Power generation energy control method for novel standby power supply system of transformer substation based on fuel cell

Abstract

The invention discloses a power generation energy control method for a novel transformer substation standby power supply system based on a fuel cell, and the method comprises the following steps: S1, constructing a complete novel transformer substation standby power supply system based on the fuel cell, and enabling the system to comprise the fuel cell used as the standby energy of the transformer substation and a super capacitor; s2, performing short-term power prediction on a fuel cell and a super capacitor in the novel standby power supply system of the transformer substation, and providing support for optimized operation and energy management of a subsequent system; and S3, energy distribution of the fuel cell and the super capacitor is optimized in real time, so that the system can quickly adapt to a continuously changing operation environment, and the energy utilization efficiency is improved. According to the invention, the system can rapidly adapt to a continuously changing operation environment, the energy utilization efficiency is improved, and the reliability and durability of the standby power supply system in long-time operation are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of backup power generation energy control of a transformer substation, and more specifically, relates to a power generation energy control method of a new backup power supply system of a transformer substation based on a fuel cell. Background Art

[0002] As an important part of the power system, substation plays a vital role in the transmission and distribution of energy. However, power outages in substations will lead to power outages in cities and industries, causing serious social and economic losses. Therefore, the reliability of the backup power system is one of the key factors in substation operation.

[0003] At present, the backup power supply of substations mainly adopts the solution of traditional lead-acid batteries and diesel generators, which has problems such as low energy density, inconvenient maintenance, and poor environmental friendliness, which is not conducive to ensuring the reliability of the power system. The proton exchange membrane fuel cell is a device that directly converts chemical energy into electrical energy through electrochemical reactions. It has the advantages of stable performance, environmental friendliness, and high energy density. At the same time, fuel cells also have the advantages of fast start-up time and high power density when dealing with sudden power accidents, which is suitable as a backup power supply for substations. However, the reliability and efficiency of fuel cells in dealing with sudden power accidents need to be further improved. At the same time, cost and system complexity are also urgent problems that need to be solved in the current field of backup power supply technology. Therefore, how to improve the reliability and efficiency of fuel cell-based backup power supply systems and reduce their cost and maintenance difficulty is an important direction of current backup power supply technology research.

[0004] Therefore, there is an urgent need to design a power generation energy control method for a new backup power supply system for a substation based on fuel cells to improve the energy utilization efficiency and the reliability and durability of the backup power supply system during long-term operation. Summary of the invention

[0005] 1. Technical issues to be resolved

[0006] Based on the defects mentioned in the above background technology, the present invention discloses a power generation energy control method for a new backup power supply system for a substation based on a fuel cell. By effectively controlling the power generation power of the fuel cell and the supercapacitor respectively, the advantages of both as backup energy sources are fully utilized, so that the system can quickly adapt to the changing operating environment, improve energy utilization efficiency, and improve the reliability and durability of the backup power supply system during long-term operation.

[0007] (II) Technical solution

[0008] The present invention discloses a method for controlling power generation energy of a new type of backup power supply system of a substation based on a fuel cell, and the control method comprises the following steps:

[0009] Step S1, constructing a complete new type of backup power supply system for a substation based on fuel cells, the system including fuel cells and supercapacitors used as backup energy sources for the substation;

[0010] Step S2: Perform short-term power forecasting on fuel cells and supercapacitors in the new backup power supply system of the substation to provide support for the subsequent system optimization operation and energy management;

[0011] Step S21, collecting historical output voltage and current data of the fuel cell and supercapacitor, and performing data preprocessing;

[0012] Data preprocessing includes:

[0013] X=[V FC , I FC , V SC , I SC ] (1)

[0014]

[0015] Among them, V FC , I FC , V SC and I SC are the output voltage and current of the fuel cell and the output voltage and current of the supercapacitor respectively. Formula (2) normalizes the historical operation data. min and X max are the maximum and minimum values ​​respectively, X norm is the normalized value of historical data X;

[0016] And calculate the output power P of the fuel cell and supercapacitor FC , P SC , as the output of the prediction, the data is also preprocessed:

[0017]

[0018] Among them, P norm is the normalized value of the output power P;

[0019] Step S22: train the LSTM model using historical data, and optimize the hyperparameters of the model using a genetic algorithm;

[0020] When using genetic algorithms to optimize the hyperparameters of the LSTM model, the fitness function uses mean square error (MSE) to evaluate model performance:

[0021]

[0022] Among them, P true (i) and P pred(i) are the actual observed value and predicted value of the i-th output power P respectively;

[0023] At the same time, in order to improve the global search ability of the genetic algorithm and avoid premature convergence, the initial population is generated using chaotic mapping, so that the population has good distribution characteristics in the search space and avoids the local concentration problem caused by traditional GA random initialization:

[0024]

[0025] Among them, n layers , n units and α are the number of LSTM layers, the number of units in each layer, and the learning rate, respectively. C0 is the original initial population, C′ is the initial population generated by chaotic mapping, and μ is the control parameter;

[0026] In addition, the adaptive mutation rate is introduced to cope with the different needs of genetic algorithms at different evolutionary stages. Usually, a larger mutation rate is needed in the early stage of the search to expand the search range, while a smaller mutation rate is needed in the later stage for a refined search.

