Alkaline and proton exchange membrane composite electrolytic cell based on renewable power

Abstract

The invention discloses an alkaline and proton exchange membrane composite electrolytic cell based on renewable power, which comprises a renewable power supply module, an alkaline electrolytic cell, a PEM (Proton Exchange Membrane) reversible fuel cell and an intelligent controller, the output end of the direct-current bus device is connected in parallel with an alkaline electrolytic cell, a PEM reversible fuel cell and a power grid; and the intelligent controller controls the working modes of the alkaline electrolytic cell and the PEM reversible fuel cell according to the direct current bus device power signal and the power grid power demand signal, wherein the working modes comprise an alkaline electrolytic cell electrolysis mode, a PEM electric pile electrolysis mode and a PEM electric pile power generation mode. Based on the combined application of the multiple electrolytic cells, the operation range of the composite electrolytic cell system can be widened, the combined stable operation of the alkaline electrolytic cell and the PEM reversible fuel cell is realized, and the hydrogen production efficiency of the composite electrolytic cell system is improved while the stable operation of a power grid is facilitated.

Description

Technical Field

[0001] The invention relates to the field of hydrogen production by electrolysis of water, and in particular to an alkaline and proton exchange membrane composite electrolyzer based on renewable electricity. Background Art

[0002] Regarding the electrolyzer configuration of green hydrogen projects, if alkaline and PEM can be properly matched, the hydrogen production efficiency can be fully improved. Early large-scale green hydrogen projects were basically equipped with only alkaline electrolyzers, but in current large-scale green hydrogen projects, the "alkaline + PEM" combination has become a new trend.

[0003] Alkaline and PEM electrolyzers each have their own advantages. Alkaline electrolyzers have a higher level of technical commercial maturity, a large amount of hydrogen produced by single-cell electrolysis, and no precious metals in the anode and cathode plates, so the production cost is relatively low; however, alkaline electrolyzers have higher requirements for power stability and are not suitable for intermittent electricity such as wind and solar power, but are more suitable for grid electrolysis and hydrogen production. Compared with alkaline electrolyzers, PEM electrolyzers have the advantages of high current density, high hydrogen purity, fast response speed, etc., and higher hydrogen production efficiency; however, precious metal materials such as iridium, platinum, and titanium are used in the production of PEM electrolyzers, and the current production cost is relatively high, about 4-6 times that of alkaline electrolyzers.

[0004] With the breakthrough of PEM electrolysis hydrogen production core technology and the reduction of cost in China, the proportion of PEM electrolyzers will increase in future market applications, and the "alkaline + PEM" combination electrolyzer will become the choice of more green hydrogen projects. However, in actual applications, there are still problems to be solved, such as the volatility of renewable electricity and the control of two electrolyzers. Summary of the invention

[0005] Purpose of the invention: The purpose of the present invention is to provide an alkaline and proton exchange membrane composite electrolyzer based on renewable electricity, so as to realize the combined stable operation of the alkaline electrolyzer and the PEM reversible fuel cell.

[0006] Technical solution: To achieve the above-mentioned purpose, the alkaline and proton exchange membrane composite electrolyzer based on renewable electricity described in the present invention includes a renewable power module, an alkaline electrolyzer, a PEM reversible fuel cell, and an intelligent controller. The output end of the renewable power module is connected to the input end of a DC bus device, and the output end of the DC bus device is connected in parallel with an alkaline electrolyzer, a PEM reversible fuel cell, and a power grid; the intelligent controller controls the working modes of the alkaline electrolyzer and the PEM reversible fuel cell according to the power signal of the DC bus device and the power demand signal of the power grid, including the alkaline electrolyzer electrolysis mode, the PEM stack electrolysis mode, and the PEM stack power generation mode.

[0007] The output end of the DC bus device is connected to the power grid through the power interface of the alkaline electrolysis step-down device and the alkaline electrolyzer, through the power interface of the PEM electrolysis step-down device and the PEM reversible fuel cell, and through the grid-connected inverter; the intelligent controller controls the working states of the grid-connected inverter, the alkaline electrolysis step-down device, the PEM electrolysis step-down device, and the PEM inverter, thereby controlling the working modes of the alkaline electrolyzer and the PEM reversible fuel cell.

[0008] Wherein, the method for determining the installed capacity of the alkaline electrolyzer and PEM reversible fuel cell is:

[0009] Q0=0.8*(Q1)(1-η)+0.2*(Q2)*η;

[0010] Q2 = Q0 - Q1;

[0011] Wherein, Q0 is the installed capacity required by the composite electrolyzer system, Q1 is the installed capacity of the alkaline electrolyzer, and Q2 is the installed capacity of the PEM reversible fuel cell; η is the response rate of the composite electrolyzer system, 0<η<1, when the response requirement of the composite electrolyzer system is high, η=0.8 is taken, and when the response of the composite electrolyzer system is low, η=0.2 is taken.

[0012] Among them, the renewable power supply module includes photovoltaic cells, wind turbines, and wind power rectifiers. The photovoltaic cells output direct current, and its output end is connected to the input end of the DC bus device. The wind turbine outputs alternating current, and its output end is connected to the input end of the DC bus device through the wind power rectifier.

