .

[0015] 2 is a circuit diagram of an example of the self-excited converter 33. The self-excited converter 33 is a two-level converter and includes AC terminals TU, TV, and TW, DC terminals TP and TN, and six IGBT modules 33up, 33un, 33vp, 33vn, 33wp, and 33wn bridge-connected between them.

[0016] Each IGBT module includes an IGBT (Insulated Gate Bipolar Transistor) and a diode connected in anti-parallel to the IGBT. The gate signal group GS includes six gate signals that control the on / off of each of the IGBTs.

[0017] The AC terminals TU, TV, and TW are electrically connected to reactors 32 (see FIG. 1 ), and the DC terminals TP and TN are electrically connected to a DC filter capacitor 34 and the cable 4. In this embodiment, a two-level converter is used as the self-excited converter 33, but the self-excited converter 33 may be a three-level converter, a five-level converter with more levels, or an MMC (Modular Multilevel Converter).

[0018] 1 , the DC filter capacitor 34 is connected to the DC side of the self-excited converter 33. The capacitor 34 is provided to realize the function of absorbing ripple current generated by the switching operation of the IGBTs in the self-excited converter 33 and the function of stabilizing the DC voltage supplied to the water electrolyzer 5. To realize this function, the resonant frequency formed by the inductance L cable [H] of the cable 4 and the capacitance Cdc [F] of the DC filter capacitor 34 is selected to be lower than the switching frequency f sw [Hz] of the self-excited converter 33.

[0019] 1, the DC filter capacitor 34 is shown as a single capacitor, but instead, a capacitor unit in which a plurality of capacitors are connected in series and parallel may be applied. In this case, the capacitance Cdc may be the equivalent capacitance of the capacitor unit.

[0020] The DC voltage sensor 12 detects the voltage across the capacitor 34 and outputs the result as a detected terminal voltage value Vdc to the waveform control device 100. As will be described in detail later, this detected terminal voltage value Vdc is used in the waveform control device 100 to control the DC current supplied to the water electrolytic cell 5.

[0021] The DC current supplied to the water electrolytic cell 5 is detected by the current sensor 13, and the waveform control device 100 performs appropriate calculations to control the power converter 3 so that the desired DC current flows into the water electrolytic cell 5. The upper level control device 200 calculates a DC current command value Idc_ref, a reactive power command value Q_ref, and a time constant τ_set in accordance with a separately determined hydrogen production plan, and outputs them to the waveform control device 100.

[0022] Here, the DC current command value Idc_ref commands the DC current to be supplied to the water electrolyzer 5. Furthermore, the reactive power command value Q_ref commands the reactive power to be output by the power converter 3 to the AC system 1. Furthermore, the time constant τ_set will be described in detail later.

[0023] Furthermore, the host control device 200 receives from the waveform control device 100 the active power P_out (input active power) input to the power converter 3, the system frequency grid_f of the AC system 1, and the active power correction amount del_Pfil (details will be described later) for providing adjustment capability. The host control device 200 includes an input unit 210 (interface), a display unit 212 (interface), a recording unit 214, and a communication unit 216. An operator of this system can input a DC current command value Idc_ref, a reactive power command value Q_ref, and a time constant τ_set via the input unit 210.

[0024] The display unit 212 displays the operating waveforms of the active power P_out, the grid frequency grid_f, and the active power correction amount del_Pfil (details will be described later), as well as other operational statuses of the water electrolytic cell system 8. The recording unit 214 stores history information P_history, which is a history of the received power adjusted according to frequency and other operational history of the water electrolytic cell system 8. The communication unit 216 transmits the history information P_history to a grid operation system (not shown).

[0025] <Waveform Controller 100> Figure 3 is a block diagram of the waveform controller 100 according to the first embodiment. As described above, the waveform controller 100 receives the DC current command value Idc_ref, the DC current detection value Idc, the reactive power command value Q_ref, and the time constant τ_set from the host controller 200. The waveform controller 100 also receives AC voltage detection values ​​Vu, Vv, and Vw, AC current detection values ​​Iu and Iw, the DC current detection value Idc, and the terminal voltage detection value Vdc from various sensors. Based on these, the waveform controller 100 calculates a group of gate signals GS and outputs them to the self-excited converter 33.

[0026] The waveform control device 100 includes a system phase / frequency detection unit A (frequency detection unit), an adjustment capability calculation / control unit B (active power adjustment unit, active power adjustment process), a system voltage / feedforward calculation unit C, a reactive power control unit D, a DC current / voltage control unit E, an AC current conversion / control unit F, and a gate signal generation unit G. Here, the system phase / frequency detection unit A includes an α-β converter 1001 and a phase detector 1002.

[0027] The regulation capability calculation / control unit B includes a regulation capability calculator 1003, a frequency deviation determiner 1004, an active current limit value calculator 1005, an amplitude calculator 1031, and an output current command value corrector 1050. The system pressure / feedforward calculation unit C includes a d-q converter 1030 and low-pass filters 1032 and 1033. The reactive power control unit D includes a subtractor 1040, an α-β converter 1041, an active / reactive power calculator 1042, a subtractor 1044, and a reactive power controller 1045.

[0028] The DC current / voltage control unit E includes a subtractor 1051, a DC current controller 1052, a subtractor 1053, and a DC voltage controller 1054. The AC current conversion / control unit F includes a d-q converter 1043 and an AC current control unit 1060. The gate signal generation unit G includes a carrier wave generator 1070 and a PWM calculator 1071.

[0029] (System Phase / Frequency Detection Unit A) First, we will explain the system phase / frequency detection unit A. The AC voltage detection values ​​Vu, Vv, and Vw detected by the AC voltage sensor 10 (see FIG. 1) are converted into two-phase quantities Vα and Vβ in an α-β converter 1001 and supplied to a phase detector 1002. The α-β converter 1001 is a matrix calculator, and its calculation content is expressed by the following equation (1).