[0027]

[0028] Among them, B mut is the mutation rate dynamically adjusted according to the population fitness, B mut,max is the maximum mutation rate, f best is the fitness of the best individual in the current generation, f avg is the average fitness of the current generation; when the fitness gap of the population is large, the mutation rate is high; and when the gap is small, the mutation rate is low;

[0029] Step S23: using the trained LSTM model to predict the short-term future power of the fuel cell and the supercapacitor;

[0030] Step S3: Optimize the energy distribution of the fuel cell and the supercapacitor in real time, so that the system can quickly adapt to the changing operating environment and improve energy utilization efficiency.

[0031] Preferably, the new backup power supply system for the substation includes a fuel preparation system, a raw material storage system, a fuel stack as a fuel cell group, a status monitoring module, a converter, a power supply detection device and a supercapacitor, the fuel preparation system is an electrolyzer, and the raw material storage system is a high-pressure hydrogen storage tank.

[0032] Preferably, the specific steps of step S1 include:

[0033] Step S11, the fuel preparation system produces hydrogen by electrolyzing water by taking electricity from the DC bus, and stores the hydrogen in the raw material storage system;

[0034] Step S12, the fuel cell stack is connected to the raw material storage system and also connected to the fuel cell status monitoring module;

[0035] Step S13, the state monitoring module is connected to the fuel preparation system, the raw material storage system and the fuel cell stack respectively;

[0036] Step S14: the converter includes a DC-DC converter and a DC-AC inverter, wherein the DC-DC converter is connected to the battery stack and the DC bus; the DC-AC inverter connects the DC bus to the AC bus to achieve bidirectional conversion of electric energy;

[0037] Step S15, the power supply detection device is divided into an AC power supply detection device and a DC power supply detection device for AC side data collection and DC side data collection, and the AC power supply detection device and the DC power supply detection device are connected to the AC side and the DC side of the converter respectively;

[0038] Step S16: The supercapacitor is directly connected to the DC bus.

[0039] Preferably, the status monitoring module measures the operating parameters of the fuel cell stack in real time through sensors, and monitors the operating parameters of the electrolyzer and the hydrogen storage tank in real time.

[0040] Preferably, the AC bus has two incoming lines connected to each other, and the AC / DC bus is provided with a power supply detection device. The detection line end of the AC power supply detection device is connected to the AC outlet end of the DC-AC inverter, and detects the power supply voltage and current signals on the AC side; the detection line end of the DC power supply detection device is connected to the outlet end of the DC-DC converter, and detects the power supply voltage and current signals of the DC bus.

[0041] Preferably, the specific steps of step S3 include:

[0042] Step S31, acquiring the voltage and current data of the fuel cell and the supercapacitor in real time, and calculating the hydrogen production capacity of the electrolyzer, measuring the real-time power generation of the fuel cell and the remaining energy storage of the supercapacitor;

[0043] Step S32: Set the energy allocation objective function and use the dynamic programming algorithm to perform energy management.

[0044] Preferably, the step S31 includes:

[0045]

[0046] The hydrogen production of the electrolyzer is The unit is mole, I elec is the electrolytic cell current in amperes, t is the operating time in seconds, and F is the Faraday constant. The theoretical power generation satisfies:

[0047]

[0048] Among them, P th (t) is the theoretical power generation in watts, and ΔH is the enthalpy change of hydrogen combustion; based on this, a reasonable upper and lower limit is set for the real-time power generation of the fuel cell:

[0049] Minimum power generation P min (t):

[0050] P min (t) = k1P th (t) (9)

[0051] Here k1 is a coefficient less than 1, representing the efficiency of the fuel cell;

[0052] Maximum power generation P max (t):

[0053] P max (t) = k2P th (t) (10)

[0054] Here k2 is a coefficient greater than 1, representing the maximum efficiency of the fuel cell under good working conditions;

[0055] The remaining energy storage of the supercapacitor can be calculated by the following formula:

[0056]

[0057] Where E SC ′(t) is the remaining energy storage of the supercapacitor in joules, C SC is the capacitance of the supercapacitor in farads, V SC (t) is the voltage of the supercapacitor in volts.

