Wind-light-water hydrogen storage micro-grid system capacity configuration optimization method based on prediction model

Abstract

The invention belongs to the technical field of micro-grids, and particularly provides a wind-light-water hydrogen storage micro-grid system capacity configuration optimization method based on a prediction model, and the method comprises the steps: 1, building a wind-light-water hydrogen storage micro-grid system architecture; 2, establishing a wind-light-water hydrogen storage micro-grid system model; 3, designing a control strategy of the wind-light-water hydrogen storage micro-grid system, and controlling the operation states of a storage battery pack, a pumped storage power station, a hydrogen fuel cell and an electrolytic cell by adopting a model prediction control algorithm; and 4, substituting the wind-light resource data and the power and hydrogen load data into the model established in the step 2, and solving the model by adopting an optimization algorithm to obtain the optimal capacity ratio of each device of the wind-light-water hydrogen storage micro-grid system. And 5, applying the control strategy in the step 3 to daily operation control of the wind-light-water hydrogen storage micro-grid system with the optimal capacity. The method can ensure the coordination and consistency of the capacity configuration result and the control strategy, thereby improving the economy and reliability of the whole life cycle of the system.

Description

Technical Field

[0001] The present invention belongs to the technical field of microgrids, and particularly relates to an optimization method for the capacity configuration of a wind-solar-hydrogen storage microgrid system based on a prediction model. Background Art

[0002] As an important integrated form of distributed energy, the microgrid system plays an increasingly important role in the energy field. The wind-solar-hydrogen storage microgrid system is a renewable energy comprehensive utilization technology with multi-energy complementarity, which is widely used to solve the problem of sustainable energy development of traditional power grids. It realizes diversified energy supply and efficient utilization by integrating various energy technologies such as wind power generation, photovoltaic power generation, hydropower, battery energy storage, and hydrogen energy utilization.

[0003] Existing research mainly focuses on the combined application of wind energy, light energy, and water energy, and less attention is paid to the comprehensive integration of these energies with hydrogen energy and battery energy storage systems. The roles and interactions of various energies in the system have not been fully considered, which limits the potential of multi-energy collaborative optimization. In the process of capacity configuration of the wind-solar-hydrogen storage microgrid system, rule-based energy control strategies are mainly adopted. These strategies are relatively rough in the capacity configuration of different devices during the optimization stage and cannot fully adapt to the complex situations of different load and resource changes. In the research on microgrid system control, advanced control methods are gradually applied, but these methods are often not matched with the previous energy configuration strategies, resulting in the difficulty of fully exerting the system performance. In addition, there is a lack of effective cooperative operation strategies among various devices, and the complementary characteristics of multiple energies are not fully utilized. The resulting problems of low device operation efficiency and imperfect system energy management lead to an increase in the overall operation cost and serious resource waste.

[0004] There are deficiencies in many aspects in the existing technology: First, the capacity configuration of devices is often based on rule-based energy control strategies, but more advanced control strategies in the subsequent research on microgrid system control strategies result in the problem that the capacity configuration results do not correspond to the control strategies, leading to a reduction in system performance such as economy and reliability, causing resource waste and system inefficiency; Second, the existing technology has insufficient prediction ability for wind-solar resources and loads during the process of optimizing the capacity configuration of the microgrid system, and cannot effectively respond to the impact of weather changes and load changes on energy supply and demand, resulting in problems such as large amplitude and high frequency of device power changes, which greatly affect the device life. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide an optimization method for the capacity configuration of a wind-solar-hydrogen storage microgrid system based on a prediction model to ensure the coordination and consistency between the capacity configuration results and the control strategies, thereby improving the economy and reliability of the system throughout its life cycle.

[0006] To solve the above technical problems, the technical solution adopted by the present invention is: an optimization method for the capacity configuration of a wind-solar-hydrogen-storage microgrid system based on a prediction model, comprising the following steps: Step 1. Establish the architecture of the wind-solar-hydrogen-storage microgrid system: Establish a wind-solar-hydrogen-storage microgrid system through a wind turbine generator set, a photovoltaic generator set, a pumped-storage power station, a battery pack, a hydrogen fuel cell pack, a hydrogen storage tank group, and an electrolyzer group, and construct the constraint conditions and objective functions of the wind-solar-hydrogen-storage microgrid system; Step 2. Establish the model of the wind-solar-hydrogen-storage microgrid system: Include the power characteristic models of the wind turbine generator set, the photovoltaic generator set, and the pumped-storage power station, the energy conversion models of the hydrogen fuel cell pack and the electrolyzer, and the state models of the pumped-storage power station, the battery pack, and the hydrogen storage tank group; Step 3. Design the control strategy of the wind-solar-hydrogen-storage microgrid system: Use the model predictive control algorithm to control the operating states of the battery pack, the pumped-storage power station, the hydrogen fuel cell, and the electrolyzer; Step 4. Input the wind-solar resource data and the electricity and hydrogen load data into the wind-solar-hydrogen-storage microgrid system model established in Step 2, and use an optimization algorithm to solve the model to obtain the optimal capacity ratio of the wind-solar-hydrogen-storage microgrid system; Step 5. Apply the control strategy in Step 3 to the daily operation control of the wind-solar-hydrogen-storage microgrid system with the optimal capacity.