[0030]

[0031] Here, assuming a rotating coordinate system that rotates at an internal oscillation angular frequency ω, and orthogonal axes in this rotating coordinate system are the d-axis and q-axis, the phase detector 1002 converts the input two-phase quantities Vα and Vβ into a d-axis component Vsd and a q-axis component Vsq, which are values ​​on the rotating coordinate system, using internally oscillating sine waves (cosωt, sinωt), as shown in the following equation (2):

[0032]

[0033] Hereinafter, this conversion process will be referred to as d-q conversion. The phase detector 1002 includes a PLL (Phase Locked Loop) circuit that uses a PI controller (Proportional-Integral Controller) to correct the internal oscillation angular frequency ω so as to set the q-axis component Vsq to "0." This technology is widely used in this technical field, so a detailed description using a block diagram will be omitted. However, using the above-mentioned calculation function, the phase detector 1002 calculates and outputs the grid frequency grid_f and grid phase grid_ph, which are the frequency and phase of the AC voltage detection values ​​Vu, Vv, and Vw.

[0034] The calculated system frequency grid_f is a frequency [Hz] obtained by dividing the angular frequency ω by 2π, and is output to a regulation capability calculation / control unit B (described later) and a higher-level control device 200. In addition, the system phase grid_ph is used in coordinate conversion in each calculation unit in the waveform control device 100 (described later).

[0035] (System Pressure / Feedforward Calculation Unit C) Next, the system pressure / feedforward calculation unit C will be described. The two-phase quantities Vα and Vβ calculated by the α-β converter 1001 are coordinate-converted in the d-q converter 1030 using cosine and sin components based on the system phase grid_ph. The calculation in the d-q converter 1030 is the same as that shown in equation (2), and grid_ph is used as phase information. As a result, the d-q converter 1030 outputs a d-axis system voltage Vgd and a q-axis system voltage Vgq.

[0036] The low-pass filters 1032 and 1033 perform low-pass filtering on the d-axis system voltage Vgd and the q-axis system voltage Vgq, respectively, and output the results to the AC current control unit 1060 as signals Vgd_fil and Vgq_fil.

[0037] Although details will be described later, the AC current control unit 1060 uses the signals Vgd_fil and Vgq_fil as voltage components for balancing with the AC system voltage and applies feedforward to the voltage command values ​​Vu_ref, Vv_ref, and Vw_ref. This makes it possible to prevent AC current with large amplitude from flowing into the power converter 3 when the gate of the self-excited converter 33 is unblocked or when the system voltage suddenly changes.

[0038] (DC Current / Voltage Control Unit E) Next, the DC current / voltage control unit E will be described. An output current command value corrector 1050 (details of which will be described later) outputs an output current command value Idc_ref2. The DC current / voltage control unit E causes the DC current detection value Idc to follow the output current command value Idc_ref2 within an allowable range in accordance with the output current command value Idc_ref2. More specifically, it calculates an active current command value Id_ref to be received from the AC system so as to control the DC voltage.

[0039] The subtractor 1051 subtracts the DC current detection value Idc detected by the current sensor 13 from the output current command value Idc_ref2 and outputs the result to the DC current controller 1052. The DC current controller 1052 is a PI controller, and receives the output of the subtractor 1051 and a flag LimiterFLG as input. As a result, the DC current controller 1052 calculates a DC voltage command value Vdc_ref such that the DC current detection value Idc follows the output current command value Idc_ref2. The DC voltage command value Vdc_ref is a command value for the DC voltage that the power converter 3 outputs to the water electrolyzer 5.

[0040] The flag LimiterFLG is a status flag that is set to "1" when the active current command value Id_ref (details of which will be described later) calculated by the DC voltage controller 1054 is limited by a limiter. When the flag LimiterFLG is set to "1", the DC current controller 1052 stops the calculation of an internal integrator (not shown). This makes it possible to avoid the accumulation of errors in the DC current during current limitation, and to prevent abrupt current changes when the grid returns to normal.

[0041] The subtractor 1053 subtracts the DC voltage command value Vdc_ref from the terminal voltage detection value Vdc of the capacitor 34 (see FIG. 1 ) detected by the DC voltage sensor 12, and outputs the subtraction result. In addition to the output of the subtractor 1053, the DC voltage controller 1054 receives an active current limit value Id_lim output from a regulation power calculation / control unit B (details of which will be described later).

[0042] The DC voltage controller 1054 is a PI controller that calculates an active current command value Id_ref by performing PI calculation on the output of the subtractor 1053. The active current limit value Id_lim is a limiter value in the power receiving direction of the PI controller, and creates regulation power for the AC grid 1 by actively limiting the active current command value Id_ref using this limiter value.

[0043] The calculated active current command value Id_ref is output to the AC current control unit 1060. In the DC voltage controller 1054, when the result of the PI control calculation for the voltage deviation is limited by the active current limit value Id_lim, the DC voltage controller 1054 sets the flag LimiterFLG to "1". In other cases, the DC voltage controller 1054 sets the flag LimiterFLG to "0". Then, the flag LimiterFLG is output to the DC current controller 1052 as described above.

[0044] (Reactive Power Control Unit D) Next, the reactive power control unit D will be described. The AC current detection values ​​Iu and Iw detected by the current sensors 11u and 11w are input to a subtractor 1040. Because the zero-phase component of the current flowing through the transformer 2 is zero, the subtractor 1040 outputs a v-phase AC current estimation value Iv.

[0045] When the AC current detection values ​​Iu, Iw and the AC current estimation value Iv are input, the α-β converter 1041 performs α-β conversion of the three-phase quantities using the matrix in the above-mentioned equation (1), and supplies the resulting two-phase quantities Iα, Iβ to the active / reactive power calculator 1042 and the d-q converter 1043.

[0046] The active / reactive power calculator 1042 outputs active power P_out and consumed reactive power Q_out based on the two-phase quantities Vα, Vβ, Iα, and Iβ. The consumed reactive power Q_out is reactive power output from the power converter 3 to the AC system 1. In this embodiment, the polarities of the active power and reactive power are defined as positive for the active power output from the power converter 3 to the AC system 1 and positive for the reactive power when the power converter 3 operates as a reactor, and the polarities of the active current and reactive current are also defined accordingly. The active power P_out is output to the upper control device 200.