[0058] Preferably, the step S32 includes:

[0059] Setting the allocation objective function, which is designed to minimize the total energy cost of the system or maximize the economic benefits of the system while satisfying the constraints of energy demand and hydrogen production;

[0060]

[0061] Where m is the total energy cost, m FC (t) is the energy cost of the fuel cell at time t, m SC (t) is the energy cost of the supercapacitor at time t;

[0062] At each time step t, the energy balance of the system should satisfy:

[0063] E FC (t)+E SC (t) = P true (t)×Δt (13)

[0064] Among them, P true (t) is the predicted output power at that moment, Δt is the time step, E FC (t) and E SC (t) is the energy supplied by the fuel cell and supercapacitor;

[0065] Fuel cell energy supply meets the constraints:

[0066] P min ×Δt≤E FC (t)≤P max ×Δt (14)

[0067] Supercapacitor energy supply meets the constraints:

[0068] E SC ′(t)≤E SC (t)≤P FC ″×Δt (15)

[0069] Among them, P FC ″ is the rated output power of the supercapacitor. Based on the above conditions and objective function, a dynamic programming algorithm is used for energy management to control and adjust the power generation of the fuel cell and supercapacitor respectively, so as to improve the responsiveness and economy of the backup power supply of the substation during power outages.

[0070] (III) Beneficial effects

[0071] (1) The new backup power supply system for the substation of the present invention uses a combination of fuel cells and supercapacitors as backup energy sources for the substation. The electric energy generated by the fuel cell is connected to the DC bus through a DC-DC converter, which can stabilize the output voltage at a specific voltage value, thereby avoiding unstable operation of the load equipment caused by output voltage fluctuations, and greatly improving energy utilization efficiency. At the same time, this backup power supply architecture can adapt the output voltage of the fuel cell to different load devices by adding a DC-DC converter, thereby increasing the flexibility and applicability of the system. Supercapacitors are used in this system architecture as emergency backup power supplies and can be used when the fuel cell group is just started or cannot work normally. At the same time, because the output voltage range of supercapacitors is large, choosing to connect the supercapacitor directly to the DC bus instead of connecting it to the DC bus through a DC-DC converter can avoid the delay caused by the start-up and control of the DC-DC converter, thereby ensuring a rapid response after the DC bus loses power and ensuring the stable provision of backup power.

[0072] (2) The present invention also aims to reduce system energy consumption and improve system durability, and studies the online real-time energy optimization scheduling method of the new backup power system of the substation of the proton exchange membrane fuel cell, by collecting historical data of the fuel cell and supercapacitor, and using the long short-term memory (LSTM) network for short-term power prediction. The LSTM network is good at processing time series data and can capture the long-term dependencies in the system output power, thereby improving the accuracy of the prediction. In addition, the hyperparameters of the LSTM model are optimized by genetic algorithms, which further improves the generalization ability and stability of the model. This method provides accurate short-term power prediction support for subsequent system energy management, can more effectively control the energy flow between the fuel cell and the supercapacitor, and optimize the overall power generation efficiency of the system.

[0073] (3) The present invention combines the real-time data of fuel cell power generation, supercapacitor remaining energy storage and electrolyzer hydrogen production capacity to establish an energy allocation objective function and optimize it using a dynamic programming algorithm. This method dynamically adjusts the energy output of fuel cells and supercapacitors through real-time monitoring and calculation of the system, ensuring that the system can quickly respond to changing load demands, thereby improving energy utilization efficiency and extending system life. Online energy optimization scheduling not only reduces the energy consumption of the system, but also enhances the reliability and durability of the system during long-term operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0074] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments:

[0075] Figure 1 It is a structural block diagram of a new type of backup power supply system for a substation based on a fuel cell in the present invention;

[0076] Figure 2 A flow chart of a method for controlling power generation energy of a new backup power supply system for a substation based on fuel cells;

[0077] Figure 3 For the present invention Figure 2 Flow chart of step S1 in FIG.

[0078] Figure 4 For the present invention Figure 2 Flow chart of step S2;

[0079] Figure 5 For the present invention Figure 2 Flow chart of step S3 in FIG. DETAILED DESCRIPTION

[0080] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution in the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0081] In view of the defects in the prior art, the present invention proposes a power generation energy control method for a new type of backup power supply system for a substation based on a fuel cell, wherein the new type of backup power supply system for a substation based on a fuel cell includes a fuel cell and a super capacitor as backup energy sources.