[0007] In a preferred solution, in Step 1, the first objective function of the wind-solar-hydrogen-storage microgrid system is the total system cost, and the expression is: ; In the formula, is the initial investment cost of the wind-solar-hydrogen-storage microgrid system, is the operation and maintenance cost of the wind-solar-hydrogen-storage microgrid system, is the replacement cost of the wind-solar-hydrogen-storage microgrid system, , are respectively the electricity purchase cost and the electricity sale revenue of the wind-solar-hydrogen-storage microgrid system; The second objective function is the load power shortage rate, and the expression is: ; In the formula, is the load power shortage amount at time is the load demand power amount at time; The third objective function is the percentage of renewable energy power generation, and the expression is: ; In the formula, is Photovoltaic power generation of the photovoltaic power generation unit at a certain moment, is wind power generation of the wind power generation unit at a certain moment, is power generation of the pumped-storage power station at a certain moment; The fourth objective function is the curtailment rate of wind and light , and the expression is: ; In the formula, is the curtailment amount of wind and light; The total objective function of the system is a combination of the total system cost, the load power outage rate, the percentage of renewable energy power generation, and the curtailment rate of wind and light. The expression is: ; In the formula, is the normalized total system cost; f1, f2, f3, and f4 are the weight coefficients of the total system cost, the load power outage rate, the percentage of renewable energy power generation, and the curtailment rate of wind and light, respectively, and are designed according to the degree of emphasis on different objectives.

[0008] In the preferred solution, in the first step, the constraint conditions of the wind-solar-hydrogen storage microgrid system include power balance constraint, equipment number constraint, power output constraint of power sources, and equipment capacity constraint; The power balance constraint expression is: ; In the formula, , , , , , , , are the load demand power, the wind turbine output, the photovoltaic output, the pumped-storage power station output, the battery output, the electrolyzer consumption power, the fuel cell output, and the grid output, respectively; The power output constraint of the power source is expressed as: ; ; ; ; ; ; In the formula, , , , , are the minimum outputs of the wind power generation unit, the photovoltaic power generation unit, the pumped storage power station, the battery pack, and the fuel cell pack, respectively. , , , , are the maximum powers of the wind power generation unit, the photovoltaic power generation unit, the pumped storage power station, the battery pack, and the fuel cell pack, respectively. is the minimum power that the electrolyzer unit can consume. is the maximum power that the electrolyzer unit can consume. , , , , , are the maximum powers of the wind power generation unit, the photovoltaic power generation unit, the pumped storage power station, the battery pack, the fuel cell pack, and the electrolyzer unit, respectively. Number of devices The constraint is expressed as: ; In the formula, and are the upper and lower limits of the installation quantity of the device, where represent the wind power generation unit, the photovoltaic power generation unit, the pumped storage power station, the battery pack, the hydrogen fuel cell pack, the hydrogen storage tank pack, and the electrolyzer unit, respectively. The device capacity constraint is expressed as: ; ; ; In the formula, represents the percentage of the remaining battery power at time t to the maximum battery capacity; represents the percentage of the hydrogen content in the hydrogen storage tank at time t to the maximum hydrogen capacity of the hydrogen storage tank; represents the percentage of the remaining water volume in the pumped storage power station reservoir at time t to the maximum capacity of the pumped storage power station reservoir.

[0009] In the preferred solution, the power characteristic models of the wind power generation unit, the photovoltaic power generation unit, and the pumped storage power station, the energy conversion models of the hydrogen fuel cell and the electrolyzer, and the state models of the pumped storage power station, the battery pack, and the hydrogen storage tank pack established in step two are as follows: The power characteristic model of the wind power generation unit is: ; In the formula, is the rated power of the wind power generation unit, in kW; is the wind speed at the hub, m / s; is the cut-in wind speed, m / s; is the cut-out wind speed, m / s; is the rated wind speed, m / s; The power characteristic model of the photovoltaic power generation unit is: ; In the formula, is the rated output power under standard rated conditions, kW; is the solar irradiance, kW / m 2 ; is the solar radiation under standard rated conditions, kW / m 2 ; is the power temperature coefficient; is the ambient temperature, °C; is the temperature under standard rated conditions, °C; The power characteristic model of the pumped-storage power station is: ; In the formula, is the turbine efficiency; is the generator efficiency; is the water flow through the turbine, ; is the head, i.e., the water level difference before and after the water flow through the turbine, m; The energy conversion model of the electrolyzer stack is: ; In the formula, is the Faraday efficiency coefficient; is the area of a single electrolyzer; is the current of the electrolyzer stack; is the temperature of the electrolyzer stack; The energy conversion model of the hydrogen fuel cell stack is: ; In the formula, is the number of fuel cell monomers; is the current, A; is the molar mass of hydrogen; is the Faraday constant; The state model of the battery pack is: ; In the formula, is the state of charge of the battery at time is the state of charge of the battery at time and is the charge-discharge efficiency; and is the charge-discharge power; is the rated capacity of the storage battery; is the time step; The state model of the hydrogen storage tank group is: ; ; In the formula, is the amount of hydrogen in the hydrogen tank at time is the amount of hydrogen generated by the electrolyzer at time is the amount of hydrogen supplied to the fuel cell at time is the capacity of the hydrogen storage tank group, kg; The state model of the pumped-storage power station is: ; ; In the formula, is the rainfall, ; is the pumping volume, / s; is the capacity of the reservoir of the pumped-storage power station, kg.