[0013] Among them, it also includes a water tank, a hydrogen storage device, and a collection module for collecting power signals of a DC bus device and power demand signals of a power grid. The water outlet of the water tank is respectively connected to the water inlet of an alkaline electrolyzer and a PEM reversible fuel cell. An alkaline electrolyzer water flow control pump is provided on a pipe connecting the water tank and the alkaline electrolyzer, and a PEM electrolyzer water flow control pump is provided on a pipe connecting the water tank and the PEM reversible fuel cell. The hydrogen storage device is respectively connected to the hydrogen outlet of the alkaline electrolyzer, the hydrogen outlet of the PEM reversible fuel cell, and the hydrogen inlet of the PEM reversible fuel cell. An alkaline electrolyzer hydrogen flow output control pump is provided on the pipe connecting the hydrogen storage device and the alkaline electrolyzer, a PEM electrolyzer hydrogen flow output control pump is provided on the pipe connecting the hydrogen storage device and the hydrogen outlet of the PEM reversible fuel cell, and a PEM electrolyzer hydrogen flow input control pump is provided on the pipe connecting the hydrogen storage device and the hydrogen inlet of the PEM reversible fuel cell. The collection module transmits the collected signals to the intelligent controller.

[0014] Wherein, the air inlet end of the PEM reversible fuel cell is provided with a PEM electrolytic cell air flow input control pump, and the power output port of the PEM reversible fuel cell is connected to the power grid through a PEM inverter.

[0015] The intelligent controller controls the working mode of the alkaline electrolyzer and the PEM reversible fuel cell according to the power signal of the DC bus device and the power demand signal of the power grid, specifically by using a fuzzy neural network for control, and the control method is:

[0016] Step 1: Determine the input and output variables,

[0017] Among them, the input variables include input variable 1: grid-connected power demand, input variable 2: DC bus power; the output variables include output variable 1: PEM stack electrolysis power, output variable 2: PEM stack power generation power, output variable 3: alkaline electrolyzer electrolysis power;

[0018] Step 2: Determine the domain and quantization factors of input and output variables;

[0019] Step 3: Design fuzzy control rules based on the domain of input and output variables: When the grid-connected power demand is greater than the DC bus power, the PEM stack power generation mode is used; when the grid-connected power demand is less than the DC bus power, and the difference between the grid-connected power demand and the DC bus power is greater than the set value, the PEM stack electrolysis mode is started first, and when the PEM stack electrolysis mode cannot meet the electrolysis demand, the alkaline electrolyzer electrolysis mode is started; when the grid-connected power demand is less than the DC bus power, and the difference between the grid-connected power demand and the DC bus power is less than the set value, only the PEM stack electrolysis mode is started;

[0020] Step 4: Output defuzzification, using the maximum membership method for defuzzification, specifically: according to the fuzzy control rules, get the output set, the output set is a fuzzy quantity, and defuzzify it. The defuzzification process is: using the maximum membership method, in the fuzzy output set, determine the element with the highest membership, and then select the exact value corresponding to the element as the output.

[0021] Among them, the signal of the DC bus power in step 1 is a composite signal, and the grid-connected power demand is defined as P1, then P1=0.7*Pt1+0.3*(Pt1-Pt2) / (t1-t2), where Pt1 is the grid-connected power demand at the current moment t1, and Pt2 is the grid-connected power demand at the previous moment t2. The moments t1 and t2 can be set to a time interval of 1 second according to the operating parameters of the composite electrolyzer system, that is, t1-t2=1 second.

[0022] The method for determining the domain and quantization factor of the input and output variables in step 2 is:

[0023] The basic domain of input variable 1 is (-1000kW, 0). After normalization, the continuous variable of the basic domain (-1000kW, 0) is converted into the continuous variable of the standard domain (-6, +6). The variables of the standard domain are then divided into 5 linguistic variables E1 {positive large PB, positive middle PM, zero ZE, negative middle NM, positive small NB};

[0024] The domain of the language variable E1 is further selected as: X1 = {-6, -5, -4, -3, -2, -1, 0, +1, +2, +3, +4, +5, +6}, then the quantization factor K1 = 0.012;

[0025] The basic domain of input variable 2 is (0, 1000kW). After normalization, the continuous variable of the basic domain (0, 1000kW) is converted into the continuous variable of the standard domain (-6, +6). The variables of the standard domain are then divided into 5 linguistic variables E2 {positive large PB, positive middle PM, zero ZE, negative middle NM, positive small NB};

[0026] The domain of the language variable E2 is further selected as: X2 = {-6, -5, -4, -3, -2, -1, 0, +1, +2, +3, +4, +5, +6}, then the quantization factor K2 = -0.012;

[0027] The basic domain of output variable 1 is (-100kW, 0). After normalization, the continuous variable of the basic domain (-100kW, 0) is converted into the continuous variable of the standard domain (-6, +6). The variables of the standard domain are then divided into 5 linguistic variables E 3 {positive large PB, positive middle PM, zero ZE, negative middle NM, positive small NB};

[0028] The domain of the language variable E3 is further selected as: Y1 = {-6, -5, -4, -3, -2, -1, 0, +1, +2, +3, +4, +5, +6}, then the quantization factor K3 = 0.12;

[0029] The basic domain of output variable 2 is (0, 100kW). After normalization, the continuous variable of the basic domain (0, 100kW) is converted into the continuous variable of the standard domain (-6, +6). The variable of the standard domain is then divided into 5 linguistic variables E4 {positive large PB, positive middle PM, zero ZE, negative middle NM, positive small NB};

[0030] The domain of the language variable E4 is further selected as: Y2 = {-6, -5, -4, -3, -2, -1, 0, +1, +2, +3, +4, +5, +6}, then the quantization factor K4 = 0.12;

[0031] The basic domain of the output variable Y3 is (-300kW, 0). After normalization, the continuous variable of the basic domain (-300kW, 0) is converted into the continuous variable of the standard domain (-6, +6). The variables of the standard domain are then divided into 5 linguistic variables E5 {positive large PB, positive middle PM, zero ZE, negative middle NM, positive small NB};

[0032] The domain of the language variable E5 is further selected as: Y3 = {-6, -5, -4, -3, -2, -1, 0, +1, +2, +3, +4, +5, +6}, then the quantization factor K5 = 0.36;

[0033] The positive value in the domain represents the emitted power, and the negative value represents the absorbed power.