[0082] The structure of the new backup power supply system for substations based on fuel cells in the present invention will be described in detail below. Figure 1 As shown in the figure, the new backup power system for substations based on fuel cells includes an electrolyzer for fuel preparation and a high-pressure hydrogen storage tank for storage, a fuel cell stack for directly converting the chemical energy of hydrogen fuel into electrical energy, a fuel cell status monitoring module, a DC-AC inverter, a DC-DC converter, a power supply detection device and a supercapacitor. The functions of each module are as follows:

[0083] (A) The proton exchange membrane fuel cell is preferably a hydrogen fuel cell, which includes an electrolyzer, a high-pressure hydrogen storage tank and a stack. The electrolyzer produces hydrogen and oxygen by taking electricity from the AC voltage bus, and the produced hydrogen and oxygen raw materials are stored in the high-pressure hydrogen storage tank, so as to continuously and stably provide raw materials to the fuel cell stack. After the fuel cell starts working, the backup power source of its substation can be supplied by the fuel cell power via the DC bus and the DC-AC inverter, so as to realize the efficient recycling of the fuel cell energy.

[0084] The fuel cell group mainly uses hydrogen and oxygen to react to generate electricity to supply various load devices. Hydrogen and oxygen are produced by electrolysis of water. This electricity comes directly from the DC bus in the substation. The whole process is environmentally friendly and efficient. Before using the fuel cell, the hydrogen is stored in a storage container. As the fuel cell group starts working, hydrogen and oxygen are continuously input into the fuel cell stack to undergo chemical reactions to generate electricity and output to the converter for the next application. The supply of oxygen and hydrogen in the whole process is accurately controlled by the fuel cell status monitoring system, which can be dynamically adjusted according to the power requirements of the load equipment and the voltage and current of the fuel cell.

[0085] (B) The status monitoring module is connected to the electrolyzer, high-pressure hydrogen storage tank and stack of the fuel cell respectively, and through this monitoring module, the output power of the fuel cell can be dynamically adjusted according to the power demand of the load equipment and the voltage and current of the fuel cell, so as to achieve the stability and reliability of the power provided by the fuel cell group. The fuel cell group formed by connecting the fuel cells in series and parallel can better improve the capacity and terminal voltage of the battery group; the status monitoring module is connected to each fuel cell, and is also connected to the electrolyzer and high-pressure hydrogen storage tank, so as to monitor the battery status in real time and control the battery power generation.

[0086] (C) The converter includes a DC-DC converter and a DC-AC inverter, wherein the DC-DC converter is connected to the fuel cell stack and the DC bus. This device can ensure the stability of the DC bus voltage and avoid unstable operation of the load equipment due to fluctuations in the fuel cell output voltage; the DC-AC inverter is specifically a bidirectional DC-AC inverter, which connects the DC bus with the AC bus to achieve DC-AC conversion and AC-DC conversion functions, breaking the traditional AC-DC separation problem. Under normal working conditions of the substation, it can supply power to the supercapacitor and electrolyzer on the DC side, thereby improving the energy utilization rate and the applicability of the system. At the same time, the converter control system is used to connect to the DC-DC converter and the DC-AC inverter respectively to control the operation of the converter. The converter control signal adopts a PWM wave, and its power supply is also supplied by the fuel cell. The control signal can be adjusted according to the load equipment of different systems.

[0087] (D) The power supply detection device is divided into an AC power supply detection device and a DC power supply detection device for AC side data collection and DC side data collection, which are respectively connected to the AC side and the DC side of the converter to facilitate the detection of the AC and DC bus voltage status;

[0088] (E) The supercapacitor is directly connected to the DC bus and serves as an emergency backup power source to ensure normal power supply to the system when the fuel cell cannot work normally.

[0089] (F) The AC bus has two incoming lines connected, namely, incoming line I and incoming line II. The two incoming lines serve as backup for each other to ensure the reliability of the power system. The DC bus is connected to the converter and is also connected to the supercapacitor to achieve the purpose of stable and reliable power supply of the DC bus. A DC power supply detection device is provided for the AC and DC voltage bus. The detection line end of the AC power supply detection device is connected to the AC outlet end of the DC-AC inverter to detect the power supply voltage and current signals on the AC side; the detection line end of the DC power supply detection device is connected to the outlet end of the DC-DC converter to detect the power supply voltage and current signals on the DC side bus. These detection signals will be transmitted to the main control room through the signal line. When a certain parameter is abnormal, an alarm will be issued to attract the attention of the staff, thereby achieving the purpose of detecting the bus status.

[0090] In the substation backup power supply architecture of the present invention, fuel cells are used as the main power supply. When the main power grid fails or loses power, the backup power supply will immediately take over the power supply to ensure the reliability and stability of the substation backup power supply. At the same time, considering that the fuel cell needs a certain startup time, a supercapacitor is added to this backup power supply system as an emergency backup power supply to supplement and solve the problem of the fuel cell startup power supply at the moment of system power failure; the supercapacitor has a large output voltage range and can meet the DC bus voltage requirements. At the moment of system power failure, it releases electrical energy to the DC bus to provide energy for the entire network. It can also make up for the lack of energy when the fuel cell supply capacity fluctuates greatly (such as changes in the operating environment) to ensure the stability of power supply.