[0010] In the preferred solution, the model predictive control algorithm in step three includes the following steps: Step 1, construct a wind resource prediction model, a light resource prediction model, and a load prediction model; Step 2, construct a prediction model for the wind-solar-hydrogen storage microgrid system; Step 3, construct the objective function and constraint conditions of the model predictive control algorithm; Step 4, input the wind resource, light resource, and load data predicted by the wind resource prediction model, the light resource prediction model, and the load prediction model into the prediction model of the wind-solar-hydrogen storage microgrid system, and obtain the optimal control sequence by optimizing the objective function under the constraints of the constraint conditions; Step 5, apply the control strategy in step three to the daily operation control of the wind-solar-hydrogen storage microgrid system with the optimal capacity.

[0011] In the preferred solution, in step 1, a time series prediction model is used to establish a wind resource prediction model, a light resource prediction model, and a load prediction model, so as to realize the function of predicting the wind resource, light resource, and load data at future moments based on the wind resource, light resource, and load data at past moments.

[0012] In the preferred solution, the time series prediction model adopted in step 1 is a long short-term memory neural network.

[0013] In the preferred solution, in step 2, the state space equation is used as the prediction model of the wind-solar-hydrogen energy storage microgrid system, and the specific expression is as follows: ; In the formula, is the load change at time , , respectively represent the maximum power generation capacities of the battery, pumped storage power station and hydrogen fuel cell.

[0014] In the preferred solution, in step 3, at time, the objective function and constraint conditions are respectively expressed as: ; ; In the formula, is the prediction time domain, is the state variable, is the control variable, is the objective function, is an index related to the state variable and control variable during the operation of the microgrid.

[0015] In the preferred solution, the optimization algorithm in step four adopts the grey wolf optimization algorithm.

[0016] A method for optimizing the capacity configuration of a wind-solar-hydrogen energy storage microgrid system based on a prediction model provided by the present invention has the following beneficial effects: 1. Improve system economy and reliability: By unifying the energy control strategy in the capacity configuration optimization process with the control strategy of the subsequent optimal capacity system, the present invention solves the problem of mismatch between capacity configuration and control strategy in the prior art, making the system operate more efficiently and effectively reducing resource waste. The optimized system can better adapt to the fluctuations of load and resources, thus improving economy and reliability.

[0017] 2. Enhance prediction ability and improve operation stability: The present invention adopts a more accurate prediction model for wind and solar resources and load, which can more effectively cope with the influence of weather changes and load fluctuations. By introducing the prediction model, the matching of energy supply and demand is optimized, the frequent adjustment during equipment operation is reduced, the power change amplitude and frequency of the equipment are reduced, thus prolonging the equipment life and improving the overall stability of the system.

[0018] 3. Improve the overall performance of the system: The present invention has made progress in the collaborative optimization of various energy forms such as wind, light, water, energy storage, and hydrogen. The scientific and reasonable capacity configuration and efficient control strategy have significantly improved the system resource utilization rate. At the same time, the flexibility and adaptability of the system operation have been enhanced, making the application of the present invention in the integrated energy management of microgrids more advantageous.

[0019] 4. Step 1 of the present invention integrates various energy and energy storage technologies such as wind power generation, photovoltaic power generation, pumped storage power stations, batteries, hydrogen fuel cells, hydrogen storage tanks, and electrolyzers, providing multi-level and multi-type energy supply and storage solutions. This architecture not only ensures the reliability of energy supply but also improves the system's adaptability to various energy fluctuations. In addition, setting reasonable constraint conditions and objective functions helps to scientifically regulate and optimize the system performance.

[0020] 5. By establishing specific power characteristic models and energy conversion models in Step 2 of the present invention, the operating characteristics and interaction relationships of each device in the system can be accurately reflected. Through the mathematical models of each subsystem such as wind power, photovoltaics, energy storage systems, and hydrogen energy conversion systems, not only can the accurate simulation of the system's dynamic behavior be achieved, but also, combined with the data of the actual operating environment, scientific predictions and optimizations of energy management in different scenarios can be made. This provides strong data support for subsequent capacity configuration and control strategy design.