[0047] The subtractor 1044 receives the reactive power command value Q_ref input from the upper level control device 200 and the consumed reactive power Q_out from the active / reactive power calculator 1042, calculates the deviation therebetween, and outputs the deviation to the reactive power controller 1045. The reactive power controller 1045 is equipped with a PI controller, performs PI calculation on the input deviation, and outputs the resulting reactive current command value Iq_ref to the AC current control unit 1060.

[0048] (AC Current Conversion / Control Unit F) Next, the AC current conversion / control unit F will be described. The two-phase quantities Iα and Iβ, which are the α and β components of the AC current, and the system phase grid_ph detected by the phase detector 1002 are input to the dq converter 1043. The dq converter 1043 performs a rotational coordinate transformation similar to that of the above-mentioned equation (2) on the two-phase quantities Iα and Iβ using the system phase grid_ph as phase information, and outputs the resulting active current Id and reactive current Iq.

[0049] The AC current control unit 1060 receives the active current command value Id_ref, the reactive current command value Iq_ref, the active current Id, the reactive current Iq, the signals Vgd_fil, Vgq_fil, and the system phase grid_ph. The AC current control unit 1060 then outputs voltage command values ​​Vu_ref, Vv_ref, and Vw_ref that command voltages to be output by the self-excited converter 33 to the AC system 1 so that the active current Id follows the active current command value Id_ref and the reactive current Iq follows the reactive current command value Iq_ref.

[0050] 4 is a block diagram of the AC current control unit 1060. A subtractor 10061 subtracts the active current Id from the active current command value Id_ref and outputs the deviation between the two. A PI controller 10063 performs PI control (proportional-integral control) calculation on this deviation and outputs the result to an adder 10065. The adder 10065 adds the calculation result of the PI control, which is a voltage correction term, to a signal Vgd_fil, which is a feedforward term, and outputs the result as a d-axis voltage command value Vd.

[0051] Furthermore, a subtractor 10062 subtracts the active current Id from the reactive current command value Iq_ref and outputs the deviation between them. A PI controller 10064 performs PI control calculation on this deviation and outputs the result to an adder 10066. The adder 10066 adds the calculation result of the PI control, which is a voltage correction term, to a signal Vgq_fil, which is a feedforward term, and outputs the result as a q-axis voltage command value Vq.

[0052] The inverse d-q converter 10067 receives the d-axis voltage command value Vd, the q-axis voltage command value Vq, and the grid phase grid_ph. Then, as shown in the following equation (3), the inverse d-q converter 10067 performs inverse d-q transformation on the d-axis voltage command value Vd and the q-axis voltage command value Vq using a cosine component and a sin component whose phase is the grid phase grid_ph, and outputs the resulting two-phase amounts Valp and Vbet to the two-phase-to-three-phase converter 10068.

[0053]

[0054] The two-phase to three-phase converter 10068 converts the two-phase amounts Valp and Vbet into three-phase voltage command values ​​Vu_ref, Vv_ref, and Vw_ref based on the following equation (4).

[0055]

[0056] (Gate Signal Generator G) Returning to Fig. 3 , the gate signal generator G will now be described. A carrier wave generator 1070 included in the gate signal generator G generates a triangular wave that is a carrier wave for PWM-modulating the voltage command values ​​Vu_ref, Vv_ref, and Vw_ref. A PWM calculator 1071 generates each gate signal included in the gate signal group GS by comparing the triangular wave with a value obtained by normalizing each of the voltage command values ​​Vu_ref, Vv_ref, and Vw_ref by half the DC voltage (Vdc / 2).

[0057] The triangular wave is a repeating waveform with a frequency of fsw [Hz] and an amplitude of "1 [-]". By generating a gate signal group GS, which is a gate signal for the self-excited converter 33, by comparing the magnitude of this triangular wave with the normalized voltage command value, the self-excited converter 33 can output AC output voltages corresponding to the voltage command values ​​Vu_ref, Vv_ref, and Vw_ref.

[0058] (Adjustment Capability Calculation / Control Unit B) Next, the adjustment capability calculation / control unit B, which is a feature of this embodiment, will be described. The adjustment capability calculation / control unit B has an adjustment capability supply characteristic calculation unit 10031 and a low-pass filter 10032, and calculates an active power correction amount del_Pfil (details will be described later) to be adjusted in accordance with the system frequency grid_f of the AC system 1. A frequency deviation determiner 1004 detects a state in which adjustment capability should be provided, and outputs a deviation determination flag F_FLG that is set to "1" when the system frequency grid_f deviates from the normal range.

[0059] The amplitude calculator 1031 calculates the square root of the sum of the squares of the d-axis system voltage Vgd and the q-axis system voltage Vgq and outputs the result as the AC system voltage amplitude VGRID to the active current limit value calculator 1005. The active current limit value calculator 1005 adjusts the limiter of the receiving direction of the active current of the power converter 3 according to the active power correction amount del_Pfil. The output current command value corrector 1050 corrects the active current command value Id_ref input from the upper control device 200 according to the deviation determination flag F_FLG and outputs the correction result to the subtractor 1051 as the above-mentioned output current command value Idc_ref2. In this way, according to the regulation capability calculation / control unit B, when the system frequency grid_f of the AC system 1 deviates from the normal range, it is possible to change the active power exchanged with the AC system 1 by dynamically adjusting the active current limit value Id_lim. Each of the above-mentioned computing units will be described in detail below.

[0060] 5 is a diagram showing an example of the characteristics of the adjustment power supply characteristic calculator 10031. As shown in the figure, the adjustment power supply characteristic calculator 10031 outputs the original correction amount del_P of the active power in response to the system frequency grid_f of the AC system 1 shown on the horizontal axis. As shown in the figure, the characteristic of the adjustment power supply characteristic calculator 10031 is a drooping characteristic having a dead band.