[0091] In general, the overall working process of the substation backup power system based on fuel cells is as follows: when the main power grid fails or is cut off, the supercapacitor is first used to provide emergency power to ensure that the system can immediately restore power supply at the moment of power failure. Then the fuel cell is started by the battery control signal and generates electricity. The output battery voltage is stabilized within a suitable range through the DC-DC converter to ensure the stability of the DC bus voltage. Secondly, the DC-AC inverter can break the problem of AC-DC separation of traditional backup power supplies, and apply the power of the fuel cell to the AC load side to achieve efficient use of the backup power supply. The power supply capacity of the fuel cell and supercapacitor of the entire backup power system can be dynamically adjusted and allocated according to the system load capacity to achieve precise control of the battery discharge amount.

[0092] exist Figure 1 On the basis of Figure 2 The flowchart is a step flow chart of a method for controlling power generation energy of a new backup power supply system for a substation based on a fuel cell, which includes the following steps:

[0093] Step S1, constructing a complete new type of backup power supply system for a substation based on fuel cells, the system including fuel cells and supercapacitors used as backup energy sources for the substation;

[0094] Step S2: Perform short-term power forecasting on fuel cells and supercapacitors in the new backup power supply system of the substation to provide support for the subsequent system optimization operation and energy management;

[0095] Step S3: Optimize the energy distribution of the fuel cell and the supercapacitor in real time, so that the system can quickly adapt to the changing operating environment and improve energy utilization efficiency.

[0096] In another embodiment, the new backup power system for the substation includes a fuel preparation system, a raw material storage system, a fuel cell stack as a fuel cell group, a state monitoring module, a converter, a power detection device and a supercapacitor. The fuel cell is specifically a hydrogen fuel cell, the fuel preparation system is an electrolyzer, and the raw material storage system is a high-pressure hydrogen storage tank.

[0097] In another embodiment, the specific steps of step S1 include:

[0098] Step S11, the fuel preparation system produces hydrogen by electrolyzing water by taking electricity from the DC bus, and stores the hydrogen in the raw material storage system;

[0099] Step S12, the fuel cell stack is connected to the raw material storage system and also connected to the fuel cell status monitoring module;

[0100] Step S13, the state monitoring module is connected to the fuel preparation system, the raw material storage system and the fuel cell stack respectively;

[0101] Step S14: the converter includes a DC-DC converter and a DC-AC inverter, wherein the DC-DC converter is connected to the battery stack and the DC bus; the DC-AC inverter connects the DC bus to the AC bus to achieve bidirectional conversion of electric energy;

[0102] Step S15, the power supply detection device is divided into an AC power supply detection device and a DC power supply detection device for AC side data collection and DC side data collection, and the AC power supply detection device and the DC power supply detection device are connected to the AC side and the DC side of the converter respectively;

[0103] Step S16: The supercapacitor is directly connected to the DC bus.

[0104] In another embodiment, the fuel cell stack formed by connecting the single fuel cells in series and parallel, the state monitoring module measures the operating parameters of the fuel cell stack in real time through sensors, and monitors the operating parameters of the electrolyzer and the hydrogen storage tank in real time. In addition, the AC bus has two incoming lines connected to each other and is connected to the DC-AC inverter at the same time; the DC bus is connected to the DC-DC converter, the DC-AC inverter and the supercapacitor. The AC and DC bus are provided with a power supply detection device, the detection line end of the AC power supply detection device is connected to the AC outlet end of the DC-AC inverter, and the power supply voltage and current signal on the AC side are detected; the detection line end of the DC power supply detection device is connected to the outlet end of the DC-DC converter, and the power supply voltage and current signal of the DC bus are detected.

[0105] In another embodiment, the specific steps of step S2 include:

[0106] Step S21: Collect historical output voltage and current data of the fuel cell and supercapacitor, and perform data preprocessing.

[0107] Data preprocessing includes:

[0108] X=[V FC , I FC , V SC , I SC ] (1)

[0109]

[0110] Among them, V FC , I FC , V SC and I SC are the output voltage and current of the fuel cell and the output voltage and current of the supercapacitor respectively. Formula (2) normalizes the historical operation data. min and X max are the maximum and minimum values ​​respectively, X norm is the normalized value of historical data X.

[0111] And calculate the output power P of the fuel cell and supercapacitor FC , P SC , as the output of the prediction, the data is also preprocessed:

[0112]

[0113] Among them, P norm is the normalized value of the output power P.

[0114] Step S22: Use historical data to train the LSTM model, and use a genetic algorithm to optimize the hyperparameters of the model.