[0021] 6. Step 3 of the present invention uses the model predictive control (MPC) algorithm to accurately control the operating states of each device. The advantage of this control strategy is that it can dynamically adjust the control scheme according to the predicted data and models to achieve the optimal operating efficiency. MPC can handle multiple interacting variables in the system and can adapt to load and resource fluctuations, reducing unnecessary device regulation and energy waste, thereby improving the stability and economy of the system. The flexibility and forward-looking nature of this control method enable the system to cope with complex and changing operating environments.

[0022] 7. The optimization process in Step 4 of the present invention uses an optimization algorithm based on wind and light resources and load data and uses the grey wolf optimization algorithm for solution. It can ensure the optimal capacity configuration of each device through a multi-dimensional optimization algorithm. By reasonably configuring the capacities of each device, the efficient operation of the system can be achieved, energy waste can be reduced, and at the same time, the system's adaptability under different operating conditions can be improved.

[0023] 8. Step five of the present invention ensures that the control strategy for subsequent practical applications is consistent with the control strategy adopted in the optimization process, thereby avoiding the mismatch problem between the control strategy adopted in the capacity configuration process in the traditional scheme and the control strategy in actual application. Many existing systems lack coordination between capacity configuration and control strategy, resulting in the comprehensive performance of the system failing to meet expectations during actual operation. The present invention directly applies the control strategy for obtaining the optimal capacity ratio to daily operations, so that the wind, solar, and water hydrogen storage microgrid can maintain a stable and optimized state under different operating conditions, thereby improving the economy, reliability, and energy efficiency of the system. Especially in actual operation, through precise control strategies, the system can achieve rapid response to various complex factors such as load fluctuations and resource changes, ensuring that the operation of the equipment always remains in the best state. This not only improves the long-term operating stability of the system, but also avoids the performance degradation caused by the incoordination between the control strategy and the equipment capacity, further improving the overall benefit of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The present invention will be further described below in conjunction with the accompanying drawings and embodiments: Figure 1 is a flow chart of the method of the present invention; Figure 2 This is the architecture diagram of the wind, solar, and water hydrogen storage microgrid system. DETAILED DESCRIPTION

[0025] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.

[0026] like Figure 1As shown in the figure, an optimization method for the capacity configuration of a wind-solar-hydrogen storage microgrid system based on a prediction model mainly includes four steps. The first step is to establish the architecture of the wind-solar-hydrogen storage microgrid system, and establish the wind-solar-hydrogen storage microgrid system through a wind turbine generator set, a photovoltaic generator set, a pumped storage power station, a battery bank, a hydrogen fuel cell stack, a hydrogen storage tank group, and an electrolyzer group; the second step is to establish a model of the wind-solar-hydrogen storage microgrid system, including the power characteristic models of the wind turbine generator set, the photovoltaic generator set, and the pumped storage power station, the energy conversion models of the hydrogen fuel cell stack and the electrolyzer, and the state models of the pumped storage power station, the battery bank, and the hydrogen storage tank group; the third step is to design the control strategy of the wind-solar-hydrogen storage microgrid system, and use the model predictive control algorithm to control the operating states of the battery bank, the pumped storage power station, the hydrogen fuel cell, and the electrolyzer; the fourth step is to input the wind-solar resource data and the electricity and hydrogen load data into the model established in the second step, and use an optimization algorithm to solve the model to obtain the optimal capacity ratio of each device in the wind-solar-hydrogen storage microgrid system. The fifth step is to apply the control strategy in the third step to the daily operation control of the wind-solar-hydrogen storage microgrid system with the optimal capacity.

[0027] Specifically described as follows: Step 1. Establish the architecture of the wind-solar-hydrogen storage microgrid system: Establish the wind-solar-hydrogen storage microgrid system through a wind turbine generator set, a photovoltaic generator set, a pumped storage power station, a battery bank, a hydrogen fuel cell stack, a hydrogen storage tank group, and an electrolyzer group, and construct the constraint conditions and objective functions of the wind-solar-hydrogen storage microgrid system.

[0028] The constructed architecture of the wind-solar-hydrogen storage microgrid system is as Figure 2 shown, including a wind turbine generator set, a wind turbine generator set, a photovoltaic generator set, a pumped storage power station, a battery bank, a hydrogen fuel cell stack, a hydrogen storage tank group, an electrolyzer group, and a control center, where the control center is used to control the operating states of each device.

[0029] The first objective function of the wind-solar-hydrogen storage microgrid system is the total system cost, and the expression is: ; In the formula, is the initial investment cost of the wind-solar-hydrogen storage microgrid system, is the operation and maintenance cost of the wind-solar-hydrogen storage microgrid system, is the replacement cost of the wind-solar-hydrogen storage microgrid system, , are the electricity purchase cost and the electricity sales revenue of the wind-solar-hydrogen storage microgrid system respectively.

[0030] The second objective function is the load power shortage rate, and the expression is: ; In the formula, is the power shortage at time t, is the power demand at time t.

[0031] The third objective function is the percentage of renewable energy generation, and its expression is: ; In the formula, is the power generation of the photovoltaic power generation unit at time t, is the power generation of the wind power generation unit at time t, is the power generation of the pumped-storage power station at time t; The fourth objective function is the wind and light curtailment rate , and its expression is: ; In the formula, is the wind and light curtailment amount.