[0034] The specific method of designing the fuzzy control rules in step 3 is as follows: Rule 1: If X1 is PB and X2 is PB, then Y1 is NB, Y2 is NB, and Y3 is NB;

[0035] Rule 2: If X1 is PB and X2 is PM, then Y1 is NB, Y2 is NM, and Y3 is NB;

[0036] Rule 3: If X1 is PB and X2 is ZE, then Y1 is NB, Y2 is ZE, and Y3 is NB;

[0037] Rule 4: If X1 is PB and X2 is NM, then Y1 is NB, Y2 is PM, and Y3 is ZE;

[0038] Rule 5: If X1 is PB and X2 is NB, then Y1 is NB, Y2 is PB, and Y3 is NB;

[0039] Rule 6: If X1 is PM and X2 is PB, then Y1 is NB, Y2 is PM, and Y3 is NB;

[0040] Rule 7: If X1 is PM, X2 is PM, then Y1 is NB, Y2 is PM, and Y3 is NB;

[0041] Rule 8: If X1 is PM and X2 is ZE, then Y1 is NB, Y2 is PM, and Y3 is NB;

[0042] Rule 9: If X1 is PM and X2 is NM, then Y1 is NB, Y2 is PM, and Y3 is ZE;

[0043] Rule 10: If X1 is PM and X2 is NB, then Y1 is NB, Y2 is PM, and Y3 is NB;

[0044] Rule 11: If X1 is ZE and X2 is PB, then Y1 is PB, Y2 is NB, and Y3 is PB;

[0045] Rule 12: If X1 is ZE and X2 is PM, then Y1 is PB, Y2 is ZE, and Y3 is PM;

[0046] Rule 13: If X1 is ZE, X2 is ZE, then Y1 is ZE, Y2 is ZE, and Y3 is ZE;

[0047] Rule 14: If X1 is ZE and X2 is NM, then Y1 is NB, Y2 is PM, and Y3 is ZE;

[0048] Rule 15: If X1 is ZE and X2 is NB, then Y1 is NB, Y2 is PB, and Y3 is NB;

[0049] Rule 16: If X1 is NM and X2 is PB, then Y1 is PB, Y2 is NB, and Y3 is ZE;

[0050] Rule 17: If X1 is NM and X2 is PM, then Y1 is PM, Y2 is NB, and Y3 is ZE;

[0051] Rule 18: If X1 is NM and X2 is ZE, then Y1 is PM, Y2 is NB, and Y3 is ZE;

[0052] Rule 19: If X1 is NM, X2 is NM, then Y1 is ZE, Y2 is ZE, and Y3 is ZE;

[0053] Rule 20: If X1 is NM and X2 is NB, then Y1 is NB, Y2 is PM, and Y3 is NB;

[0054] Rule 21: If X1 is NB and X2 is PB, then Y1 is PM, Y2 is ZE, and Y3 is PM;

[0055] Rule 22: If X1 is NB and X2 is PM, then Y1 is PB, Y2 is NB, and Y3 is PM;

[0056] Rule 23: If X1 is NB and X2 is ZE, then Y1 is PM, Y2 is NB, and Y3 is PM;

[0057] Rule 24: If X1 is NB and X2 is NM, then Y1 is PM, Y2 is NB, and Y3 is ZE;

[0058] Rule 25: If X1 is NB, X2 is NB, then Y1 is ZE, Y2 is ZE, and Y3 is NB.

[0059] Beneficial effects: The present invention has the following advantages: 1. The combination of multiple electrolyzers can broaden the operating range of the composite electrolyzer system, realize the joint operation of the alkaline electrolyzer and the PEM reversible fuel cell, improve the hydrogen production efficiency of the composite electrolyzer system, and reduce the hydrogen production cost;

[0060] 2. The present invention optimizes and controls the working modes of the alkaline electrolyzer and the PEM reversible fuel cell based on the fuzzy neural network, and can adjust the operating states of the alkaline electrolyzer and the PEM reversible fuel cell in real time according to the grid power demand and the DC bus power, which is helpful to stabilize the operation of the power grid and can cope with the complex and changeable power system environment;

[0061] 3. When the present invention processes the real-time control problem of renewable energy power generation equipment based on fuzzy neural network, it can be more efficient, intuitive and easy to implement compared with traditional control algorithms, especially in application scenarios that require fast response and strong robustness;

[0062] 4. The present invention can promote the advancement of alkaline water electrolysis hydrogen production technology, improve the problem of safety risks in the low-load operation range, and at the same time form a MW-level standardized container-type high-safety, high-dynamic and high-intelligence next-generation advanced alkaline water electrolysis hydrogen production system, realizing remote monitoring, unmanned operation, safety warning, and system health status monitoring of 100% off-grid renewable energy hydrogen production system control. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] Figure 1 This is a schematic diagram of the structure of the composite electrolytic cell of the present invention.