[0115] When using genetic algorithms to optimize the hyperparameters of the LSTM model, the fitness function uses mean square error (MSE) to evaluate model performance:

[0116]

[0117] Among them, P true (i) and P pred (i) are the actual observed value and predicted value of the i-th output power P, respectively.

[0118] At the same time, in order to improve the global search ability of the genetic algorithm and avoid premature convergence, the initial population is generated using chaotic mapping, so that the population has good distribution characteristics in the search space and avoids the local concentration problem caused by traditional GA random initialization:

[0119]

[0120] Among them, n layers , n units and α are the number of LSTM layers, the number of units in each layer, and the learning rate, respectively. C0 is the original initial population, C′ is the initial population generated by chaotic mapping, and μ is a control parameter, the specific value of which depends on the actual case.

[0121] In addition, an adaptive mutation rate is introduced to cope with the different needs of genetic algorithms at different evolutionary stages. Usually, a larger mutation rate is required in the early stage of the search to expand the search range, while a smaller mutation rate is required in the later stage for a refined search.

[0122]

[0123] Among them, B mut is the mutation rate dynamically adjusted according to the population fitness, B mut,max is the maximum mutation rate, f best is the fitness of the best individual in the current generation, f avg is the average fitness of the current generation; when the fitness gap of the population is large, the mutation rate is high; and when the gap is small, the mutation rate is low.

[0124] Step S23: Use the trained LSTM model to predict the short-term future power of the fuel cell and the supercapacitor.

[0125] The hyperparameters of the LSTM model trained with historical data are optimized through genetic algorithms, so that accurate short-term power forecasts can help energy management systems more effectively allocate and schedule energy flows between fuel cells and supercapacitors to improve the overall efficiency and life of the system. The LSTM network can better handle long-term dependency issues in time series data to improve the accuracy of short-term power forecasts. The parameter configuration of the LSTM model is automatically optimized through genetic algorithms to improve the generalization ability and stability of the forecasting model.

[0126] In another embodiment, the specific steps of step S3 include:

[0127] Step S31, real-time acquisition of voltage and current data of the fuel cell and the supercapacitor, and calculation of the hydrogen production capacity of the electrolyzer, measuring the real-time power generation of the fuel cell and the remaining energy storage of the supercapacitor.

[0128] The step S31 comprises:

[0129]

[0130] The hydrogen production of the electrolyzer is The unit is mole, I elec is the electrolytic cell current in amperes, t is the operating time in seconds, and F is the Faraday constant. The theoretical power generation satisfies:

[0131]

[0132] Among them, P th (t) is the theoretical power generation in watts, and ΔH is the enthalpy change of hydrogen combustion; based on this, a reasonable upper and lower limit is set for the real-time power generation of the fuel cell:

[0133] Minimum power generation P min (t):

[0134] P min (t) = k1P th (t) (9)

[0135] Here k1 is a coefficient less than 1, representing the efficiency of the fuel cell.

[0136] Maximum power generation P max (t):

[0137] P max (t) = k2P th (t) (10)

[0138] Here k2 is a coefficient greater than 1, representing the maximum efficiency of the fuel cell under good working conditions.

[0139] The remaining energy storage of the supercapacitor can be calculated by the following formula:

[0140]

[0141] Where E SC ′(t) is the remaining energy storage of the supercapacitor in joules, C SC is the capacitance of the supercapacitor in farads, V SC (t) is the voltage of the supercapacitor in volts.

[0142] Step S32: Set the energy allocation objective function and use the dynamic programming algorithm to perform energy management.

[0143] The step S32 includes: setting a distribution objective function designed to minimize the total energy cost of the system or maximize the economic benefits of the system while satisfying the constraints of energy demand and hydrogen production.

[0144]

[0145] Where m is the total energy cost, m FC (t) is the energy cost of the fuel cell at time t, m SC (t) is the energy cost of the supercapacitor at time t.

[0146] At each time step t, the energy balance of the system should satisfy:

[0147] E FC (t)+E SC (t) = P true (t)×Δt (13)

[0148] Among them, P true (t) is the predicted output power at that moment, Δt is the time step, E FC (t) and E SC (t) is the energy supplied by the fuel cell and supercapacitor.

[0149] Fuel cell energy supply meets the constraints:

[0150] P min ×Δt≤E FC (t)≤P max ×Δt (14)

[0151] Supercapacitor energy supply meets the constraints:

[0152] E SC ′(t)≤E SC (t)≤P FC ″×Δt (15)

[0153] Among them, PFC ″ is the rated output power of the supercapacitor. Based on the above conditions and objective function, a dynamic programming algorithm is used for energy management to control and adjust the power generation of the fuel cell and the supercapacitor respectively, so as to improve the responsiveness and economy of the backup power supply of the substation during power outages.