[0032] The total objective function of the system is a combination of the system total cost, load shedding rate, percentage of renewable energy generation, and wind and light curtailment rate, and its expression is: ; In the formula, is the normalized system total cost; f1, f2, f3, and f4 are the weight coefficients of the system total cost, load shedding rate, percentage of renewable energy generation, and wind and light curtailment rate, respectively, and are designed according to the degree of emphasis on different objectives.

[0033] The constraint conditions of the wind-solar-hydrogen-storage microgrid system include power balance constraint, equipment number constraint, power output constraint of power sources, and equipment capacity constraint; The power balance constraint expression is: ; In the formula, , , , , , , , are the load demand power, wind turbine output, photovoltaic output, pumped-storage power station output, battery output (discharge is positive, charge is negative), electrolyzer consumption power, fuel cell output, and grid output (power purchase is positive, power sale is negative), respectively.

[0034] The power output constraint of power sources can be expressed as: ; ; ; ; ; ; wherein, , , , , are the minimum outputs of the wind power generation unit, photovoltaic power generation unit, pumped-storage power station, battery pack, and fuel cell pack, respectively, , , , , are the maximum powers of the wind power generation unit, photovoltaic power generation unit, pumped-storage power station, battery pack, and fuel cell pack, respectively, is the minimum power that the electrolyzer bank can consume, is the maximum power that the electrolyzer bank can consume, , , , , , are the maximum powers of the wind power generation unit, photovoltaic power generation unit, pumped-storage power station, battery pack, fuel cell pack, and electrolyzer bank, respectively.

[0035] Number of devices The constraint is expressed as: ; wherein, and are the upper and lower limits of the installation quantity of the devices, respectively, where represent the wind power generation unit, photovoltaic power generation unit, pumped-storage power station, battery pack, hydrogen fuel cell pack, hydrogen storage tank bank, and electrolyzer bank, respectively.

[0036] The device capacity constraint is expressed as: ; ; ; wherein, represents the percentage of the remaining battery power at time t to the maximum battery capacity; represents the percentage of the hydrogen content in the hydrogen storage tank at time t to the maximum hydrogen capacity of the hydrogen storage tank; represents the percentage of the remaining water volume in the pumped-storage power station reservoir at time t to the maximum capacity of the pumped-storage power station reservoir.

[0037] Step 2. Establish a wind-solar-hydrogen energy storage microgrid system model: including the power characteristic models of wind turbines, photovoltaic generators and pumped-storage power stations, the energy conversion models of hydrogen fuel cell stacks and electrolyzers, and the state models of pumped-storage power stations, battery packs and hydrogen storage tank groups.

[0038] The established power characteristic models of wind turbines, photovoltaic generators and pumped-storage power stations, the energy conversion models of hydrogen fuel cell stacks and electrolyzers, and the state models of pumped-storage power stations, battery packs and hydrogen storage tank groups are expressed as follows: The power characteristic model of a wind turbine is: ; In the formula, is the rated power of the wind turbine, kW; is the wind speed at the hub, m / s; is the cut-in wind speed, m / s; is the cut-out wind speed, m / s; is the rated wind speed, m / s.

[0039] The power characteristic model of a photovoltaic generator is: ; In the formula, is the rated output power under standard rated conditions, kW; is the solar irradiance, kW / m 2 ; is the solar radiation under standard rated conditions, kW / m 2 ; is the power temperature coefficient; is the ambient temperature, °C; is the temperature under standard rated conditions, °C. The power characteristic model of a pumped-storage power station is: ; In the formula, is the turbine efficiency; is the generator efficiency; is the water flow rate through the turbine, ; is the head, i.e., the water level difference before and after the water flows through the turbine, m.

[0040] The energy conversion model of the electrolyzer group is: ; In the formula, is the Faraday efficiency coefficient; is the area of a single electrolyzer; is the current of the electrolyzer group; is the temperature of the electrolyzer group.

[0041] The energy conversion model of the hydrogen fuel cell stack is: ; In the formula, is the number of fuel cell monomers; is the current, A; is the molar mass of hydrogen; is the Faraday constant.

[0042] The state model of the battery pack is: ; In the formula, is the state of charge of the battery at time is the state of charge of the battery at time and are the charge and discharge efficiencies; and are the charge and discharge powers; is the rated capacity of the battery; is the time step; The state model of the hydrogen storage tank group is: ; ; In the formula, is the amount of hydrogen in the hydrogen tank at time is the amount of hydrogen generated by the electrolyzer at time is the amount of hydrogen supplied to the fuel cell at time is the capacity of the hydrogen storage tank group, kg; The state model of the pumped-storage power station is: ; ; In the formula, is the rainfall, ; is the pumping volume, / s; is the capacity of the reservoir of the pumped-storage power station, kg.