[0061] The adjustment power supply characteristic calculator 10031 sets a predetermined dead band for the rated frequency f_rated of the AC system 1. That is, the dead band ranges from a predetermined frequency f_2 lower than the rated frequency f_rated to a predetermined frequency f_3 higher than the rated frequency f_rated.

[0062] Then, the adjustment power supply characteristic calculator 10031 sets the correction amount to a positive value when the detected system frequency grid_f is lower than the frequency f_2. That is, the adjustment power supply characteristic calculator 10031 outputs the original correction amount del_P such that active power correction is performed in the direction of supplying active power to the AC system 1 (in the direction of reducing the power received from the AC system 1).

[0063] Hereinafter, the result of subtracting the rated frequency f_rated from the grid frequency grid_f may be referred to as the "frequency deviation Δf" (not shown). Furthermore, the active power P_out when the frequency deviation Δf is "0" may be referred to as the "reference active power P_s" (not shown). The above-mentioned active power correction amount del_Pfil is the result of subtracting the reference active power P_s from the active power P_out. The original correction amount del_P shown in FIG. 5 is a value that is the basis for calculating the active power correction amount del_Pfil.

[0064] On the other hand, the adjustment power supply characteristic calculator 10031 sets the correction amount to a negative value when the detected system frequency grid_f is higher than the frequency f_3. That is, it outputs an original correction amount del_P that performs active power correction in the direction of receiving active power from the AC system 1 (increasing the power received from the AC system 1). As shown in Fig. 5 , these correction amounts have drooping characteristics that are determined by an amount proportional to the deviation of the system frequency grid_f from f_2 or f_3.

[0065] Furthermore, the adjustment power supply characteristic calculator 10031 sets upper and lower limits of ±ΔP_max for the original correction amount del_P. That is, when the original correction amount del_P reaches ΔP_max at f_1, which is a frequency lower than f_2, the correction amount is maintained at ΔP_max even if the frequency drops further. Conversely, when the original correction amount del_P reaches −ΔP_max at f_4, which is a frequency higher than f_3, the correction amount is maintained at −ΔP_max even if the frequency rises further. Here, ΔP_max is set to prevent deterioration of the water electrolyzer 5 due to excessive output adjustment, and is preferably set to less than 50% (0.5 pu) of the rated power.

[0066] Returning to Figure 3, the explanation will continue. The output of the adjustment power supply characteristic calculator 10031 is output to a low-pass filter 10032. In addition to the output of the adjustment power supply characteristic calculator 10031, a time constant τ_set set by the upper control device 200 is also input to the low-pass filter 10032. The time constant τ_set is a time constant that should be selected taking into consideration the initial characteristics, degree of deterioration, temperature, etc. of the water electrolyzer 5. The time constant of the current power grid itself is about 3 seconds, and considering that this will be strengthened by the water electrolysis system, it is appropriate to select τ_set at about 0.1 to 1 second as an initial condition.

[0067] Fig. 6 is a diagram showing an example of calculation in the adjustment capability calculator 1003. That is, Fig. 6 shows example waveforms of the system frequency grid_f, the original correction amount del_P output by the adjustment capability supply characteristic calculator 10031, and the active power correction amount del_Pfil output by the low-pass filter 10032. Here, the frequency f_2 is set to 49.98 Hz, the frequency f_3 is set to 51.02 Hz, ΔP_max is set to 0.1 pu, and the time constant τ_set is set to 0.1 sec. Also, it is assumed that a system fault occurs at time 1.5 seconds, causing the system frequency grid_f to swing.

[0068] As shown in the waveform before 1.5 seconds in Fig. 6 , even if high-frequency pulsation is mixed in the grid frequency grid_f, as long as it is within the normal range, the active power correction amount del_Pfil is calculated as zero. However, once the grid frequency grid_f deviates from the normal range, the active power correction amount del_Pfil is calculated in a direction that returns the grid frequency grid_f to the rated value, and it is confirmed that fluctuations in the active power correction amount del_Pfil are suppressed for high-frequency components included in the grid frequency grid_f.

[0069] Then, as described above, the system frequency grid_f, the active power correction amount del_Pfil, and the active power P_out are transmitted to the host controller 200. As described above, the host controller 200 is provided with the display unit 212 that displays the received data, and therefore the operator of the water electrolytic cell 5 can confirm that high-frequency components have been removed from the active power correction amount del_Pfil.

[0070] The active power correction amount del_Pfil is also output to the active current limit value calculator 1005. In this embodiment, a low-pass filter 10032 is provided as a filter for removing high-frequency components contained in the original correction amount del_P output by the adjustment power supply characteristic calculator 10031. However, a moving average filter can be applied instead of the low-pass filter 10032. In this case, the filter characteristics set by the upper control device 200 are not the time constant τ_set but the width of the moving average window of the moving average.

[0071] Furthermore, the effect of applying the low-pass filter 10032 is to "suppress high-frequency components," which can be stated in other words as "calculating the adjustment force with a lower gain for the variable components contained in the original correction amount del_P that have a period shorter than a predetermined period."

[0072] Furthermore, when the time constant τ_set is in the range of 0.1 to 1 second, the gain of the fluctuation amplitude of the active power correction amount del_Pfil relative to the fluctuation amplitude of the frequency deviation Δf when the fluctuation period of the frequency deviation Δf is a predetermined first period TV1 (e.g., 1.1 seconds) is defined as a first gain GA1 (not shown). Furthermore, the gain of the fluctuation amplitude of the active power correction amount del_Pfil relative to the fluctuation amplitude of the frequency deviation Δf when the fluctuation period of the frequency deviation Δf is a predetermined second period TV2 (e.g., 0.09 seconds) that is shorter than the first period TV1 is defined as a second gain GA2 (not shown). It can also be said that the low-pass filter 10032 sets the active power correction amount del_Pfil so that the second gain GA2 is smaller than the first gain GA1.