[0064] Among them: 1-photovoltaic cell, 2-wind turbine, 3-wind power rectifier, 4-DC bus device, 5-grid-connected inverter, 6-alkaline electrolysis step-down device, 7-PEM electrolysis step-down device, 8-PEM inverter, 9-alkaline electrolyzer, 10-PEM reversible fuel cell, 11-acquisition module, 12-intelligent controller, 13-water tank, 14-hydrogen storage device, 15-alkaline electrolysis cell water flow control pump, 16-alkaline electrolysis cell hydrogen flow output control pump, 17-PEM electrolysis cell water flow control pump, 18-PEM electrolysis cell hydrogen flow output control pump, 19-PEM electrolysis cell hydrogen flow input control pump, 20-PEM electrolysis cell air flow input control pump, 21-grid. DETAILED DESCRIPTION

[0065] The technical solution of the present invention is described in detail below in conjunction with the embodiments and drawings.

[0066] like Figure 1 As shown, the alkaline and proton exchange membrane composite electrolyzer based on renewable electricity of the present invention includes a power module, an alkaline electrolyzer 9, a PEM reversible fuel cell 10, a collection module 11, an intelligent controller 12, a water tank 13, and a hydrogen storage device 14.

[0067] The power module includes a photovoltaic cell 1, a wind turbine 2, and a wind power rectifier 3. The photovoltaic cell 1 outputs direct current, and its output end is directly connected to the input end of a direct current bus device 4. The wind turbine 2 outputs alternating current, and its output end is connected to the input end of a direct current bus device 4 through the wind power rectifier 3.

[0068] The output end of the DC bus device 4 is connected to the power interface of the alkaline electrolytic cell 9 through the alkaline electrolytic voltage reducer 6, connected to the power interface of the PEM reversible fuel cell 10 through the PEM electrolytic voltage reducer 7, and connected to the power grid 21 through the grid-connected inverter 5.

[0069] The water outlet of the water tank 13 is connected to the water inlet of the alkaline electrolyzer 9 and the PEM reversible fuel cell 10 respectively. An alkaline electrolyzer water flow control pump 15 is provided on the connecting pipe between the water tank 13 and the alkaline electrolyzer 9, and a PEM electrolyzer water flow control pump 17 is provided on the connecting pipe between the water tank 13 and the PEM reversible fuel cell 10.

[0070] The hydrogen storage device 14 is respectively connected to the hydrogen outlet of the alkaline electrolyzer 9, the hydrogen outlet of the PEM reversible fuel cell 10, and the hydrogen inlet of the PEM reversible fuel cell 10. The pipeline connecting the hydrogen storage device 14 and the alkaline electrolyzer 9 is provided with an alkaline electrolyzer hydrogen flow output control pump 16, the pipeline connecting the hydrogen storage device 14 and the hydrogen outlet of the PEM reversible fuel cell 10 is provided with a PEM electrolyzer hydrogen flow output control pump 18, and the pipeline connecting the hydrogen storage device 14 and the hydrogen inlet of the PEM reversible fuel cell 10 is provided with a PEM electrolyzer hydrogen flow input control pump 19. The air inlet end of the PEM reversible fuel cell 10 is provided with a PEM electrolyzer air flow input control pump 20, and the power output port of the PEM reversible fuel cell 10 is connected to the power grid 21 through the PEM inverter 8.

[0071] The acquisition module 11 is used to collect the power signal of the photovoltaic cell 1 (the photovoltaic cell 1 is connected to the DC bus, the DC bus voltage and power are related to the photovoltaic cell 1, and the voltage and power of the DC bus are the sum of the outputs of the photovoltaic cell 1 and the wind turbine 2), the power signal of the DC bus device 4, and the power demand signal of the power grid 21. The intelligent controller 12 controls the working state of the grid-connected inverter 5, the alkaline electrolysis step-down device 6, the PEM electrolysis step-down device 7, and the PEM inverter 8 according to the signal collected by the acquisition module 11, thereby controlling the working mode of the alkaline electrolyzer 9 and the PEM reversible fuel cell 10, namely, the alkaline electrolyzer electrolysis mode, the PEM stack electrolysis mode and the PEM stack power generation mode.

[0072] The control of the working states of the grid-connected inverter 5, the alkaline electrolysis voltage reducer 6, the PEM electrolysis voltage reducer 7, and the PEM inverter 8, thereby controlling the working modes of the alkaline electrolyzer 9 and the PEM reversible fuel cell 10, refers to: output variable 1: the output signal of the PEM electrolysis power controls the PEM electrolysis voltage reducer (7); output variable 2: the output signal of the PEM power generation power controls the PEM inverter (8) and the grid-connected inverter (5); output variable 3: the output signal of the alkaline electrolysis power controls the alkaline electrolysis voltage reducer (6).

[0073] In the alkaline electrolyzer electrolysis mode and the PEM stack electrolysis mode, the DC bus device 4 and the water tank 13 provide the alkaline electrolyzer 9 and the PEM reversible fuel cell 10 with necessary DC power and electrolyzed water in the hydrogen production process, and the hydrogen produced by the alkaline electrolyzer 9 and the PEM reversible fuel cell 10 is stored in the hydrogen storage device 14. In the PEM stack power generation mode, the hydrogen storage device 14 provides the PEM reversible fuel cell 10 with the hydrogen required for power generation, and introduces external air through the PEM electrolyzer air flow input control pump 20, and the electric energy generated by the PEM reversible fuel cell 10 is incorporated into the power grid 21 through the PEM inverter 8.

