An optimization control method and system for a hybrid electrolyzed water hydrogen production system

Abstract

The present invention discloses a method and system for optimizing and controlling a hybrid water electrolysis hydrogen production system. The method adopted by the present invention includes: establishing a hybrid water electrolysis hydrogen production system including a wind turbine, photovoltaics, an alkaline electrolyzer, a proton exchange membrane electrolyzer, a solid oxide electrolyzer and an energy storage device; establishing a unified and universal mathematical model; considering the differences in dynamic response speeds of different types of electrolyzers, combining a low-pass filtering algorithm, and adopting a multi-stage controlled optimized control of the hybrid water electrolysis hydrogen production system; constructing an objective function with minimizing operating costs as the control target, considering the electricity purchase cost, energy storage use cost, wind and solar power abandonment cost, hydrogen sales income and electrolyzer start-up and shutdown cost, and using the branch and bound method to solve the control scheme of the hybrid water electrolysis hydrogen production system. The present invention realizes the coordinated control of multiple types of electrolyzers and the differentiated utilization of different types of electrolyzers, thereby improving the operating economy and energy utilization efficiency of the hybrid water electrolysis hydrogen production system.

Description

Technical Field

[0001] The present invention belongs to the technical field of optimization control of water electrolysis hydrogen production systems, and in particular to an optimization control method and system for a hybrid water electrolysis hydrogen production system that takes into account the differences in dynamic response speeds of different water electrolysis hydrogen production technologies. Background Art

[0002] The installed capacity of photovoltaic and wind turbines is constantly increasing. Wind and solar resources are discontinuous and unstable. Direct grid connection will have an impact on the power grid. Water electrolysis hydrogen production technology can convert and store fluctuating clean electricity such as wind and solar power into high-quality hydrogen energy, thereby improving energy utilization efficiency.

[0003] Currently, common electrolyzers include alkaline electrolysis cells (AECs), proton exchange membrane electrolysis cells (PEMECs), and solid oxide electrolysis cells (SOECs). AEC technology is the most mature and has the lowest cost, but hydrogen-oxygen diffusion occurs at low loads, resulting in a narrow operating load range and slow response speed. PEMEC technology significantly improves upon AEC in terms of load range and response speed, but is slightly more expensive. SOEC offers the highest energy conversion efficiency, but also has the highest cost and the slowest response speed.

[0004] Existing technologies have studied water electrolysis hydrogen production systems. On the one hand, most electrolyzer models are usually represented by a fixed hydrogen-to-electricity conversion coefficient. Few studies have considered changes in electrolysis efficiency with electrolyzer operating conditions, and the constraints on electrolyzer operation are not considered in detail and in depth. On the other hand, most studies only consider the application of one water electrolysis technology, and few studies have comprehensively considered the differences in operating characteristics of different electrolyzers, and loaded renewable energy with different fluctuation characteristics according to their dynamic response speed differences, to achieve differentiated utilization of AEC, PEMEC and SOEC.

[0005] Therefore, it is necessary to propose an optimization control method and system for a hybrid water electrolysis hydrogen production system that takes into account the differences in operating characteristics of different electrolyzers. Summary of the Invention

[0006] In order to solve the problems existing in the above-mentioned prior art, the present invention proposes an optimization control method and system for a hybrid water electrolysis hydrogen production system to achieve coordinated control and differentiated utilization of multiple types of electrolyzers.

[0007] To this end, a technical solution adopted by the present invention is: a hybrid water electrolysis hydrogen production system optimization control method, which includes the steps of:

[0008] 1) Establish a hybrid water electrolysis hydrogen production system including wind turbines, photovoltaics, alkaline electrolyzers, proton exchange membrane electrolyzers, solid oxide electrolyzers and batteries;

[0009] 2) Establish a refined, unified, and universal mathematical model for alkaline electrolyzers, proton exchange membrane electrolyzers, and solid oxide electrolyzers;

[0010] 3) Considering the differences in dynamic response speeds of different types of electrolyzers, a multi-stage control hybrid water electrolysis hydrogen production system is optimized by combining a low-pass filtering algorithm;

[0011] 4) Construct an objective function with the minimization of operating costs as the control goal, taking into account the electricity purchase cost, energy storage usage cost, wind and solar power curtailment cost, hydrogen sales revenue and electrolyzer start-up and shutdown costs, and use the branch and bound method to solve the control scheme of the hybrid water electrolysis hydrogen production system.

[0012] The hybrid water electrolysis hydrogen production system of the present invention takes into account different types of electrolyzers and comprehensively utilizes the differences in operating characteristics of different types of electrolyzers to better absorb renewable energy.

[0013] Another technical solution adopted by the present invention is: an optimization control system for a hybrid water electrolysis hydrogen production system, which includes:

[0014] Hybrid water electrolysis hydrogen production system establishment unit: Establish a hybrid water electrolysis hydrogen production system including wind turbines, photovoltaics, alkaline electrolyzers, proton exchange membrane electrolyzers, solid oxide electrolyzers and batteries;

[0015] Unified and universal mathematical model building unit: Establish refined unified and universal mathematical models for alkaline electrolyzers, proton exchange membrane electrolyzers, and solid oxide electrolyzers;

[0016] Optimization control unit: Considering the differences in dynamic response speeds of different types of electrolyzers, combined with a low-pass filtering algorithm, a multi-stage control hybrid water electrolysis hydrogen production system is optimized;

[0017] Control scheme solving unit: Constructs an objective function with minimizing operating costs as the control goal, taking into account the electricity purchase cost, energy storage usage cost, wind and solar power curtailment cost, hydrogen sales revenue and electrolyzer start-up and shutdown costs, and uses the branch and bound method to solve the control scheme of the hybrid water electrolysis hydrogen production system.