[0073] 7 is a block diagram of the frequency deviation determiner 1004. The frequency deviation determiner 1004 includes a low-pass filter 10041, a normal frequency deviation determiner 10042, and an off-delay calculator 10043. The low-pass filter 10041 performs low-pass filtering of the grid frequency grid_f. The time constant of the low-pass filter 10041 is set to a value that prevents erroneous detection of the grid frequency grid_f due to noise or the operation of the water electrolytic cell system 8 itself. It is desirable to select this time constant to be approximately 0.1 to 1 second.

[0074] The frequency normal value deviation determiner 10042 outputs a binary signal that is "1" if the output of the low-pass filter 10041 deviates from the frequency dead band (f_2 to f_3) described in FIG. 5 , and is "0" otherwise. The off-delay calculator 10043 outputs a deviation determination flag F_FLG based on the binary signal. First, when the binary signal changes from "0" to "1", the off-delay calculator 10043 immediately sets the deviation determination flag F_FLG to "1". On the other hand, when the binary signal changes from "1" to "0", the off-delay calculator 10043 holds the deviation determination flag F_FLG at "1" for a predetermined holding time, and sets the deviation determination flag F_FLG to "0" if the binary signal remains "0" during that holding time.

[0075] As a result, when the system frequency grid_f swings due to a system fault or the like, it becomes possible to maintain the deviation determination flag F_FLG at "1", and a stable regulation power can be supplied to the AC system 1. Taking into consideration the time required for the system fault to be resolved, it is desirable to select this holding time to be approximately 5 to 10 seconds.

[0076] 8 is a block diagram of the active current limit value calculator 1005. The active current limit value calculator 1005 receives the active power correction amount del_Pfil, the deviation determination flag F_FLG, the AC system voltage amplitude VGRID, and the active current command value Id_ref from the adjustment capability calculator 1003, the frequency deviation determiner 1004, the amplitude calculator 1031, and the DC voltage controller 1054. Based on these values, when the deviation determination flag F_FLG becomes "1" (when the system frequency grid_f deviates from the normal range), the active current limit value calculator 1005 sets the active current limit value Id_lim so as to correct the adjustment capability using the operating point up to that point as a starting point.

[0077] The active current limit value calculator 1005 includes low-pass filters 10051 and 10052 , a divider 10053 , a multiplier 10054 , switches 10055 and 10059 , a previous value holder 10056 , an adder 10057 , and a limiter 10058 .

[0078] Low-pass filters 10051 and 10052 perform low-pass filtering on the active current command value Id_ref and the AC grid voltage amplitude VGRID, respectively. A divider 10053 divides the active power correction amount del_Pfil by the low-pass filtered AC grid voltage amplitude VGRID. A multiplier 10054 multiplies the output of the divider 10053 by a constant (√(3)) to convert it into a value corresponding to the active current.

[0079] The active current command value Id_ref after low-pass filtering output from the low-pass filter 10051 is input to one input terminal of a switch 10055. The output signal of a previous value holder 10056 is input to the other input terminal of the switch 10055. The previous value holder 10056 outputs the previous output value of the switch 10055, i.e., the output value one control cycle before.

[0080] When the deviation determination flag F_FLG is "0" (the frequency is in the normal range), the switch 10055 selects and outputs the active current command value Id_ref after low-pass filtering. On the other hand, when the deviation determination flag F_FLG is "1" (a frequency deviation has occurred), the switch 10055 selects and outputs the output value of the previous value holder 10056. As a result, when a deviation of the grid frequency grid_f occurs and the deviation determination flag F_FLG becomes "1," the switch 10055 continues to output the active current command value Id_ref after low-pass filtering that was before the deviation occurred.

[0081] The adder 10057 adds the outputs of the multiplier 10054 and the switch 10055 and outputs the sum to the limiter 10058. The limiter 10058 limits the input signal to a value between 0 and Id_rated, which is a negative value obtained by multiplying the rated current value of the power converter 3 by "-1", and outputs the limited result to the switch 10059.

[0082] Id_rated is supplied to one input terminal of a switch 10059, and the output signal of limiter 10058 is supplied to the other input terminal. When the deviation determination flag F_FLG is "0" (the frequency is within the normal range), the switch 10059 selects Id_rated and outputs it as the active current limit value Id_lim. On the other hand, when the deviation determination flag F_FLG is "1" (a frequency deviation has occurred), the switch 10059 selects the output value of limiter 10058 and outputs it as the active current limit value Id_lim.

[0083] With the above configuration, starting from an operating state in which the grid frequency grid_f is within the normal range, limiter adjustment of the active current corresponding to the regulation capability can be realized when the grid frequency grid_f deviates from the normal range.

[0084] 3 , the active current limit value Id_lim is output as a negative limiter to the DC voltage controller 1054. When the deviation determination flag F_FLG is "0", the output current command value corrector 1050 outputs the same value as the DC current command value Idc_ref as the output current command value Idc_ref2. On the other hand, when the deviation determination flag F_FLG is "1", the output current command value corrector 1050 sets the rated current of the water electrolytic bath 5 as the output current command value Idc_ref2.

[0085] As a result, when the system frequency grid_f becomes higher than the frequency f_3 under partial load operation as an initial condition, the DC current controller 1052 is set to increase the DC voltage command value Vdc_ref. As a result, the DC voltage controller 1054 corrects the active current command value Id_ref in a direction that sticks to the negative limiter. Therefore, it is possible to increase the power received from the AC system 1, and to correct the active power in a direction that contributes to stabilizing the frequency of the AC system 1.

[0086] As described above, according to this embodiment, when the frequency of the AC grid 1 connected to the water electrolytic cell system 8 deviates from the normal range, active power adjustment can be performed to stabilize the frequency. Furthermore, even when the system frequency grid_f includes high-frequency pulsating components, the active power correction performed by the water electrolytic cell system 8 can suppress the influence of the high-frequency pulsating components, thereby suppressing deterioration of the water electrolytic cell 5.