[0074] The method by which the intelligent controller 12 controls the working modes of the alkaline electrolyzer 9 and the PEM reversible fuel cell 10 is:

[0075] Step 1: Determine the input and output variables

[0076] The input variables include input variable 1: grid-connected power demand, input variable 2: DC bus power;

[0077] Among them, the signal of DC bus power is a composite signal, and the grid-connected power demand is defined as P1, then P1 = 0.7*Pt1+0.3*(Pt1-Pt2) / (t1-t2), where Pt1 is the grid-connected power demand at the current moment t1, and Pt2 is the grid-connected power demand at the previous moment t2. The moments t1 and t2 can be set to a time interval of 1 second according to the system operation parameters, that is, t1-t2 = 1 second;

[0078] The output variables include output variable 1: PEM stack electrolysis power, output variable 2: PEM stack power generation power, and output variable 3: alkaline electrolyzer electrolysis power.

[0079] Step 2: Determine the domain and quantization factors of input and output variables

[0080] The basic domain of input variable 1 is (-1000kW, 0). After normalization, the continuous variable of the basic domain (-1000kW, 0) is converted into the continuous variable of the standard domain (-6, +6). The variables of the standard domain are then divided into 5 linguistic variables E1 {positive large PB, positive middle PM, zero ZE, negative middle NM, positive small NB};

[0081] The domain of the language variable E1 is further selected as: X1 = {-6, -5, -4, -3, -2, -1, 0, +1, +2, +3, +4, +5, +6}, then the quantization factor K1 = 0.012;

[0082] The basic domain of input variable 2 is (0, 1000kW). After normalization, the continuous variable of the basic domain (0, 1000kW) is converted into the continuous variable of the standard domain (-6, +6). The variables of the standard domain are then divided into 5 linguistic variables E2 {positive large PB, positive middle PM, zero ZE, negative middle NM, positive small NB};

[0083] The domain of the language variable E2 is further selected as: X2 = {-6, -5, -4, -3, -2, -1, 0, +1, +2, +3, +4, +5, +6}, then the quantization factor K2 = -0.012;

[0084] The basic domain of output variable 1 is (-100kW, 0). After normalization, the continuous variable of the basic domain (-100kW, 0) is converted into the continuous variable of the standard domain (-6, +6). The variables of the standard domain are then divided into 5 linguistic variables E 3 {positive large PB, positive middle PM, zero ZE, negative middle NM, positive small NB};

[0085] The domain of the language variable E3 is further selected as: Y1 = {-6, -5, -4, -3, -2, -1, 0, +1, +2, +3, +4, +5, +6}, then the quantization factor K3 = 0.12;

[0086] The basic domain of output variable 2 is (0, 100kW). After normalization, the continuous variable of the basic domain (0, 100kW) is converted into the continuous variable of the standard domain (-6, +6). The variable of the standard domain is then divided into 5 linguistic variables E4 {positive large PB, positive middle PM, zero ZE, negative middle NM, positive small NB};

[0087] The domain of the language variable E4 is further selected as: Y2 = {-6, -5, -4, -3, -2, -1, 0, +1, +2, +3, +4, +5, +6}, then the quantization factor K4 = 0.12;

[0088] The basic domain of the output variable Y3 is (-300kW, 0). After normalization, the continuous variable of the basic domain (-300kW, 0) is converted into the continuous variable of the standard domain (-6, +6). The variables of the standard domain are then divided into 5 linguistic variables E5 {positive large PB, positive middle PM, zero ZE, negative middle NM, positive small NB};

[0089] The domain of the language variable E5 is further selected as: Y3 = {-6, -5, -4, -3, -2, -1, 0, +1, +2, +3, +4, +5, +6}, then the quantization factor K5 = 0.36;

[0090] The positive value in the domain represents the emitted power, and the negative value represents the absorbed power.

[0091] Step 3: Design fuzzy control rules based on the domain of input and output variables (: When the grid-connected power demand is greater than the DC bus power, the PEM stack power generation mode is used; when the grid-connected power demand is less than the DC bus power, and the difference between the grid-connected power demand and the DC bus power is greater than the set value, the PEM stack electrolysis mode is started first. When the PEM stack electrolysis mode cannot meet the electrolysis demand, the alkaline electrolyzer electrolysis mode is started; when the grid-connected power demand is less than the DC bus power, and the difference between the grid-connected power demand and the DC bus power is less than the set value, only the PEM stack electrolysis mode is started. The specific control rules are:

[0092] Rule 1: If X1 is PB, X2 is PB, then Y1 is NB, Y2 is NB, and Y3 is NB;

[0093] Rule 2: If X1 is PB and X2 is PM, then Y1 is NB, Y2 is NM, and Y3 is NB;

[0094] Rule 3: If X1 is PB and X2 is ZE, then Y1 is NB, Y2 is ZE, and Y3 is NB;

[0095] Rule 4: If X1 is PB and X2 is NM, then Y1 is NB, Y2 is PM, and Y3 is ZE;

[0096] Rule 5: If X1 is PB and X2 is NB, then Y1 is NB, Y2 is PB, and Y3 is NB;

[0097] Rule 6: If X1 is PM and X2 is PB, then Y1 is NB, Y2 is PM, and Y3 is NB;

[0098] Rule 7: If X1 is PM, X2 is PM, then Y1 is NB, Y2 is PM, and Y3 is NB;

[0099] Rule 8: If X1 is PM and X2 is ZE, then Y1 is NB, Y2 is PM, and Y3 is NB;

[0100] Rule 9: If X1 is PM and X2 is NM, then Y1 is NB, Y2 is PM, and Y3 is ZE;

[0101] Rule 10: If X1 is PM and X2 is NB, then Y1 is NB, Y2 is PM, and Y3 is NB;

[0102] Rule 11: If X1 is ZE and X2 is PB, then Y1 is PB, Y2 is NB, and Y3 is PB;

[0103] Rule 12: If X1 is ZE and X2 is PM, then Y1 is PB, Y2 is ZE, and Y3 is PM;