[0018] The present invention comprehensively considers the operating characteristics of different electrolyzers, coordinates the output of three different electrolyzers, and makes them carry net loads with different fluctuation characteristics according to the differences in dynamic response speed. Ultimately, it achieves the optimized control goals of maximizing the absorption of wind turbine and photovoltaic output, minimizing operating costs, and differentially utilizing AEC, PEMEC, and SOEC, thereby improving the operating economy and energy utilization efficiency of the hybrid water electrolysis hydrogen production system. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some implementation cases of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0020] Figure 1 Schematic diagram of the structure of the hybrid water electrolysis hydrogen production system of the present invention;

[0021] Figure 2 This is a schematic diagram of the multi-stage optimization control principle of the hybrid water electrolysis hydrogen production system of the present invention;

[0022] Figure 3 This is a flow chart of the multi-stage optimization control method for the hybrid water electrolysis hydrogen production system of the present invention;

[0023] Figure 4 This is the day-ahead control decision result diagram of the SOEC of the present invention;

[0024] Figure 5 This is the intraday control decision result diagram of the AEC of the present invention;

[0025] Figure 6 This is a real-time control decision result diagram of the PEMEC of the present invention;

[0026] Figure 7 This is a control decision comparison result diagram of the AEC electrolyzer of the present invention;

[0027] Figure 8 This is a control decision comparison result diagram of the PEMEC electrolyzer of the present invention;

[0028] Figure 9 This is a control decision comparison result diagram of the SOEC electrolyzer of the present invention. DETAILED DESCRIPTION

[0029] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without creative work are within the scope of protection of the present invention.

[0030] Example 1

[0031] This embodiment provides an optimization control method for a hybrid water electrolysis hydrogen production system, the specific contents of which are as follows:

[0032] 1) Establish a hybrid water electrolysis hydrogen production system including wind turbines, photovoltaics, alkaline electrolyzers, proton exchange membrane electrolyzers, solid oxide electrolyzers and batteries, taking into account different types of electrolyzers and comprehensively utilizing the differences in operating characteristics of different types of electrolyzers to better absorb renewable energy.

[0033] 2) Establish a refined unified and universal mathematical model for alkaline electrolyzers, proton exchange membrane electrolyzers, and solid oxide electrolyzers. Different from other mathematical models, the refined unified and universal mathematical model established in the present invention includes an electrolyzer start-stop model, an electrolyzer output model, an electrolyzer power model, and an electrolyzer temperature model. The technical indicators considered mainly include start-up delay, start-stop capability, output characteristics, working range, starting power, climbing capability, and temperature constraints.

[0034] 3) Considering the differences in dynamic response speeds of different types of electrolyzers, combined with a low-pass filtering algorithm, a multi-stage control hybrid water electrolysis hydrogen production system optimization control is adopted. The optimization control of different types of electrolyzers is divided into three stages: day-ahead, intraday, and real-time. Differentiated optimization control is performed according to the operating characteristics of different types of electrolyzers.

[0035] 4) Construct an objective function with minimizing operating costs as the control goal, taking into account the electricity purchase cost, energy storage usage cost, wind and solar power curtailment cost, hydrogen sales revenue and electrolyzer start-up and shutdown costs, taking into account the wind turbine output constraints, photovoltaic output constraints, battery operation constraints and electrolyzer operation constraints, and use the branch and bound method to solve the control scheme of the mixed water electrolysis hydrogen production system.

[0036] In the step 1), the hybrid water electrolysis hydrogen production system is established as shown in the attached Figure 1 As shown;

[0037] In step 2), the refined unified universal electrolytic cell mathematical model of the alkaline electrolyzer, proton exchange membrane electrolyzer, and solid oxide electrolyzer is as follows:

[0038] In the following text, the superscript M of all symbols represents different electrolyzers, M{A,P,S}, where A represents alkaline electrolyzer (AEC), P represents proton exchange membrane electrolyzer (PEMEC), and S represents solid oxide electrolyzer (SOEC); the subscript k is the electrolyzer number; the subscript t represents the unit operating time period, and T is the total operating time period, where 1≤t≤T.

[0039] (1) Electrolyzer start-stop model

[0040] The electrolytic cell start-stop model considering the start-up delay is as follows:

[0041]

[0042]

[0043]

[0044] Among them: 0-1 variable Indicates the on / off status of the electrolytic cell. Indicates the start-up action of the electrolytic cell. Indicates the start and stop of the electrolytic cell; α M Indicates the startup delay;

[0045] Constraints on the number of electrolytic cell starts and stops:

[0046]

[0047]

[0048] in: They represent the upper limit of the number of startup and shutdown times of M electrolytic cell in a day.

[0049] (2) Electrolyzer output model

[0050]

[0051]

[0052]

[0053] in: is the working efficiency of the electrolytic cell; U HHV =1.48V is a constant; is the working current of the electrolytic cell; is the working power of the electrolytic cell; The energy corresponding to the chemical energy in hydrogen represents the mass of hydrogen produced, in kg; γ = 7.2 is the conversion factor from mol / s to kg / h of hydrogen produced per unit time period; F = 96485 C / mol is the Faraday constant;

[0054] With working power The relationship is approximately linearized as follows:

[0055]

[0056]

[0057] Where: a and b are the linearization parameters of the electrolyzer output model; is the electrolytic cell temperature; T M,max and T M,min are the upper and lower limits of the electrolytic cell temperature respectively; is the rated power of the electrolyzer; is the thermal power generated by the electrolyzer;

[0058] (3) Electrolyzer power model

[0059] Electrolyzer operating upper and lower power constraints:

[0060]

[0061]

[0062] Where: P M，min / P M，max Respectively represent the upper / lower limit of the working power of M electrolyzer in the power-on state; P M,boot Indicates the electric power consumed during the startup of the electrolytic cell.