[0087] Second Embodiment Figure 9 is a block diagram of a water electrolytic cell system 8 according to a second embodiment. In the following description, parts corresponding to those in the first embodiment described above are denoted by the same reference numerals, and their description may be omitted. The configuration of the water electrolytic cell system 8 of this embodiment is similar to that of the first embodiment (see Figure 1), except that the AC voltage sensor 10 is connected to the primary side of the transformer 2. In particular, when the percentage impedance of the transformer 2 is large, setting it to the primary side as shown in Figure 9 can reduce the effect of the operation of the power converter 3 on the detection of the grid frequency grid_f. This enables the supply of a more stable regulation power.

[0088] [Third Embodiment] Figure 10 is a block diagram of a water electrolysis cell system 8 according to a third embodiment. In the following description, parts corresponding to those in the other embodiments described above are given the same reference numerals, and their description may be omitted. In this embodiment, a waveform control device 100a is provided instead of the waveform control device 100 (see Figure 1) in the first embodiment. Also, an AC system 1a is applied instead of the AC system 1 (see Figure 1).

[0089] The AC system 1a is assumed to be an AC system in which the resistance value R of the system impedance cannot be ignored compared to the reactance X (for example, R / X≈1 / 2 to 1 / 4), such as a power distribution system. In Fig. 10, the AC system 1a is represented as a series circuit of a resistor and an inductance.

[0090] The waveform control device 100a is characterized by having a learning function for stabilizing the voltage of the AC system 1a, and suppressing voltage fluctuations in the AC system during supply of regulating power through reactive power control using this learning function. That is, the waveform control device 100a calculates a reactive power command value itself based on state quantities detected by various sensors, rather than the reactive power command value Q_ref from the upper control device 200, and controls the reactive power to follow that value.

[0091] By implementing an adjustment capability supply function based on the system frequency in the water electrolytic cell system 8, the active power received from the AC system 1a can be actively changed. In cases such as the AC system 1a, where the resistance value R in the system impedance is not negligible compared to the reactance X, changes in the active power supplied to the water electrolytic cell system 8 can change the voltage of the AC system 1a, potentially deteriorating the quality of power supplied to other devices connected to this AC system 1a. Therefore, in this embodiment, the reactive power is controlled in accordance with the active power supplied to the water electrolytic cell system 8, thereby stabilizing the voltage of the AC system 1a.

[0092] 11 is a block diagram of a waveform control device 100a according to the third embodiment. The waveform control device 100a includes a reactive power command calculator 1100 (power factor adjustment unit), but is otherwise configured similarly to the waveform control device 100 of the first embodiment (see FIG. 3). Here, the reactive power command calculator 1100 receives the AC system voltage amplitude VGRID, the active power P_out, and the reactive current command Iq_out, and based on these, sets the power factor of the power input to the power converter 3 and calculates a reactive power command value Q_ref.

[0093] 12 is a block diagram of the reactive power command calculator 1100. The reactive power command calculator 1100 includes a multiplier 11001, an adder 11002, band-pass filters 11003 and 11004, a multiplier 11005 (correlation calculator), an integrator 11006 (reactive power command unit), and a multiplier 11007 (reactive power command unit).

[0094] The multiplier 11001 multiplies the active current command value Id_ref by a constant and outputs the result. As a result, the multiplier 11001 outputs a voltage value corresponding to the voltage drop caused by the leakage inductance in the transformer 2. The adder 11002 adds the AC system voltage amplitude VGRID and the voltage value output by the multiplier 11001. As a result, the adder 11002 outputs an amplitude estimation value of the primary side voltage of the transformer 2. The bandpass filter 11003 performs bandpass filtering on this amplitude estimation value and outputs the fluctuation component of the amplitude estimation value.

[0095] Meanwhile, the active power P_out is input to the bandpass filter 11004. The bandpass filter 11004 performs bandpass filtering on the active power P_out to output the fluctuation components of the active power P_out. The bandpass filters 11003 and 11004 have the same cut-in and cutout frequency characteristics. As a result, the bandpass filters 11003 and 11004 perform bandpass filtering with the same frequency characteristics on the primary side voltage amplitude estimate value output from the adder 11002 and the active power P_out.

[0096] The multiplier 11005 multiplies the output values ​​of the bandpass filters 11003 and 11004 to output a correlation value between the estimated amplitude value of the primary side voltage and the active power P_out. The integrator 11006 calculates an average correlation value by integrating the correlation value output by the multiplier 11005. This average correlation value is a power factor in the direction of suppressing voltage fluctuations, and the integrator 11006 can learn this power factor. The multiplier 11007 multiplies the active power P_out by this power factor to output a reactive power command value Q_ref that suppresses fluctuations in voltage amplitude. By controlling the reactive power of the power converter 3 in accordance with the reactive power command calculated as described above, it is possible to suppress amplitude fluctuations in the system voltage that fluctuate due to fluctuations in the active power P_out. In this embodiment, the output of the multiplier 11007 is used as the reactive power command value, but in order to specify the reactive power while the frequency is within the normal range, a fixed value may be added to the output of the multiplier 11007 and the sum may be used as the reactive power command value Q_ref. Also, the same effect can be achieved by using this fixed value as the reactive power command value transmitted from the upper control device 200.

[0097] As described above, according to this embodiment, similar to the first and second embodiments, when the system frequency grid_f of the AC system 1 connected to the water electrolytic cell system 8 deviates from the normal range, active power adjustment is possible to stabilize the frequency. Furthermore, similar to the first and second embodiments, even when the system frequency grid_f has a high-frequency pulsating component, the active power correction performed by the water electrolytic cell system 8 can suppress the influence of the high-frequency pulsating component, thereby suppressing deterioration of the water electrolytic cell 5. Furthermore, according to the water electrolytic cell system 8 of this embodiment, a reactive power command value Q_ref that suppresses voltage amplitude fluctuations in the connected AC system 1 is calculated as a result of learning, and reactive power control that follows this reactive power command can be realized. This makes it possible to suppress deterioration in the power quality of the connected AC system 1 while providing the AC system 1 with adjustment capability using the system frequency grid_f of the AC system 1 as an input.