[0154] It can be seen that the present invention aims to reduce system energy consumption and improve system durability, and studies the online real-time energy optimization scheduling method of the new backup power supply system of the substation of the proton exchange membrane fuel cell. It utilizes the high efficiency, low pollution, high reliability and other characteristics of the fuel cell, combines supercapacitors, DC-DC converters and DC-AC inverters and energy optimization technologies, and realizes the rapid response, precise control and high applicability of the backup power supply system.

[0155] In addition, the main advantages of the present invention are:

[0156] The fuel preparation and storage system of the present invention can realize the energy reserve of fuel cells; the fuel cell stack as the main energy supply of the backup power supply system can realize the continuous and reliable power supply of the backup power supply; the fuel cell state monitoring system can realize the monitoring and control of the backup power supply state, ensuring the rapid response to the main network failure and the normal operation of the backup power supply; the converter is used to stabilize the voltage and solve the problem of the split of the AC and DC system of the traditional backup power supply; the power supply detection device detects the bus voltage state and sends the corresponding signal to the centralized monitoring unit; the supercapacitor is used as an emergency backup power supply to ensure the fast and seamless backup power after the DC bus loses power. And the fuel cell-supercapacitor system is predicted for short-term power to provide important data support and decision-making basis for the subsequent system optimization operation. Dynamically adjusting the energy distribution of the fuel cell and the supercapacitor can enable the system to quickly adapt to the changing operating environment and improve the overall energy utilization efficiency. The present invention uses fuel cells and supercapacitors as backup power sources, so that the substation backup power supply system has the characteristics of high reliability, high efficiency and low pollution, and even when a single point failure occurs, it still helps to achieve the stability and reliability of the system as a whole. Therefore, the present invention has practicality and promotion value.

[0157] Finally, it is noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for controlling power generation energy of a new type of backup power supply system for a substation based on a fuel cell, characterized in that: The control method comprises the following steps: Step S1, constructing a complete new type of backup power supply system for a substation based on fuel cells, the system including fuel cells and supercapacitors used as backup energy sources for the substation; Step S2: Perform short-term power forecasting on fuel cells and supercapacitors in the new backup power supply system of the substation to provide support for the subsequent system optimization operation and energy management; Step S21, collecting historical output voltage and current data of the fuel cell and supercapacitor, and performing data preprocessing; Data preprocessing includes: X=[V FC ,I FC ,V SC ,I SC ] (1) Among them, V FC , I FC , V SC and I SC are the output voltage and current of the fuel cell and the output voltage and current of the supercapacitor respectively. Formula (2) normalizes the historical operation data. min and X max are the maximum and minimum values ​​respectively, X norm is the normalized value of historical data X; And calculate the output power P of the fuel cell and supercapacitor FC , P SC , as the output of the prediction, the data is also preprocessed: Among them, P norm is the normalized value of the output power P; Step S22: train the LSTM model using historical data, and optimize the hyperparameters of the model using a genetic algorithm; When using genetic algorithms to optimize the hyperparameters of the LSTM model, the fitness function uses mean square error (MSE) to evaluate model performance: Among them, P true (i) and P pred (i) are the actual observed value and predicted value of the i-th output power P respectively; At the same time, in order to improve the global search ability of the genetic algorithm and avoid premature convergence, the initial population is generated using chaotic mapping, so that the population has good distribution characteristics in the search space and avoids the local concentration problem caused by traditional GA random initialization: Among them, n layers , n units and α are the number of LSTM layers, the number of units in each layer, and the learning rate, respectively. C0 is the original initial population, C ′ is the initial population generated by chaotic mapping, μ is the control parameter; In addition, the adaptive mutation rate is introduced to cope with the different needs of genetic algorithms at different evolutionary stages. Usually, a larger mutation rate is needed in the early stage of the search to expand the search range, while a smaller mutation rate is needed in the later stage for a refined search. Among them, B mut is the mutation rate dynamically adjusted according to the population fitness, B mut,max is the maximum mutation rate, f best is the fitness of the best individual in the current generation, f avg is the average fitness of the current generation; when the fitness gap of the population is large, the mutation rate is high; and when the gap is small, the mutation rate is low; Step S23: using the trained LSTM model to predict the short-term future power of the fuel cell and the supercapacitor; Step S3: Optimize the energy distribution of the fuel cell and the supercapacitor in real time, so that the system can quickly adapt to the changing operating environment and improve energy utilization efficiency.

2. The power generation energy control method of the new backup power supply system of a substation based on a fuel cell according to claim 1 is characterized in that: The new backup power supply system for the substation includes a fuel preparation system, a raw material storage system, a fuel cell stack as a fuel cell group, a status monitoring module, a converter, a power supply detection device and a supercapacitor. The fuel preparation system is an electrolyzer, and the raw material storage system is a high-pressure hydrogen storage tank.