[0063] Electrolyzer ramp power constraint:

[0064]

[0065] Where: ΔP M，max Indicates the maximum ramp power per unit time period when the M electrolyzer is in the on state, P M，max is the upper limit of the working power of the electrolyzer;

[0066] (4) Electrolytic cell temperature model

[0067]

[0068]

[0069]

[0070] Where: T a is the ambient temperature; C e is the lumped heat capacity of the electrolytic cell, which can be measured by heating experiments; R e is the lumped thermal resistance; is the lost heat power; is the heat power output outside the system; is the electrolytic cell temperature; T M,max and T M,min They are the upper and lower limits of electrolytic cell power respectively.

[0071] In step 3), the low-pass filtering algorithm is as follows:

[0072] Q t =nM t +(1-n)Q t-1 (17)

[0073] Where: Mt is the original data input at time t; n is the filter coefficient; Q t is the filtered low-frequency data output at time t.

[0074] The SOEC optimization control process in the day-ahead stage is as follows:

[0075] 1) Input the wind turbine, photovoltaic, and load forecast data for the previous 24 hours. Subtract the predicted load power from the sum of the wind turbine and photovoltaic forecast output to obtain the net load power for the previous day:

[0076]

[0077] in: is the day-ahead net load power; and They are the wind turbine, photovoltaic and load forecast data for the day before;

[0078] 2) Perform the first low-pass filtering on the net load:

[0079]

[0080] Where: n is the filter coefficient, is the day-ahead low-frequency net load at time t;

[0081] 3) Under the condition that SOEC operation constraints are met, SOEC is used to absorb the low-frequency net load after the first low-pass filtering:

[0082] Constraints:

[0083] Power balance constraints:

[0084]

[0085] Where: P t S is the electric power actually consumed by SOEC during period t; is the difference between the actual electric power consumed by SOEC and the day-ahead low-frequency net load;

[0086] Other constraints: electrolyzer operation constraints, i.e., equations (1)-(16);

[0087] Optimization goal:

[0088] F riqian =min(f1-f2+f3) (21)

[0089]

[0090]

[0091]

[0092] Among them: F riqian is the objective function of the day-ahead phase, f1 is the cost of starting and stopping SOEC, C S,boot and C S,shut are the startup cost and shutdown cost of SOEC electrolyzer respectively; and are the state variables of SOEC startup and shutdown actions respectively; f2 is the income from selling hydrogen produced by SOEC; The price of hydrogen sold, in RMB / kg; is the mass of hydrogen produced by SOEC during period t; f3 is the sum of the absolute values of the difference between the electric power actually consumed by SOEC and the low-frequency net load on the previous day.

[0093] 4) Use the branch and bound method to solve the day-ahead control scheme of SOEC.

[0094] The intraday AEC optimization control process is as follows:

[0095] 1) Input the wind turbine, photovoltaic and load forecast data for the next 8 hours. Use the new forecast data to calculate the new net load power and then subtract the output of SOEC to obtain the net load power within the day:

[0096]

[0097] in: is the net load power during the day; and are the wind turbine, photovoltaic and load forecast data for the day; P t S is the actual electric power consumed by SOEC obtained in the day-ahead phase;

[0098] 2) Low-pass filter the net load power within the day again:

[0099]

[0100] Where: n is the filter coefficient, is the intraday low-frequency net load at time t;

[0101] 3) Under the condition of meeting the AEC operation constraints, use AEC to absorb the intraday low-frequency net load after low-pass filtering:

[0102] Constraints:

[0103] Power balance constraints:

[0104]

[0105] Where: P t Ais the electric power actually consumed by AEC during period t; The difference between the actual electric power consumed by the AEC and the net low-frequency load during the day;

[0106] Other constraints: electrolyzer operation constraints, i.e., equations (1)-(16);

[0107] Optimization goal:

[0108] F rinei =min(f4-f5+f6) (28)

[0109]

[0110]

[0111]

[0112] Among them: F rinei is the intraday objective function, f4 is the AEC start-stop cost, C A,boot and C A,shut are the startup cost and shutdown cost of the AEC electrolyzer, respectively; and are the state variables of the AEC startup and shutdown actions respectively; f5 is the income from selling the hydrogen generated by the AEC; The price of hydrogen sold, in RMB / kg; is the mass of hydrogen produced by AEC during period t; f6 is the sum of the absolute values of the difference between the actual electric power consumed by AEC and the net low-frequency load during the day.

[0113] 4) Use the branch and bound method to solve the intraday control solution of AEC.

[0114] 5) Continuously update the forecast data for the next 8 hours and perform the above optimization process until the AEC's 24-hour control plan is obtained.

[0115] The real-time stage PEMEC optimization control process is as follows:

[0116] 1) Input the wind turbine, photovoltaic and load forecast data for the next 4 hours, use the new forecast data to calculate the new net load power, and then subtract the output of SOEC and AEC to obtain the real-time net load power:

[0117]

[0118] in: is the real-time net load power; and are the wind turbine, photovoltaic and load forecast data for the day; P t SP is the actual electric power consumed by SOEC obtained in the day-ahead stage; t A is the actual electric power consumed by the AEC calculated during the day;

[0119] 2) Using PEMEC to absorb real-time net load while meeting PEMEC operation constraints:

[0120] Constraints:

[0121] Wind turbine photovoltaic output constraints:

[0122]

[0123] Power balance constraints:

[0124]

[0125] Purchased power constraints:

[0126]

[0127] Where: P t P is the actual electric power consumed by PEMEC obtained during the day-ahead period; and are charging and discharging power respectively; P PV,t and P WT,t are the actual outputs of photovoltaic and wind turbines respectively; P e,t The power of purchased electricity; The upper limit of the power purchased;

[0128] Battery operation constraints:

[0129]

[0130] in: and They are energy storage charging and discharging states respectively; and are the upper limits of energy storage charge and discharge; η ch and η dis are the battery charging efficiency and discharging efficiency respectively; E 0 and E T They are the initial and final energy storage capacity, which need to be balanced; E t is the energy storage capacity at time t, E max and E min They are the upper and lower limits of energy storage capacity respectively.