[0098] [Computer Configuration] Figure 13 is a block diagram of a computer 980. The waveform control devices 100, 100a and the host control device 200 shown in Figures 1, 9, and 10 each include one or more computers 980 shown in Figure 13. In Figure 13, the computer 980 includes a CPU (Central Processing Unit) 981, a storage unit 982, a communication I / F (interface) 983, an input / output I / F 984, and a DSP (Digital Signal Processor) 985. Here, the storage unit 982 includes a RAM 982a and a ROM 982b.

[0099] The communication I / F 983 is connected to a communication circuit 986. The input / output I / F 984 is connected to an input / output device 987. The ROM 982b stores control programs executed by the CPU 981 and DSP 985, various data, etc. The CPU 981 and DSP 985 execute these control programs to realize various functions. The interior of the waveform control devices 100 and 100a shown in Figures 3 and 11 is shown as blocks representing functions realized by the control programs, etc.

[0100] Effects of the embodiment As described above, the water electrolytic cell system 8 includes an active power adjustment unit (B) that sets the active power correction amount del_Pfil such that the input active power (P_out) when the frequency deviation Δf is “0” is the reference active power P_s, the result obtained by subtracting the reference active power P_s from the input active power (P_out) is the active power correction amount del_Pfil, the first gain GA1 is the gain of the fluctuation amplitude of the active power correction amount del_Pfil with respect to the fluctuation amplitude of the frequency deviation Δf when the fluctuation period of the frequency deviation Δf is a predetermined first period TV1, and the second gain GA2 is the gain of the fluctuation amplitude of the active power correction amount del_Pfil with respect to the fluctuation amplitude of the frequency deviation Δf when the fluctuation period of the frequency deviation Δf is a predetermined second period TV2 that is shorter than the first period TV1, and that sets the active power correction amount del_Pfil so that the second gain GA2 is smaller than the first gain GA1.

[0101] This allows the water electrolytic cell system 8 to be appropriately connected to the AC system 1. That is, when the system frequency grid_f of the AC system 1 to which the water electrolytic cell system 8 is connected deviates from the normal range, active power adjustment can be made to stabilize the frequency. Furthermore, even when the system frequency grid_f has a high-frequency pulsating component, the influence of the high-frequency pulsating component can be suppressed in the active power correction performed by the water electrolytic cell system 8, thereby suppressing deterioration of the water electrolytic cell 5.

[0102] Furthermore, the active power adjustment unit (B) applies a gain to the frequency deviation Δf to determine the active power correction amount del_Pfil, and it is more preferable that the active power adjustment unit (B) has a calculation function that makes the gain for the active power correction amount del_Pfil smaller for fluctuating frequency components of the frequency deviation Δf that are higher than a predetermined frequency than the gain for other frequency components, thereby more appropriately suppressing the effects of high-frequency pulsating components.

[0103] Furthermore, the active power adjustment unit (B) applies a gain to the frequency deviation Δf to determine the active power correction amount del_Pfil, and it is more preferable that the active power adjustment unit (B) has a calculation function that makes the gain for the active power correction amount del_Pfil smaller for fluctuation period components of the frequency deviation Δf that are shorter than a predetermined fluctuation period than the gain for other fluctuation period components. This also makes it possible to more appropriately suppress the effects of high-frequency pulsating components.

[0104] Furthermore, it is more preferable that the water electrolytic cell system 8 includes an interface (210, 212) for an operator to set a characteristic value for adjusting the gain characteristic for the fluctuating frequency component or fluctuating periodic component of the frequency deviation Δf. This allows the operator to set an appropriate characteristic value depending on the state of the water electrolytic cell 5 or the AC system 1.

[0105] Furthermore, it is more preferable that the characteristic value is the time constant of a low-pass filter, whereby the gain characteristic can be adjusted using the time constant of the low-pass filter.

[0106] Furthermore, it is more preferable that the characteristic value is the width of the moving average window of the moving average filter, thereby making it possible to adjust the gain characteristic using the width of the moving average window of the moving average filter.

[0107] Furthermore, as in the third embodiment, it is more preferable that the water electrolytic cell system 8 further includes a power factor adjustment unit (1100) that sets the power factor of the power input to the power converter 3. This allows an appropriate power factor to be set for the power input to the power converter 3.

[0108] Furthermore, as in the third embodiment, it is more preferable that the power factor adjustment unit (1100) includes a correlation calculator (11005) that calculates a correlation value between a fluctuation component of the voltage amplitude in the AC system 1 and a fluctuation component of the input active power (P_out), and reactive power command units (11006, 11007) that calculate a reactive power command value by multiplying the learned power factor and active power based on the correlation value. This makes it possible to set the power factor in the power converter 3 based on the correlation value between the fluctuation component of the voltage amplitude in the AC system 1 and the fluctuation component of the input active power (P_out).

[0109] According to another aspect, each of the above-described embodiments comprises a self-excited power converter 3 that converts AC power received from an AC system 1 into DC power, a water electrolytic cell 5 that produces hydrogen using the DC power supplied from the power converter 3, a connection circuit (4) that is connected between the power converter 3 and the water electrolytic cell 5 and whose impedance has an inductance component, and a capacitor circuit (34) that is connected between a pair of DC output terminals of the power converter 3, wherein a resonant frequency determined by the equivalent capacitor capacitance of the capacitor circuit (34) and the inductance component in the connection circuit (4) is lower than a switching frequency (fsw) of the power converter 3. This makes it possible to suppress ripple components contained in the DC voltage supplied to the water electrolytic cell 5.

[0110] [Modifications] The present invention is not limited to the above-described embodiment, and various modifications are possible. The above-described embodiment is provided as an example to facilitate understanding of the present invention, and is not necessarily limited to an embodiment including all of the described configurations. Furthermore, it is possible to replace part of the configuration of one embodiment with the configuration of another embodiment, or to add the configuration of another embodiment to the configuration of one embodiment. Furthermore, it is possible to delete part of the configuration of each embodiment, or to add or replace other configurations. Furthermore, the control lines and information lines shown in the figures are those considered necessary for explanation, and do not necessarily represent all control lines and information lines necessary for the product. In reality, it is possible to consider that almost all components are interconnected. Possible modifications of the above-described embodiment include, for example, the following.