[0104] Rule 13: If X1 is ZE, X2 is ZE, then Y1 is ZE, Y2 is ZE, and Y3 is ZE;

[0105] Rule 14: If X1 is ZE and X2 is NM, then Y1 is NB, Y2 is PM, and Y3 is ZE;

[0106] Rule 15: If X1 is ZE and X2 is NB, then Y1 is NB, Y2 is PB, and Y3 is NB;

[0107] Rule 16: If X1 is NM and X2 is PB, then Y1 is PB, Y2 is NB, and Y3 is ZE;

[0108] Rule 17: If X1 is NM and X2 is PM, then Y1 is PM, Y2 is NB, and Y3 is ZE;

[0109] Rule 18: If X1 is NM and X2 is ZE, then Y1 is PM, Y2 is NB, and Y3 is ZE;

[0110] Rule 19: If X1 is NM, X2 is NM, then Y1 is ZE, Y2 is ZE, and Y3 is ZE;

[0111] Rule 20: If X1 is NM and X2 is NB, then Y1 is NB, Y2 is PM, and Y3 is NB;

[0112] Rule 21: If X1 is NB and X2 is PB, then Y1 is PM, Y2 is ZE, and Y3 is PM;

[0113] Rule 22: If X1 is NB and X2 is PM, then Y1 is PB, Y2 is NB, and Y3 is PM;

[0114] Rule 23: If X1 is NB and X2 is ZE, then Y1 is PM, Y2 is NB, and Y3 is PM;

[0115] Rule 24: If X1 is NB and X2 is NM, then Y1 is PM, Y2 is NB, and Y3 is ZE;

[0116] Rule 25: If X1 is NB, X2 is NB, then Y1 is ZE, Y2 is ZE, and Y3 is NB.

[0117] Step 4: Output defuzzification,

[0118] The maximum membership method is used for defuzzification. According to the above fuzzy control rules, an output set is obtained. The output set is a fuzzy quantity and needs to be defuzzified. The defuzzification process is: using the maximum membership method, in the fuzzy output set, determine which element has the highest membership, and then select the exact value corresponding to the element as the output. The output includes output variable 1: PEM stack electrolysis power, output variable 2: PEM stack power generation power, output variable 3: alkaline electrolyzer electrolysis power.

[0119] The present invention can provide a fuel cell full life cycle monitoring and evaluation system based on the alkaline and proton exchange membrane composite electrolyzer of renewable electricity. Under standard operating conditions, the system can monitor the output voltage of the fuel cell stack in real time, and compare it with the standard voltage and the voltage value 1 hour ago, respectively, to obtain the difference with the standard voltage and the voltage attenuation value of 1 hour, and then obtain the performance index of the fuel cell stack through the fuzzy algorithm. Under non-standard conditions, by detecting the hydrogen and oxygen input flow values ​​of the battery stack and the current operating temperature, the reference voltage value under the current working condition is obtained through the BP application network algorithm, and the obtained reference voltage is used as the standard voltage required by the fuzzy algorithm, so as to obtain the battery stack performance index under the actual working condition.

[0120] In addition, this embodiment also provides a method for determining the installed capacity of an alkaline electrolyzer 9 and a PEM reversible fuel cell 10, as Figure 1 The reference basis for the implementation of the composite electrolyzer system shown is based on the economic and kinetic composite model:

[0121] Q0=0.8*(Q1)(1-η)+0.2*(Q2)*η;

[0122] Q2 = Q0 - Q1;

[0123] In the formula, Q0 is Figure 1 The installed capacity required by the composite electrolyzer system is shown in FIG. 1 , where Q1 is the installed capacity of the alkaline electrolyzer 9, and Q2 is the installed capacity of the PEM reversible fuel cell 10; η is the system response rate, 0

Claims

1. An alkaline and proton exchange membrane composite electrolyzer based on renewable electricity, comprising a renewable power module, an alkaline electrolyzer (9), a PEM reversible fuel cell (10), and an intelligent controller (12), characterized in that: The output end of the renewable power module is connected to the input end of a DC bus device (4); the output end of the DC bus device (4) is connected in parallel with an alkaline electrolyzer (9), a PEM reversible fuel cell (10), and a power grid (21); the intelligent controller (12) controls the working modes of the alkaline electrolyzer (9) and the PEM reversible fuel cell (10) according to a power signal of the DC bus device (4) and a power demand signal of the power grid (21), including an alkaline electrolyzer electrolysis mode, a PEM stack electrolysis mode, and a PEM stack power generation mode.

2. The alkaline and proton exchange membrane composite electrolyzer based on renewable electricity according to claim 1, characterized in that: The output end of the DC bus device (4) is connected to the power supply interface of the alkaline electrolysis step-down device (6) and the alkaline electrolytic cell (9), the power supply interface of the PEM electrolysis step-down device (7) and the PEM reversible fuel cell (10), and the grid-connected inverter (5) and the power grid (21); the intelligent controller (12) controls the working states of the grid-connected inverter (5), the alkaline electrolysis step-down device (6), the PEM electrolysis step-down device (7), and the PEM inverter (8), thereby controlling the working modes of the alkaline electrolytic cell (9) and the PEM reversible fuel cell (10).