[0131] Other constraints: electrolyzer operation constraints, i.e., equations (1)-(16);

[0132] Optimization goal:

[0133] F shishi =min(f7-f8+f9+f 10 +f 11 ) (37)

[0134]

[0135]

[0136]

[0137]

[0138]

[0139] Among them: F shishi is the real-time objective function, f7 is the PEMEC start-stop cost, C P,boot and C P,shut are the startup cost and shutdown cost of the PEMEC electrolyzer, respectively; and are the state variables of the PEMEC startup and shutdown actions respectively; f8 is the income from selling the hydrogen produced by PEMEC; The price of hydrogen sold, in RMB / kg; is the mass of hydrogen produced by PEMEC during period t; f9 is the cost of electricity purchase; is the electricity purchase price; f 10 The penalty cost for curtailing wind and solar power; α1 and β are the penalty coefficients for curtailing solar power and wind power, respectively; f 11 Energy storage usage cost; ch and λ dis is the charging and discharging cost coefficient;

[0140] 3) Use the branch and bound method to solve the real-time control solution of PEMEC.

[0141] 4) Continuously update the forecast data for the next 4 hours and perform the above optimization process until the real-time 24-hour control solution of PEMEC is obtained.

[0142] In this embodiment, the structure of the hybrid water electrolysis hydrogen production system is as follows: Figure 1As shown, it includes a wind turbine, photovoltaics, an alkaline electrolyzer, a proton exchange membrane electrolyzer, a solid oxide electrolyzer, an energy storage device and a load. The electrical energy is converted into hydrogen by the electrolyzer to improve energy utilization efficiency. In order to obtain the optimal control scheme for hybrid hydrogen production, the present invention first establishes a refined unified universal mathematical model of the electrolyzer, and then combines the dynamic response speed differences of different types of electrolyzers and the low-pass filtering algorithm to propose an optimization control method for a hybrid water electrolysis hydrogen production system including three stages: day-ahead, intraday and real-time. The operation control principle is as follows Figure 2 The specific optimization process is as shown in Figure 3 shown.

[0143] Day-ahead stage: Optimize the SOEC control. First, input the wind turbine, photovoltaic, and load forecast data for the 24 hours before the day. Subtract the forecast load power from the sum of the wind turbine and photovoltaic forecast output to obtain the day-ahead net load power. Then, perform a first low-pass filter on the net load. Subject to the SOEC's operating constraints, the SOEC is used to absorb the day-ahead low-frequency net load after the first low-pass filter. Finally, the branch-and-bound method is used to solve the SOEC's day-ahead control scheme.

[0144] Intraday stage: Optimize the AEC control by first inputting the forecast data for wind turbines, photovoltaics, and loads for the next 8 hours. The new forecast data is used to calculate the new net load power, which is then subtracted from the SOEC output to obtain the intraday net load power. The intraday net load power is then low-pass filtered again. Subject to the AEC operating constraints, the AEC is used to absorb the intraday low-frequency net load after low-pass filtering. Finally, the branch-and-bound method is used to solve the AEC's intraday control plan. The forecast data for the next 8 hours is continuously updated, and the above optimization process is repeated until the AEC's intraday 24-hour control plan is determined.

[0145] Real-time stage: Optimize the control of PEMEC. First, input the wind turbine, photovoltaic, and load forecast data for the next four hours. Use the new forecast data to calculate the new net load power, and then subtract the output of the SOEC and AEC to obtain the real-time net load power. Then, under the condition that the PEMEC operating constraints are met, the PEMEC is used to absorb the real-time net load. Finally, the branch and bound method is used to solve the real-time control plan of PEMEC. The forecast data for the next four hours is continuously updated and the above optimization process is repeated until the real-time 24-hour control plan of PEMEC is obtained.

[0146] Figure 4 、 Figure 5 and Figure 6 are the operation control results of SOEC, AEC and PEMEC respectively, Figure 7-9The comparison of the three results shows that the SOEC is primarily responsible for absorbing the portion of the net load with the smallest fluctuations, the AEC is responsible for absorbing a portion of the net load with fluctuations, but also primarily the relatively stable portion. Due to its fast response and quick start-stop speeds, the PEMEC is responsible for absorbing the portion of the net load with the highest frequency fluctuations. The optimized control method for the hybrid water electrolysis hydrogen production system proposed in this invention achieves differentiated utilization of different types of electrolyzers.

[0147] Example 2

[0148] This embodiment provides an optimization control system for a hybrid water electrolysis hydrogen production system, which consists of a hybrid water electrolysis hydrogen production system establishment unit, a unified universal mathematical model establishment unit, an optimization control unit, and a control scheme solving unit.

[0149] Hybrid water electrolysis hydrogen production system establishment unit: Establish a hybrid water electrolysis hydrogen production system including wind turbines, photovoltaics, alkaline electrolyzers, proton exchange membrane electrolyzers, solid oxide electrolyzers and batteries.

[0150] Unified and universal mathematical model establishment unit: establish refined unified and universal mathematical models for alkaline electrolyzers, proton exchange membrane electrolyzers and solid oxide electrolyzers.