3. The power generation energy control method of the new backup power supply system of a substation based on a fuel cell according to claim 2 is characterized in that: The specific steps of step S1 include: Step S11, the fuel preparation system produces hydrogen by electrolyzing water by taking electricity from the DC bus, and stores the hydrogen in the raw material storage system; Step S12, the fuel cell stack is connected to the raw material storage system and also connected to the fuel cell status monitoring module; Step S13, the state monitoring module is connected to the fuel preparation system, the raw material storage system and the fuel cell stack respectively; Step S14: the converter includes a DC-DC converter and a DC-AC inverter, wherein the DC-DC converter is connected to the battery stack and the DC bus; the DC-AC inverter connects the DC bus to the AC bus to achieve bidirectional conversion of electric energy; Step S15, the power supply detection device is divided into an AC power supply detection device and a DC power supply detection device for AC side data acquisition and DC side data acquisition, and the AC power supply detection device and the DC power supply detection device are connected to the AC side and the DC side of the converter respectively; Step S16: The supercapacitor is directly connected to the DC bus.

4. The power generation energy control method of the new backup power supply system of a substation based on a fuel cell according to claim 3 is characterized in that: The state monitoring module measures the operating parameters of the fuel cell stack in real time through sensors, and monitors the operating parameters of the electrolyzer and hydrogen storage tank in real time.

5. The power generation energy control method of the new backup power supply system of a substation based on a fuel cell according to claim 4 is characterized in that: The AC bus has two incoming lines connected to each other, and the AC and DC bus is equipped with a power supply detection device. The detection line end of the AC power supply detection device is connected to the AC outlet end of the DC-AC inverter, and detects the power supply voltage and current signals on the AC side; the detection line end of the DC power supply detection device is connected to the outlet end of the DC-DC converter, and detects the power supply voltage and current signals of the DC bus.

6. The power generation energy control method of the new backup power supply system of a substation based on a fuel cell according to claim 1 is characterized in that: The specific steps of step S3 include: Step S31, acquiring the voltage and current data of the fuel cell and the supercapacitor in real time, and calculating the hydrogen production capacity of the electrolyzer, measuring the real-time power generation of the fuel cell and the remaining energy storage of the supercapacitor; Step S32: Set the energy allocation objective function and use the dynamic programming algorithm to perform energy management.

7. The power generation energy control method of the new backup power supply system of a substation based on a fuel cell according to claim 6 is characterized in that: The step S31 comprises: The hydrogen production of the electrolyzer is The unit is mole, I elec is the electrolytic cell current in amperes, t is the operating time in seconds, and F is the Faraday constant. The theoretical power generation satisfies: Among them, P th (t) is the theoretical power generation in watts, and ΔH is the enthalpy change of hydrogen combustion; based on this, a reasonable upper and lower limit is set for the real-time power generation of the fuel cell: Minimum power generation P min (t): P min (t)=k1P th (t) (9) Here k1 is a coefficient less than 1, representing the efficiency of the fuel cell; Maximum power generation P max (t): P max (t)=k2P th (t) (10) Here k2 is a coefficient greater than 1, representing the maximum efficiency of the fuel cell under good working conditions; The remaining energy storage of the supercapacitor can be calculated by the following formula: Where E SC ′(t) is the remaining energy storage of the supercapacitor in joules, C SC is the capacitance of the supercapacitor in farads, V SC (t) is the voltage of the supercapacitor in volts.

8. The power generation energy control method of the new backup power supply system of a substation based on a fuel cell according to claim 7 is characterized in that: The step S32 comprises: Setting the allocation objective function, which is designed to minimize the total energy cost of the system or maximize the economic benefits of the system while satisfying the constraints of energy demand and hydrogen production; Where m is the total energy cost, m FC (t) is the energy cost of the fuel cell at time t, m SG (t) is the energy cost of the supercapacitor at time t; At each time step t, the energy balance of the system should satisfy: E FC (t)+E SC (t)=P true (t)×Δt (13) Among them, P true (t) is the predicted output power at that moment, Δt is the time step, E FC (t) and E SG (t) is the energy supplied by the fuel cell and supercapacitor; Fuel cell energy supply meets the constraints: P min ×Δt≤E FC (t)≤P max ×Δt (14) Supercapacitor energy supply meets the constraints: E SC ′(t)≤E SC (t)≤P FC "×Δt (15) Among them, P FC ″ is the rated output power of the supercapacitor. Based on the above conditions and objective function, a dynamic programming algorithm is used for energy management to control and adjust the power generation of the fuel cell and supercapacitor respectively, so as to improve the responsiveness and economy of the substation backup power supply during power outages.

Images