3. The alkaline and proton exchange membrane composite electrolyzer based on renewable electricity according to claim 2, characterized in that: The method for determining the installed capacity of the alkaline electrolyzer (9) and the PEM reversible fuel cell (10) is as follows: Q0=0.8*(Q1)(1-η)+0.2*(Q2)*η; Q2 = Q0 - Q1; In the formula, Q0 is the installed capacity required by the composite electrolyzer system, Q1 is the installed capacity of the alkaline electrolyzer (9), and Q2 is the installed capacity of the PEM reversible fuel cell (10); η is the response rate of the composite electrolyzer system, 0<η<1, when the response requirement of the composite electrolyzer system is high, η=0.8 is taken, and when the response of the composite electrolyzer system is low, η=0.2 is taken.

4. The alkaline and proton exchange membrane composite electrolyzer based on renewable electricity according to claim 1, characterized in that: The renewable power supply module comprises a photovoltaic cell (1), a wind turbine generator set (2), and a wind power rectifier (3); the photovoltaic cell (1) outputs direct current, and its output end is connected to the input end of a direct current bus device (4); the wind turbine generator set (2) outputs alternating current, and its output end is connected to the input end of the direct current bus device (4) through the wind power rectifier (3).

5. The alkaline and proton exchange membrane composite electrolyzer based on renewable electricity according to claim 1, characterized in that: The invention also comprises a water tank (13), a hydrogen storage device (14), and a collection module (11) for collecting power signals of a DC bus device (4) and power demand signals of a power grid (21); the water outlet of the water tank (13) is respectively connected to the water inlet of an alkaline electrolyzer (9) and a PEM reversible fuel cell (10); an alkaline electrolyzer water flow control pump (15) is provided on a pipe connecting the water tank (13) and the alkaline electrolyzer (9); and a PEM electrolyzer water flow control pump (17) is provided on a pipe connecting the water tank (13) and the PEM reversible fuel cell (10); the hydrogen storage device (14) is respectively connected to the hydrogen inlet of the alkaline electrolyzer (9) and the PEM reversible fuel cell (10); The outlet of the hydrogen storage device (14), the hydrogen outlet of the PEM reversible fuel cell (10), and the hydrogen inlet of the PEM reversible fuel cell (10) are connected; an alkaline electrolytic cell hydrogen flow output control pump (16) is provided on the connecting pipe between the hydrogen storage device (14) and the alkaline electrolytic cell (9); a PEM electrolytic cell hydrogen flow output control pump (18) is provided on the connecting pipe between the hydrogen storage device (14) and the hydrogen outlet of the PEM reversible fuel cell (10); and a PEM electrolytic cell hydrogen flow input control pump (19) is provided on the connecting pipe between the hydrogen storage device (14) and the hydrogen inlet of the PEM reversible fuel cell (10); and the acquisition module (11) transmits the acquired signal to the intelligent controller (12).

6. The alkaline and proton exchange membrane composite electrolyzer based on renewable electricity according to claim 1, characterized in that: The air inlet end of the PEM reversible fuel cell (10) is provided with a PEM electrolytic cell air flow input control pump (20), and the power output port of the PEM reversible fuel cell (10) is connected to the power grid (21) through a PEM inverter (8).

7. The alkaline and proton exchange membrane composite electrolyzer based on renewable electricity according to claim 1, characterized in that: The intelligent controller (12) controls the working modes of the alkaline electrolyzer (9) and the PEM reversible fuel cell (10) according to the power signal of the DC bus device (4) and the power demand signal of the power grid (21), specifically using a fuzzy neural network for control, and the control method is: Step 1: Determine the input and output variables, Among them, the input variables include input variable 1: grid-connected power demand, input variable 2: DC bus power; the output variables include output variable 1: PEM stack electrolysis power, output variable 2: PEM stack power generation power, output variable 3: alkaline electrolyzer electrolysis power; Step 2: Determine the domain and quantization factors of input and output variables; Step 3: Design fuzzy control rules based on the domain of input and output variables: When the grid-connected power demand is greater than the DC bus power, the PEM stack power generation mode is used; when the grid-connected power demand is less than the DC bus power, and the difference between the grid-connected power demand and the DC bus power is greater than the set value, the PEM stack electrolysis mode is started first, and when the PEM stack electrolysis mode cannot meet the electrolysis demand, the alkaline electrolyzer electrolysis mode is started; when the grid-connected power demand is less than the DC bus power, and the difference between the grid-connected power demand and the DC bus power is less than the set value, only the PEM stack electrolysis mode is started; Step 4: Output defuzzification, using the maximum membership method for defuzzification, specifically: according to the fuzzy control rules, get the output set, the output set is a fuzzy quantity, and defuzzify it. The defuzzification process is: using the maximum membership method, in the fuzzy output set, determine the element with the highest membership, and then select the exact value corresponding to the element as the output.

8. The alkaline and proton exchange membrane composite electrolyzer based on renewable electricity according to claim 7, characterized in that: The signal of the DC bus power in step 1 is a composite signal, and the grid-connected power demand is defined as P1, then P1=0.7*Pt1+0.3*(Pt1-Pt2) / (t1-t2), where Pt1 is the grid-connected power demand at the current moment t1, and Pt2 is the grid-connected power demand at the previous moment t2. The moments t1 and t2 can be set to a time interval of 1 second according to the operating parameters of the composite electrolyzer system, that is, t1-t2=1 second.