[0151] Optimization control unit: Taking into account the differences in dynamic response speeds of different types of electrolyzers, combined with a low-pass filtering algorithm, a multi-stage controlled hybrid water electrolysis hydrogen production system is optimized.

[0152] Control scheme solving unit: Constructs an objective function with minimizing operating costs as the control goal, taking into account the electricity purchase cost, energy storage usage cost, wind and solar power curtailment cost, hydrogen sales revenue and electrolyzer start-up and shutdown costs, and uses the branch and bound method to solve the control scheme of the hybrid water electrolysis hydrogen production system.

[0153] Specifically, the unified and universal mathematical model includes an electrolytic cell start-stop model, an electrolytic cell output model, an electrolytic cell power model, and an electrolytic cell temperature model. The technical indicators considered include start-up delay, start-stop capability, output characteristics, operating range, starting power, climbing capability, and temperature constraints.

[0154] More specifically, the unified and universal mathematical model is as follows:

[0155] In the following text, the superscript M in all symbols represents different electrolytic cells, M{A,P,S}, where A represents alkaline electrolytic cell, P represents proton exchange membrane electrolytic cell, and S represents solid oxide electrolytic cell; the subscript k is the electrolytic cell number; the subscript t represents the unit operating time period, and T is the total operating time period, where 1≤t≤T;

[0156] (1) Electrolyzer start-stop model

[0157] The electrolytic cell start-stop model considering the start-up delay is as follows:

[0158]

[0159]

[0160]

[0161] Among them: 0-1 variable Indicates the on / off status of the electrolytic cell. Indicates the start-up action of the electrolytic cell. Indicates the start and stop of the electrolytic cell; α M Indicates the startup delay; Indicates the switch state of the electrolytic cell at time t-1; represents t-α M The moment the electrolytic cell starts to start;

[0162] Constraints on the number of electrolytic cell starts and stops:

[0163]

[0164]

[0165] in: They represent the upper limit of the number of startup and shutdown times of M electrolytic cell per day;

[0166] (2) Electrolyzer output model

[0167]

[0168]

[0169]

[0170] in: is the working efficiency of the electrolytic cell; U HHV =1.48V is a constant; is the working current of the electrolytic cell; is the working power of the electrolytic cell; The energy corresponding to the chemical energy in hydrogen is converted; represents the mass of hydrogen produced, in kg; γ = 7.2 is the conversion factor from mol / s to kg / h of hydrogen produced per unit time period; F = 96485 C / mol is the Faraday constant;

[0171] Working power of electrolyzer The relationship is approximately linearized as follows:

[0172]

[0173]

[0174] Where: a and b are linearized parameters of the electrolytic cell output model; is the electrolytic cell temperature; T M,max and T M,min are the upper and lower limits of the electrolytic cell temperature respectively; is the rated power of the electrolyzer; is the thermal power generated by the electrolyzer;

[0175] (3) Electrolyzer power model

[0176] The upper and lower power limits of the electrolytic cell are:

[0177]

[0178]

[0179] Where: P M，min / P M，max Respectively represent the upper / lower limit of the working power of M electrolyzer in the power-on state; P M,boot Indicates the electric power consumed during the startup of the electrolytic cell; τ indicates the time unit from the start-up of the electrolytic cell to the time when the electrolytic cell enters the working state; Indicates the start-up action of the electrolytic cell at time t-τ;

[0180] Electrolyzer ramp power constraint:

[0181]

[0182] Where: ΔP M，max Indicates the maximum ramp power per unit time period of M electrolyzer when it is on;

[0183] (4) Electrolytic cell temperature model

[0184]

[0185]

[0186]

[0187] Where: T a is the ambient temperature; C e is the lumped heat capacity of the electrolytic cell; R e is the lumped thermal resistance; is the lost heat power; is the thermal power output outside the system; Δt represents unit time.

[0188] Specifically, the hybrid water electrolysis hydrogen production system optimization control takes into account wind turbine output constraints, photovoltaic output constraints, battery operation constraints and electrolyzer operation constraints.

[0189] Specifically, the hybrid water electrolysis hydrogen production system optimization control divides the optimization control of different types of electrolyzers into three stages: day-ahead, intraday, and real-time, and performs differentiated optimization control based on the operating characteristics of different types of electrolyzers.

[0190] Specifically, the low-pass filtering algorithm is as follows:

[0191] Q t =nM t +(1-n)Q t-1 (17)

[0192] Where: M t is the original data input at time t; n is the filter coefficient; Q t is the filtered low-frequency data output at time t.

[0193] Specifically, the SOEC optimization control process in the day-ahead stage is as follows:

[0194] 1) Input the wind turbine, photovoltaic, and load forecast data for the previous 24 hours. Subtract the predicted load power from the sum of the wind turbine and photovoltaic forecast output to obtain the net load power for the previous day:

[0195]

[0196] in: is the day-ahead net load power; and They are the wind turbine, photovoltaic and load forecast data for the day before;

[0197] 2) Perform the first low-pass filtering on the net load:

[0198]

[0199] Where: n is the filter coefficient, is the day-ahead low-frequency net load at time t;

[0200] 3) Under the condition that SOEC operation constraints are met, SOEC is used to absorb the low-frequency net load after the first low-pass filtering:

[0201] Constraints:

[0202] Power balance constraints:

[0203]

[0204] Where: P t Sis the electric power actually consumed by SOEC during period t; is the difference between the actual electric power consumed by SOEC during period t and the low-frequency net load on the day before;

[0205] Other constraints: electrolyzer operation constraints, i.e., equations (1)-(16);

[0206] Optimization goal:

[0207] F riqian =min(f1-f2+f3) (21)

[0208]

[0209]

[0210]

[0211] Among them: F riqian is the objective function of the day-ahead phase, f1 is the cost of starting and stopping SOEC, C S,boot and C S,shut are the startup cost and shutdown cost of SOEC electrolyzer respectively; and are the state variables of SOEC startup and shutdown actions respectively; f2 is the income from selling hydrogen produced by SOEC; The price of hydrogen sold, in RMB / kg; is the mass of hydrogen produced by SOEC during period t; f3 is the sum of the absolute values of the difference between the actual electric power consumed by SOEC and the low-frequency net load on the previous day;

[0212] 4) Use the branch and bound method to solve the day-ahead control scheme of SOEC.