[0111] (1) Since the hardware of the waveform control devices 100, 100a and the upper control device 200 in the above embodiment can be realized by a general computer, the above-mentioned block diagrams and other programs for executing the various processes described above may be stored on a storage medium (a computer-readable storage medium on which a program is recorded) or distributed via a transmission path.

[0112] (2) In the above embodiment, the above-described block diagrams and various other processes are described as software processes using programs. However, some or all of them may be replaced with hardware processes using an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array), etc.

[0113] (3) The various processes executed in the above embodiment may be executed by a server computer via a network (not shown), and the various data stored in the above embodiment may also be stored in the server computer.

[0114] 1 AC system 3 Power converter 4 Cable (connection circuit) 5 Water electrolyzer 8 Water electrolyzer system 34 Capacitor (capacitor circuit) 210 Input unit (interface) 212 Display unit (interface) 1100 Reactive power command calculator (power factor adjustment unit) 11005 Multiplier (correlation calculator) 11006 Integrator (reactive power command unit) 11007 Multiplier (reactive power command unit) A System phase / frequency detection unit (frequency detection unit) B Adjustment power calculation / control unit (active power adjustment unit, active power adjustment process) Δf Frequency deviation GA1 First gain GA2 Second gain P_s Reference active power TV1 First period TV2 Second period fsw Frequency, switching frequency P_out Active power (input active power) τ_set Time constant grid_f System frequency f_rated Rated frequency del_Pfil Active power correction amount

Claims

1. A water electrolytic cell system comprising: a self-excited power converter that converts AC power received from an AC system into DC power; a water electrolytic cell that produces hydrogen using the DC power supplied from the power converter; a frequency detection unit that detects a system frequency that is the frequency of the AC system; and an active power adjustment unit that defines an input active power as the active power received from the AC system by the power converter, a frequency deviation as a result of subtracting a predetermined rated frequency from the system frequency, a reference active power as the input active power when the frequency deviation is "0", a result of subtracting the reference active power from the input active power as an active power correction amount, a first gain as a gain of the fluctuation amplitude of the active power correction amount relative to the fluctuation amplitude of the frequency deviation when the fluctuation period of the frequency deviation is a predetermined first period, and a second gain as a gain of the fluctuation amplitude of the active power correction amount relative to the fluctuation amplitude of the frequency deviation when the fluctuation period of the frequency deviation is a predetermined second period that is shorter than the first period, and sets the active power correction amount so that the second gain is smaller than the first gain.

2. The water electrolysis cell system according to claim 1, characterized in that the active power adjustment unit determines the active power correction amount by applying a gain to the frequency deviation, and has a calculation function that makes the gain for the active power correction amount for a fluctuating frequency component of the frequency deviation that is higher than a predetermined frequency smaller than the gain for other frequency components.

3. The water electrolysis cell system according to claim 1, characterized in that the active power adjustment unit determines the active power correction amount by applying a gain to the frequency deviation, and has a calculation function that makes the gain for the active power correction amount for a fluctuation period component of the frequency deviation that is shorter than a predetermined fluctuation period smaller than the gain for other fluctuation period components.

4. A water electrolysis cell system as claimed in claim 2 or 3, characterized in that it is provided with an interface for an operator to set a characteristic value for adjusting the gain characteristic for the fluctuating frequency component or fluctuating periodic component of the frequency deviation.

5. The water electrolysis cell system according to claim 4, characterized in that the characteristic value is a time constant of a low-pass filter.

6. The water electrolysis cell system according to claim 4, characterized in that the characteristic value is a width of a moving average window of a moving average filter.

7. The water electrolytic cell system according to claim 1, further comprising a power factor adjustment unit that sets the power factor of the power input to the power converter.

8. The water electrolysis cell system according to claim 7, characterized in that the power factor adjustment unit comprises: a correlation calculator that calculates a correlation value between a fluctuation component of a voltage amplitude in the AC system and a fluctuation component of the input effective power; and a reactive power command unit that calculates a reactive power command value by multiplying the effective power by a power factor learned based on the correlation value.

9. A water electrolyzer system comprising: a self-excited power converter that converts AC power received from an AC system into DC power; a water electrolyzer that produces hydrogen using the DC power supplied from the power converter; a connection circuit connected between the power converter and the water electrolyzer and having an impedance having an inductance component; and a capacitor circuit connected between a pair of DC output terminals of the power converter, wherein a resonant frequency determined by the equivalent capacitor capacitance of the capacitor circuit and the inductance component in the connection circuit is lower than a switching frequency of the power converter.

10. The water electrolytic cell system according to claim 1, further comprising: a connection circuit connected between the power converter and the water electrolytic cell, the impedance of which has an inductance component; and a capacitor circuit connected between a pair of DC output terminals of the power converter, wherein a resonant frequency determined by the equivalent capacitance of the capacitor circuit and the inductance component in the connection circuit is lower than a switching frequency in the power converter.

11. A control method for a water electrolyzer system comprising: a self-excited power converter that converts AC power received from an AC system into DC power; a water electrolyzer that produces hydrogen using the DC power supplied from the power converter; and a frequency detection unit that detects a system frequency that is the frequency of the AC system, the control method comprising: a step of acquiring input active power, which is active power received from the AC system by the power converter; and an active power adjustment step of setting a frequency deviation as a result of subtracting a predetermined rated frequency from the system frequency, setting the input active power when the frequency deviation is "0" as a reference active power, setting a result of subtracting the reference active power from the input active power as an active power correction amount, setting a first gain as a gain of the fluctuation amplitude of the active power correction amount relative to the fluctuation amplitude of the frequency deviation when the fluctuation period of the frequency deviation is a predetermined first period, setting a second gain as a gain of the fluctuation amplitude of the active power correction amount relative to the fluctuation amplitude of the frequency deviation when the fluctuation period of the frequency deviation is a predetermined second period that is shorter than the first period, and setting the active power correction amount so that the second gain is smaller than the first gain. A method for controlling a water electrolysis cell system.

Images