9. The alkaline and proton exchange membrane composite electrolyzer based on renewable electricity according to claim 7, characterized in that: The method for determining the domain and quantization factor of the input and output variables described in step 2 is: The basic domain of input variable 1 is (-1000kW, 0). After normalization, the continuous variable of the basic domain (-1000kW, 0) is converted into the continuous variable of the standard domain (-6, +6). The variables of the standard domain are then divided into 5 linguistic variables E1 {positive large PB, positive middle PM, zero ZE, negative middle NM, positive small NB}; The domain of the language variable E1 is further selected as: X1 = {-6, -5, -4, -3, -2, -1, 0, +1, +2, +3, +4, +5, +6}, then the quantization factor K1 = 0.012; The basic domain of input variable 2 is (0, 1000kW). After normalization, the continuous variable of the basic domain (0, 1000kW) is converted into the continuous variable of the standard domain (-6, +6). The variables of the standard domain are then divided into 5 linguistic variables E2 {positive large PB, positive middle PM, zero ZE, negative middle NM, positive small NB}; The domain of the language variable E2 is further selected as: X2 = {-6, -5, -4, -3, -2, -1, 0, +1, +2, +3, +4, +5, +6}, then the quantization factor K2 = -0.012; The basic domain of output variable 1 is (-100kW, 0). After normalization, the continuous variable of the basic domain (-100kW, 0) is converted into the continuous variable of the standard domain (-6, +6). The variables of the standard domain are then divided into 5 linguistic variables E 3 {positive large PB, positive middle PM, zero ZE, negative middle NM, positive small NB}; The domain of the language variable E3 is further selected as: Y1 = {-6, -5, -4, -3, -2, -1, 0, +1, +2, +3, +4, +5, +6}, then the quantization factor K3 = 0.12; The basic domain of output variable 2 is (0, 100kW). After normalization, the continuous variable of the basic domain (0, 100kW) is converted into the continuous variable of the standard domain (-6, +6). The variable of the standard domain is then divided into 5 linguistic variables E4 {positive large PB, positive middle PM, zero ZE, negative middle NM, positive small NB}; The domain of the language variable E4 is further selected as: Y2 = {-6, -5, -4, -3, -2, -1, 0, +1, +2, +3, +4, +5, +6}, then the quantization factor K4 = 0.12; The basic domain of the output variable Y3 is (-300kW, 0). After normalization, the continuous variable of the basic domain (-300kW, 0) is converted into the continuous variable of the standard domain (-6, +6). The variables of the standard domain are then divided into 5 linguistic variables E5 {positive large PB, positive middle PM, zero ZE, negative middle NM, positive small NB}; The domain of the language variable E5 is further selected as: Y3 = {-6, -5, -4, -3, -2, -1, 0, +1, +2, +3, +4, +5, +6}, then the quantization factor K5 = 0.36; The positive value in the domain represents the emitted power, and the negative value represents the absorbed power.

10. The alkaline and proton exchange membrane composite electrolyzer based on renewable electricity according to claim 9, characterized in that: The specific method of designing fuzzy control rules in step 3 is as follows: Rule 1: If X1 is PB, X2 is PB, then Y1 is NB, Y2 is NB, and Y3 is NB; Rule 2: If X1 is PB and X2 is PM, then Y1 is NB, Y2 is NM, and Y3 is NB; Rule 3: If X1 is PB and X2 is ZE, then Y1 is NB, Y2 is ZE, and Y3 is NB; Rule 4: If X1 is PB, X2 is NM, then Y1 is NB, Y2 is PM, and Y3 is ZE; Rule 5: If X1 is PB and X2 is NB, then Y1 is NB, Y2 is PB, and Y3 is NB; Rule 6: If X1 is PM and X2 is PB, then Y1 is NB, Y2 is PM, and Y3 is NB; Rule 7: If X1 is PM, X2 is PM, then Y1 is NB, Y2 is PM, and Y3 is NB; Rule 8: If X1 is PM and X2 is ZE, then Y1 is NB, Y2 is PM, and Y3 is NB; Rule 9: If X1 is PM and X2 is NM, then Y1 is NB, Y2 is PM, and Y3 is ZE; Rule 10: If X1 is PM and X2 is NB, then Y1 is NB, Y2 is PM, and Y3 is NB; Rule 11: If X1 is ZE and X2 is PB, then Y1 is PB, Y2 is NB, and Y3 is PB; Rule 12: If X1 is ZE and X2 is PM, then Y1 is PB, Y2 is ZE, and Y3 is PM; Rule 13: If X1 is ZE, X2 is ZE, then Y1 is ZE, Y2 is ZE, and Y3 is ZE; Rule 14: If X1 is ZE and X2 is NM, then Y1 is NB, Y2 is PM, and Y3 is ZE; Rule 15: If X1 is ZE and X2 is NB, then Y1 is NB, Y2 is PB, and Y3 is NB; Rule 16: If X1 is NM and X2 is PB, then Y1 is PB, Y2 is NB, and Y3 is ZE; Rule 17: If X1 is NM and X2 is PM, then Y1 is PM, Y2 is NB, and Y3 is ZE; Rule 18: If X1 is NM and X2 is ZE, then Y1 is PM, Y2 is NB, and Y3 is ZE; Rule 19: If X1 is NM, X2 is NM, then Y1 is ZE, Y2 is ZE, and Y3 is ZE; Rule 20: If X1 is NM and X2 is NB, then Y1 is NB, Y2 is PM, and Y3 is NB; Rule 21: If X1 is NB and X2 is PB, then Y1 is PM, Y2 is ZE, and Y3 is PM; Rule 22: If X1 is NB and X2 is PM, then Y1 is PB, Y2 is NB, and Y3 is PM; Rule 23: If X1 is NB and X2 is ZE, then Y1 is PM, Y2 is NB, and Y3 is PM; Rule 24: If X1 is NB and X2 is NM, then Y1 is PM, Y2 is NB, and Y3 is ZE; Rule 25: If X1 is NB, X2 is NB, then Y1 is ZE, Y2 is ZE, and Y3 is NB.

Images