[0213] Specifically, the intraday AEC optimization control process is as follows:

[0214] 1) Input the wind turbine, photovoltaic and load forecast data for the next 8 hours. Use the new forecast data to calculate the new net load power and then subtract the output of SOEC to obtain the net load power within the day:

[0215]

[0216] in: is the net load power during the day; and are the wind turbine, photovoltaic and load forecast data for the day; P t S is the actual electric power consumed by SOEC obtained in the day-ahead phase;

[0217] 2) Low-pass filter the net load power within the day again:

[0218]

[0219] Where: n is the filter coefficient, is the intraday low-frequency net load at time t;

[0220] 3) Under the condition of meeting the AEC operation constraints, use AEC to absorb the intraday low-frequency net load after low-pass filtering:

[0221] Constraints:

[0222] Power balance constraints:

[0223]

[0224] Where: P t A is the electric power actually consumed by AEC during period t; The difference between the actual electric power consumed by the AEC and the net low-frequency load during the day;

[0225] Other constraints: electrolyzer operation constraints, i.e., equations (1)-(16);

[0226] Optimization goal:

[0227] F rinei =min(f4-f5+f6) (28)

[0228]

[0229]

[0230]

[0231] Among them: F rinei is the intraday objective function, f4 is the AEC start-stop cost, C A,boot and C A,shut are the startup cost and shutdown cost of the AEC electrolyzer, respectively; and are the state variables of the AEC startup and shutdown actions respectively; f5 is the income from selling the hydrogen generated by the AEC; The price of hydrogen sold, in RMB / kg; is the mass of hydrogen produced by the AEC during period t; f6 is the sum of the absolute values of the difference between the actual electric power consumed by the AEC and the net low-frequency load during the day;

[0232] 4) Use branch and bound method to solve the intraday control scheme of AEC;

[0233] 5) Continuously update the forecast data for the next 8 hours and perform the above optimization process until the AEC's 24-hour control plan is obtained.

[0234] Specifically, the real-time PEMEC optimization control process is as follows:

[0235] 1) Input the wind turbine, photovoltaic and load forecast data for the next 4 hours, use the new forecast data to calculate the new net load power, and then subtract the output of SOEC and AEC to obtain the real-time net load power:

[0236]

[0237] in: is the real-time net load power; and are the wind turbine, photovoltaic and load forecast data for the day; P t S P is the actual electric power consumed by SOEC obtained in the day-ahead stage; t A is the actual electric power consumed by the AEC calculated during the day;

[0238] 2) Using PEMEC to absorb real-time net load while meeting PEMEC operation constraints:

[0239] Constraints:

[0240] Wind turbine photovoltaic output constraints:

[0241]

[0242] Power balance constraints:

[0243]

[0244] Purchased power constraints:

[0245]

[0246] Where: P t P is the actual electric power consumed by PEMEC obtained during the day-ahead period; and are charging and discharging power respectively; P PV,t and P WT,t are the actual outputs of photovoltaic and wind turbines respectively; P e,t The power of purchased electricity; The upper limit of the power purchased;

[0247] Battery operation constraints:

[0248]

[0249] in: and They are energy storage charging and discharging states respectively; and are the upper limits of energy storage charge and discharge respectively; η ch and η dis are the battery charging efficiency and discharging efficiency respectively; E 0 and E T They are the initial and final energy storage capacity, which need to be balanced; E t is the energy storage capacity at time t, E max and E min They are the upper and lower limits of energy storage capacity respectively;

[0250] Other constraints: electrolyzer operation constraints, i.e., equations (1)-(16);

[0251] Optimization goal:

[0252] F shishi =min(f7-f8+f9+f 10 +f 11 ) (37)

[0253]

[0254]

[0255]

[0256]

[0257]

[0258] Among them: F shishi is the real-time objective function, f7 is the PEMEC start-stop cost, C P,boot and C P,shut are the startup cost and shutdown cost of the PEMEC electrolyzer, respectively; and are the state variables of the PEMEC startup and shutdown actions respectively; f8 is the income from selling the hydrogen produced by PEMEC; The price of hydrogen sold, in RMB / kg; is the mass of hydrogen produced by PEMEC during period t; f9 is the cost of electricity purchase; is the electricity purchase price; f 10 The penalty cost for curtailing wind and solar power; α1 and β are the penalty coefficients for curtailing solar power and wind power, respectively; f 11 Energy storage usage cost; ch and λ dis is the charging and discharging cost coefficient;

[0259] 3) Use branch and bound method to solve the real-time control scheme of PEMEC;

[0260] 4) Continuously update the forecast data for the next 4 hours and perform the above optimization process until the real-time 24-hour control solution of PEMEC is obtained.

[0261] Although the above describes the specific embodiments of the present invention in conjunction with the accompanying drawings, it is not intended to limit the scope of protection of the present invention. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art on the basis of the technical solution of the present invention without any creative work are still within the scope of protection of the present invention.