[0043] Step 3: Design the control strategy of the wind-solar-hydrogen storage microgrid system: Use the model predictive control algorithm to control the operating states of the battery pack, pumped-storage power station, hydrogen fuel cell, and electrolyzer.

[0044] The model predictive control algorithm includes the following steps: Step 1: Construct a wind resource prediction model, a light resource prediction model, and a load prediction model.

[0045] Adopt a time series prediction model to establish a wind resource prediction model, a light resource prediction model, and a load prediction model, so as to realize the function of predicting the wind resource, light resource, and load data at future moments based on the wind resource, light resource, and load data at past moments.

[0046] In this embodiment, the time series prediction model adopted is a long short-term memory neural network (LSTM). Constructing a wind resource prediction model, a light resource prediction model, and a load prediction model through a long short-term memory neural network includes the following steps: S101: Divide the wind speed, light intensity, and load data into training sets, validation sets, and test sets respectively, and create input and output sequences through the sliding window technique. Taking the wind speed as an example, that is, predicting the wind speed at the th to to ( is an integer greater than 1) moment based on the wind speed data of the previous

[0047] moments. Similarly, the data set division of light intensity and load is also carried out according to the above operations.

[0048] S102: Design the LSTM model architecture, including an input layer, an LSTM layer, and an output layer, and set the number of units in the LSTM layer, that is, the number of hidden units.

[0049] S103: Set the optimizer, loss function, and evaluation metrics of the model.

[0050] S104: Train the LSTM model with the training set data, and update the weight and bias parameters of the model through the backpropagation algorithm to minimize the loss function. Train the LSTM model with the training set data, and update the weight and bias parameters of the model through the backpropagation algorithm combined with the Adam optimizer to minimize the cross-entropy loss function. The training process adopts batch processing, and the performance of the model on the validation set is evaluated after each iteration cycle.

[0051] S105: Verify the model with the validation set data, and adjust the model parameters according to the verification results, such as the learning rate, batch size, number of training epochs, etc.

[0052] S106: Evaluate the trained LSTM model with the test set data, and optimize the model according to the evaluation results, such as adjusting the number of LSTM layers, the number of units, optimizer parameters, etc.

[0053] Step 2: Construct a prediction model for the wind-solar-hydrogen energy storage microgrid system.

[0054] The prediction model of the model predictive control algorithm emphasizes its prediction function. The state-space equation can accurately capture the dynamic behavior of the system, handle multiple input-output variables, and adapt to nonlinear and unknown disturbances. Taking the state-space equation as an example of the prediction model, the specific expression of the prediction model for the wind-solar-hydrogen energy storage microgrid system can be expressed as follows: ; In the formula, is the load change at time , , respectively represent the maximum power generation capacities of the battery, pumped-storage power station, and hydrogen fuel cell.

[0055] Step 3: Construct the objective function and constraint conditions of the model predictive control algorithm.

[0056] At time, the objective function and constraint conditions can be respectively expressed as: ; ; In the formula, is the prediction horizon, t is time, is the state variable, is the control variable, is the objective function, is an index related to the state variable and control variable during the operation of the microgrid.

[0057] Step 4: Input the wind resource, light resource, and load data predicted by the wind resource prediction model, light resource prediction model, and load prediction model into the prediction model of the wind-solar-hydrogen energy storage microgrid system, and obtain the optimal control sequence by optimizing the objective function under the constraints of the constraint conditions.

[0058] Step 4: Input the wind and solar resource data, as well as the electricity and hydrogen load data, into the wind-solar-hydrogen energy storage microgrid system model established in Step 2, and use an optimization algorithm to solve the model to obtain the optimal capacity ratio of the wind-solar-hydrogen energy storage microgrid system.

[0059] The optimization algorithm in Step 4 emphasizes the function of reasonably allocating resources through this method to achieve the optimal efficiency. The grey wolf optimization algorithm has the characteristics of strong global search ability, fast convergence speed, and strong robustness.

[0060] The optimization algorithm in Step 4 uses the grey wolf optimization algorithm.

[0061] Taking the grey wolf optimization algorithm as an example, this invention solves the capacity configuration of the wind-solar-hydrogen-storage microgrid system through the grey wolf optimization algorithm. The solution process is as follows: Step 4.1: Set the objective function and constraints for optimizing the capacities of various devices in the wind-solar-hydrogen-storage microgrid system; Step 4.2: Set the evaluation index of the wind-solar-hydrogen-storage microgrid system and evaluate the microgrid systems composed of different device capacities; Step 4.3: Initialize the grey wolf population: Set the population size and initialize the positions of grey wolves; Step 4.4: Evaluate the fitness of grey wolves; Step 4.5: Update the positions of grey wolves: Update the positions of each grey wolf according to the social hierarchy and hunting strategies of grey wolves; Step 4.6: Calculate the current fitness of grey wolves and classify the grey wolves; Step 4.7: Check the termination condition: If the condition is met, execute Step 8.7; if the condition is not met, repeat Steps 4.5 - 4.7; Step 4.8: Output the optimal solution.

[0062] Step Five: Apply the control strategy in Step Three to the daily operation control of the wind-solar-hydrogen-storage microgrid system with the optimal capacity.

[0063] The above embodiments are only the preferred technical solutions of this invention and should not be regarded as limitations on this invention. The embodiments and the features in the embodiments in this application can be arbitrarily combined with each other without conflict. The protection scope of this invention should be the technical solutions recorded in the claims, including the equivalent replacement solutions of the technical features in the technical solutions recorded in the claims. That is, the equivalent replacement improvements within this scope are also within the protection scope of this invention.

Claims

1. A method for optimizing the capacity configuration of a wind-solar-water hydrogen storage microgrid system based on a prediction model, characterized in that: The following steps are involved: Step 1: Establish the wind-solar-water hydrogen storage microgrid system architecture: Establish a wind-solar-water hydrogen storage microgrid system through wind turbines, photovoltaic generators, pumped storage power stations, battery groups, hydrogen fuel cell groups, hydrogen storage tank groups and electrolyzer groups, and construct the constraints and objective functions of the wind-solar-water hydrogen storage microgrid system; Step 2: Establish a wind-solar-water-hydrogen storage microgrid system model: including the power characteristic model of wind turbines, photovoltaic generators and pumped storage power stations, the energy conversion model of hydrogen fuel cell groups and electrolyzers, and the state model of pumped storage power stations, battery groups and hydrogen storage tank groups; Step 3: Design the control strategy of the wind-solar-water hydrogen storage microgrid system: Use the model predictive control algorithm to control the operating status of the battery pack, pumped storage power station, hydrogen fuel cell and electrolyzer; Step 4: Bring the wind and solar resource data and the electricity and hydrogen load data into the wind, solar, and water hydrogen storage microgrid system model established in step 2, use the optimization algorithm to solve the model, and obtain the optimal capacity ratio of the wind, solar, and water hydrogen storage microgrid system; Step 5: Apply the control strategy in step 3 to the daily operation control of the wind, solar, water and hydrogen storage microgrid system with optimal capacity.

2. According to the prediction model-based wind, solar, and water hydrogen storage microgrid system capacity configuration optimization method according to claim 1, it is characterized in that: In step 1, the first objective function of the wind-solar-water hydrogen storage microgrid system is the total system cost, which is expressed as: ； In the formula, is the initial investment cost of the wind-solar-water hydrogen storage microgrid system, The operation and maintenance costs of the wind, solar, and water hydrogen storage microgrid system. The replacement cost of the wind-solar-water hydrogen storage microgrid system is , They are respectively the electricity purchase cost and electricity sales revenue of the wind, solar, water and hydrogen storage microgrid system; The second objective function is the load power failure rate, which is expressed as: ； In the formula, for The load is short of power at all times. for The load demand power at every moment; The third objective function is the percentage of renewable energy generation, expressed as: ； In the formula, for The power generation of photovoltaic generators at each moment, for The power generation of wind turbines at any moment, for The power generation of pumped storage power station at any moment; The fourth objective function is the wind and solar power abandonment rate , the expression is: ； In the formula, The amount of wind and solar power abandoned; The overall objective function of the system is a combination of the total system cost, load power shortage rate, renewable energy generation percentage, and wind and solar power abandonment rate, expressed as: ； In the formula, is the normalized total system cost; f1, f2, f3, and f4 are the weight coefficients of the total system cost, load power shortage rate, percentage of renewable energy power generation, and wind and solar power abandonment rate, respectively, and are designed according to the degree of importance attached to different objectives.

3. The method for optimizing the capacity configuration of a wind-solar-water hydrogen storage microgrid system based on a prediction model according to claim 1 is characterized in that: In the step 1, the constraints of the wind-solar-water hydrogen storage microgrid system include power balance constraints, equipment number constraints, power supply output power constraints, and equipment capacity constraints; The power balance constraint expression is: ； In the formula, , , , , , , , They are load demand power, wind turbine output, photovoltaic output, pumped storage power station output, battery output, electrolyzer power consumption, fuel cell output and grid output; The power supply output power constraint is expressed as: ； ； ； ； ； ； In the formula, , , , , are the minimum outputs of wind turbines, photovoltaic generators, pumped storage power stations, batteries and fuel cells, respectively. , , , , are the maximum power of wind turbines, photovoltaic generators, pumped storage power stations, batteries and fuel cells respectively. is the minimum power that the electrolyzer group can consume, is the maximum power that the electrolyzer group can consume, , , , , , They are the maximum power of wind turbines, photovoltaic generators, pumped storage power plants, batteries, fuel cells and electrolyzers respectively; Number of devices The constraints are expressed as: ； In the formula, and are the upper and lower limits of the number of equipment installed, respectively. They represent wind turbines, photovoltaic generators, pumped storage power plants, batteries, hydrogen fuel cells, hydrogen storage tanks and electrolyzers respectively; The equipment capacity constraint is expressed as: ； ； ； In the formula, Represents the percentage of the battery's remaining capacity at time t to the battery's maximum capacity; Represents the percentage of hydrogen content in the hydrogen storage tank at time t to the maximum hydrogen capacity of the hydrogen storage tank; It represents the percentage of the remaining water in the pumped-storage power station reservoir at time t to the maximum capacity of the pumped-storage power station reservoir.