Claims

1. A hybrid water electrolysis hydrogen production system optimization control method, characterized in that: Including steps: 1) Establish a hybrid water electrolysis hydrogen production system including wind turbines, photovoltaics, alkaline electrolyzers, proton exchange membrane electrolyzers, solid oxide electrolyzers and batteries; 2) Establish a refined, unified, and universal mathematical model for alkaline electrolyzers, proton exchange membrane electrolyzers, and solid oxide electrolyzers; 3) Considering the differences in dynamic response speeds of different types of electrolyzers, a multi-stage control hybrid water electrolysis hydrogen production system is optimized by combining a low-pass filtering algorithm; 4) Construct an objective function with the goal of minimizing operating costs, taking into account the cost of electricity purchase, energy storage usage, wind and solar power curtailment costs, hydrogen sales revenue, and electrolyzer start-up and shutdown costs, and use the branch and bound method to solve the control scheme of the hybrid water electrolysis hydrogen production system; The unified and universal mathematical model includes an electrolytic cell start-stop model, an electrolytic cell output model, an electrolytic cell power model, and an electrolytic cell temperature model. The technical indicators considered include start-up delay, start-stop capability, output characteristics, operating range, starting power, climbing capability, and temperature constraints. In step 2), the unified and universal mathematical model is as follows: In the following text, the superscript M in all symbols represents different electrolytic cells, M{A,P,S}, where A represents alkaline electrolytic cell, P represents proton exchange membrane electrolytic cell, and S represents solid oxide electrolytic cell; the subscript k is the electrolytic cell number; the subscript t represents the unit operating time period, and T is the total operating time period, where 1≤t≤T; (1) Electrolyzer start-stop model The electrolytic cell start-stop model considering the start-up delay is as follows: Among them: 0-1 variable Indicates the switch status of the electrolytic cell, Indicates the start-up action of the electrolytic cell. Indicates the start and stop of the electrolytic cell; α M Indicates the startup delay; Indicates the switch state of the electrolytic cell at time t-1; represents t-α M The moment the electrolytic cell starts to start; Constraints on the number of starts and stops of the electrolyzer: in: They represent the upper limit of the number of startup and shutdown times of M electrolytic cell per day; (2) Electrolyzer output model in: is the working efficiency of the electrolytic cell; U HHV =1.48V is a constant; is the working current of the electrolytic cell; is the working power of the electrolytic cell; The energy corresponding to the chemical energy in hydrogen is converted; represents the mass of hydrogen produced, in kg; γ = 7.2 is the conversion factor from mol / s to kg / h of hydrogen produced per unit time period; F = 96485 C / mol is the Faraday constant; Working power of electrolyzer The relationship is approximately linearized as follows: Where: a and b are the linearization parameters of the electrolyzer output model; is the electrolytic cell temperature; T M,max and T M,min are the upper and lower limits of the electrolytic cell temperature respectively; is the rated power of the electrolyzer; is the thermal power generated by the electrolyzer; (3) Electrolyzer power model The upper and lower power limits of the electrolytic cell are: Where: P M，min / P M，max Respectively represent the upper / lower limit of the working power of M electrolyzer in the power-on state; P M,boot Indicates the electric power consumed during the startup of the electrolytic cell; τ indicates the time unit from the start-up of the electrolytic cell to the time when the electrolytic cell enters the working state; Indicates the start-up action of the electrolytic cell at time t-τ; Electrolyzer ramp power constraint: Where: ΔP M，max Indicates the maximum ramp power per unit time period of M electrolyzer when it is on; (4) Electrolytic cell temperature model Where: T a is the ambient temperature; C e is the lumped heat capacity of the electrolytic cell; R e is the lumped thermal resistance; is the lost heat power; is the thermal power output outside the system; Δt represents unit time; The hybrid water electrolysis hydrogen production system optimization control takes into account wind turbine output constraints, photovoltaic output constraints, battery operation constraints and electrolyzer operation constraints; the hybrid water electrolysis hydrogen production system optimization control divides the optimization control of different types of electrolyzers into three stages: day-ahead, intraday and real-time, and performs differentiated optimization control based on the operating characteristics of different types of electrolyzers.

2. The optimization control method for a hybrid water electrolysis hydrogen production system according to claim 1, wherein: In step 3), the low-pass filtering algorithm is as follows: Q t =nM t +(1-n)Q t-1 (17) Where: M t is the original data input at time t; n is the filter coefficient; Q t is the filtered low-frequency data output at time t.

3. The optimization control method for a hybrid water electrolysis hydrogen production system according to claim 1, wherein: The SOEC optimization control process in the day-ahead stage is as follows: 1) Input the wind turbine, photovoltaic, and load forecast data for the previous 24 hours. Subtract the predicted load power from the sum of the wind turbine and photovoltaic forecast output to obtain the net load power for the previous day: in: is the day-ahead net load power; and They are the wind turbine, photovoltaic and load forecast data for the day before; 2) Perform the first low-pass filtering on the net load: Where: n is the filter coefficient, is the day-ahead low-frequency net load at time t; 3) Under the condition that SOEC operation constraints are met, SOEC is used to absorb the low-frequency net load after the first low-pass filtering: Constraints: Power balance constraints: Where: P t S is the electric power actually consumed by SOEC during period t; is the difference between the actual electric power consumed by SOEC during period t and the low-frequency net load on the day before; Other constraints: electrolyzer operation constraints, i.e., equations (1)-(16); Optimization goal: <h2 style=";text-align:left;direction:ltr">F<h2 style=";text-align:left;direction:ltr"> riqian <h2 style=";text-align:left;direction:ltr"> = min(f1-f2+f3) (21) Among them: F riqian is the objective function of the day-ahead phase, f1 is the cost of starting and stopping SOEC, C S,boot and C S,shut are the startup cost and shutdown cost of SOEC electrolyzer respectively; and are the state variables of SOEC startup and shutdown actions respectively; f2 is the income from selling hydrogen produced by SOEC; The price of hydrogen sold, in RMB / kg; is the mass of hydrogen produced by SOEC during period t; f3 is the sum of the absolute values of the difference between the actual electric power consumed by SOEC and the low-frequency net load on the previous day; 4) Use the branch and bound method to solve the day-ahead control scheme of SOEC.