4. The method for optimizing capacity configuration of a wind-solar-water hydrogen storage microgrid system based on a prediction model according to claim 1 is characterized in that: The power characteristic models of the wind turbine generator set, photovoltaic generator set and pumped storage power station, the energy conversion models of the hydrogen fuel cell group and the electrolyzer, and the state models of the pumped storage power station, the battery group and the hydrogen storage tank group established in step 2 are expressed as follows: The power characteristic model of the wind turbine generator set is: ； In the formula, is the rated power of the wind turbine, kW; is the wind speed at the hub, m / s; is the cut-in wind speed, m / s; is the cut-out wind speed, m / s; is the rated wind speed, m / s; The power characteristic model of the photovoltaic generator set is: ； In the formula, is the rated output power under standard rated conditions, kW; is the solar irradiance, kW / m 2 ; is the solar radiation under standard rated conditions, kW / m 2 ; is the power temperature coefficient; is the ambient temperature, °C; is the temperature under standard rated conditions, °C; The power characteristic model of the pumped storage power station is: ； In the formula, is the turbine efficiency; is the generator efficiency; is the water flow through the turbine, ; is the water head, i.e. the water level difference before and after the water flows through the turbine, m; The energy conversion model of the electrolytic cell group is: ； In the formula, is the Faraday efficiency coefficient; is the area of ​​a single electrolytic cell; is the electrolytic cell group current; is the temperature of the electrolytic cell group; The energy conversion model of the hydrogen fuel cell group is: ； In the formula, is the number of fuel cell monomers; is the current, A; is the molar mass of hydrogen; is the Faraday constant; The state model of the battery pack is: ； In the formula, for The battery's state of charge at all times; for The battery's state of charge at all times; and is the charge and discharge efficiency; and is the charge and discharge power; is the rated capacity of the battery; is the time step; The state model of the hydrogen storage tank group is: ； ； In the formula, for The amount of hydrogen in the hydrogen tank at the moment, kg; for The amount of hydrogen produced by the electrolyzer at the moment, kg; for The amount of hydrogen supplied to the fuel cell at any given moment, kg; is the capacity of the hydrogen storage tank group, kg; The state model of the pumped storage power station is: ； ； In the formula, is the rainfall, ; is the water pumping volume, / s; is the capacity of the pumped storage power station reservoir, kg.

5. The method for optimizing capacity configuration of a wind-solar-water hydrogen storage microgrid system based on a prediction model according to claim 1, characterized in that: The model predictive control algorithm in step 3 includes the following steps: Step 1: Construct wind resource prediction model, light resource prediction model, and load prediction model; Step 2: Construct a prediction model for the wind, solar, water and hydrogen storage microgrid system; Step 3: Construct the objective function and constraints of the model predictive control algorithm; Step 4: The wind resource, light resource and load data predicted by the wind resource prediction model, light resource prediction model and load prediction model are input into the wind-solar-water-hydrogen storage microgrid system prediction model, and the optimal control sequence is obtained by optimizing the objective function under the constraints; Step 5: Apply the control strategy in step 3 to the daily operation control of the wind-solar-water-hydrogen storage microgrid system with optimal capacity.

6. The method for optimizing the capacity configuration of a wind-solar-water hydrogen storage microgrid system based on a prediction model according to claim 5 is characterized in that: In step 1, a time series prediction model is used to establish a wind resource prediction model, a light resource prediction model and a load prediction model, so as to realize the function of predicting the wind resource, light resource and load data at future moments based on the wind resource, light resource and load data at past moments.

7. The method for optimizing capacity configuration of a wind-solar-water hydrogen storage microgrid system based on a prediction model according to claim 6 is characterized in that: The time series prediction model adopted in step 1 is a long short-term memory neural network.

8. The method for optimizing capacity configuration of a wind-solar-water hydrogen storage microgrid system based on a prediction model according to claim 5 is characterized in that: In step 2, the state space equation is used as the prediction model of the wind-solar-water hydrogen storage microgrid system, and the specific expression is as follows: ； In the formula, for The load variation at each moment, , , They represent the maximum power generation capacity of batteries, pumped storage power stations and hydrogen fuel cells respectively.

9. The method for optimizing capacity configuration of a wind-solar-water hydrogen storage microgrid system based on a prediction model according to claim 5 is characterized in that: In step 3, At this moment, the objective function and constraints are expressed as: ； ； In the formula, For the prediction time domain, is the state variable, is the control variable, is the objective function, It is an indicator related to state variables and control variables during the operation of the microgrid.

10. A method for optimizing capacity configuration of a wind-solar-water hydrogen storage microgrid system based on a prediction model according to claim 1, characterized in that: The optimization algorithm in step 4 adopts the Grey Wolf Optimization Algorithm.

Images