4. The optimization control method for a hybrid water electrolysis hydrogen production system according to claim 1, wherein: The AEC optimization control process for the intraday stage is as follows: 1) Input the wind turbine, photovoltaic and load forecast data for the next 8 hours. Use the new forecast data to calculate the new net load power and then subtract the output of SOEC to obtain the net load power within the day: in: is the net load power during the day; and are the wind turbine, photovoltaic and load forecast data for the day; P t S is the actual electric power consumed by SOEC obtained in the day-ahead phase; 2) Low-pass filter the net load power within the day again: Where: n is the filter coefficient, is the intraday low-frequency net load at time t; 3) Under the condition of meeting the AEC operation constraints, use AEC to absorb the intraday low-frequency net load after low-pass filtering: Constraints: Power balance constraints: Where: P t A is the electric power actually consumed by AEC during period t; The difference between the actual electric power consumed by the AEC and the net low-frequency load during the day; Other constraints: electrolyzer operation constraints, i.e., equations (1)-(16); Optimization goal: F rinei =min(f4-f5+f6) (28) Among them: F rinei is the intraday objective function, f4 is the AEC start-stop cost, C A,boot and C A,shut are the startup cost and shutdown cost of the AEC electrolyzer, respectively; and are the state variables of the AEC startup and shutdown actions respectively; f5 is the income from selling the hydrogen generated by the AEC; The price of hydrogen sold, in RMB / kg; is the mass of hydrogen produced by the AEC during period t; f6 is the sum of the absolute values of the difference between the actual electric power consumed by the AEC and the net low-frequency load during the day; 4) Use branch and bound method to solve the intraday control scheme of AEC; 5) Continuously update the forecast data for the next 8 hours and perform the above optimization process until the AEC's 24-hour control plan is obtained.

5. The optimization control method for a hybrid water electrolysis hydrogen production system according to claim 1, wherein: The real-time PEMEC optimization control process is as follows: 1) Input the wind turbine, photovoltaic and load forecast data for the next 4 hours, use the new forecast data to calculate the new net load power, and then subtract the output of SOEC and AEC to obtain the real-time net load power: in: is the real-time net load power; and are the wind turbine, photovoltaic and load forecast data for the day; P t S P is the actual electric power consumed by SOEC obtained in the day-ahead stage; t A is the actual electric power consumed by the AEC calculated during the day; 2) Using PEMEC to absorb real-time net load while meeting PEMEC operation constraints: Constraints: Wind turbine photovoltaic output constraints: Power balance constraints: Power purchase power constraints: Where: P t P is the actual electric power consumed by PEMEC obtained during the day-ahead period; and are charging and discharging power respectively; P PV,t and P WT,t are the actual outputs of photovoltaic and wind turbine respectively; P e,t The power of purchased electricity; The upper limit of the power purchased; Battery operation constraints: in: and They are energy storage charging and discharging states respectively; and are the upper limits of energy storage charge and discharge respectively; η ch and η dis are the battery charging efficiency and discharging efficiency respectively; E 0 and E T They are the initial and final energy storage capacity, which need to be balanced; E t is the energy storage capacity at time t, E max and E min They are the upper and lower limits of energy storage capacity respectively; Other constraints: electrolyzer operation constraints, i.e., equations (1)-(16); Optimization goal: F shishi =min(f7-f8+f9+f 10 +f 11 ) (37) Among them: F shishi is the real-time objective function, f7 is the PEMEC start-stop cost, C P,boot and C P,shut are the startup cost and shutdown cost of the PEMEC electrolyzer, respectively; and are the state variables of the PEMEC startup and shutdown actions respectively; f8 is the income from selling the hydrogen produced by PEMEC; The price of hydrogen sold, in RMB / kg; is the mass of hydrogen produced by PEMEC during period t; f9 is the cost of electricity purchase; is the electricity purchase price; f 10 The penalty cost for curtailing wind and solar power; α1 and β are the penalty coefficients for curtailing solar power and wind power, respectively; f 11 Energy storage usage cost; ch and λ dis is the charging and discharging cost coefficient; 3) Use branch and bound method to solve the real-time control scheme of PEMEC; 4) Continuously update the forecast data for the next 4 hours and perform the above optimization process until the real-time 24-hour control solution of PEMEC is obtained.

6. A hybrid water electrolysis hydrogen production system optimization control system, used to implement the hybrid water electrolysis hydrogen production system optimization control method according to any one of claims 1 to 5, characterized in that: include: Hybrid water electrolysis hydrogen production system establishment unit: Establish a hybrid water electrolysis hydrogen production system including wind turbines, photovoltaics, alkaline electrolyzers, proton exchange membrane electrolyzers, solid oxide electrolyzers and batteries; Unified and universal mathematical model building unit: Establish refined unified and universal mathematical models for alkaline electrolyzers, proton exchange membrane electrolyzers, and solid oxide electrolyzers; Optimization control unit: Considering the differences in dynamic response speeds of different types of electrolyzers, combined with a low-pass filtering algorithm, a multi-stage control hybrid water electrolysis hydrogen production system is optimized; Control scheme solving unit: Constructs an objective function with minimizing operating costs as the control goal, taking into account the electricity purchase cost, energy storage usage cost, wind and solar power curtailment cost, hydrogen sales revenue and electrolyzer start-up and shutdown costs, and uses the branch and bound method to solve the control scheme of the hybrid water electrolysis hydrogen production system.

Images