Energy balance optimization method and equipment for wind power off-grid energy storage hydrogen production system

Abstract

The present invention discloses an energy balance optimization method and equipment for an off-grid wind power energy storage hydrogen production system, which relates to the technical field related to wind power optimization, including: obtaining wind power factors within a predetermined time zone, predicting power generation based on the wind power factors, and obtaining predicted output power; monitoring and obtaining the current energy storage capacity of the starting node of the predetermined time zone; taking optimization restriction conditions as constraints, taking maximum hydrogen production, minimum hydrogen production fluctuation and minimum energy storage aging as optimization purposes, and performing multi-stage optimization of charging and discharging strategies and electrolyzer workloads based on predicted output power and current energy storage capacity, and outputting the optimal energy balance control plan; executing energy storage hydrogen production operations in the predetermined time zone according to the optimal energy balance control plan. The technical problems existing in the prior art of large volatility of wind power generation, uneven utilization of energy storage capacity, and strong volatility of hydrogen production process are solved, and the technical effects of increasing hydrogen production, reducing hydrogen production fluctuations, and increasing the service life of energy storage hydrogen production systems are achieved.

Description

Technical Field

[0001] The present application relates to the technical field related to wind power optimization, and specifically to an energy balance optimization method and equipment for an off-grid wind power energy storage and hydrogen production system. Background Art

[0002] Among many clean energy technologies, wind energy, as a widely used and renewable energy, has invested a lot of resources in energy production. The intermittent and unstable characteristics of wind energy make its integration in traditional power grids face certain challenges, especially when combined with energy storage systems. How to effectively balance and optimize energy becomes a key link. However, in the actual application of wind power off-grid energy storage hydrogen production system, the volatility and uncertainty of wind power generation bring difficulties to energy storage and conversion. It is difficult to predict the power generation capacity of wind power and reasonably arrange the charging and discharging process of the energy storage system. In addition, the production process of hydrogen not only depends on the stable output of wind power, but also needs to consider the capacity of energy storage equipment, aging effects and other factors. It is difficult to balance the output power of wind power, energy storage capacity and the load of the electrolyzer, which makes the system operation unstable, thereby reducing the hydrogen production, the life of the energy storage system and the stability of production.

[0003] Therefore, in the current relevant technologies, there are technical problems such as large volatility in wind power generation, uneven utilization of energy storage capacity, and strong volatility in the hydrogen production process. Summary of the invention

[0004] This application solves the technical problems existing in the prior art of large volatility in wind power generation, uneven utilization of energy storage capacity, and strong volatility in the hydrogen production process by providing an energy balance optimization method and equipment for an off-grid wind power energy storage hydrogen production system, thereby achieving the technical effects of increasing hydrogen production, reducing hydrogen production fluctuations, and increasing the service life of the energy storage hydrogen production system.

[0005] The present application provides an energy balance optimization method for an off-grid wind power energy storage hydrogen production system, comprising: obtaining a wind power factor within a predetermined time zone, performing power generation prediction according to the wind power factor, and obtaining a predicted output power; monitoring and obtaining the current energy storage capacity of a starting node of the predetermined time zone; configuring an optimization restriction condition based on the off-grid wind power energy storage hydrogen production system, taking the optimization restriction condition as a constraint, taking maximum hydrogen production, minimum hydrogen production fluctuation, and minimum energy storage aging as optimization purposes, performing multi-stage optimization of charging and discharging strategies and electrolyzer workloads according to the predicted output power and current energy storage capacity, and outputting an optimal energy balance control plan; and executing the energy storage hydrogen production operation in the predetermined time zone according to the optimal energy balance control plan.

[0006] In a possible implementation, power generation prediction is performed according to the wind power factors to obtain the predicted output power, and the following processing is also performed: query the wind power generation log, collect sample wind power factor sets and sample output power sets; configure N prediction operators, wherein the prediction operators include at least long short-term memory networks, BP neural networks and random forests, and N is an integer greater than or equal to 3; based on historical data, the N prediction accuracies of the N prediction operators are counted, and N prediction scales are configured according to the N prediction accuracies, wherein the prediction scale is the number of branches and is positively correlated with the prediction accuracy; the sample wind power factor set and the sample output power set are used as training data and divided into N equal parts to obtain N training sets; using the N training sets, based on the N prediction scales, the N prediction operators are supervised and trained until convergence, and N prediction branches are integrated to construct a wind power generation prediction plug-in; using the wind power generation prediction plug-in, power generation prediction is performed according to the wind power factors to obtain the predicted output power.

[0007] In a possible implementation, the current energy storage capacity of the starting node in the predetermined time zone is monitored and obtained, and the following processing is also performed: a tolerance interval for energy storage capacity acquisition is configured, wherein the tolerance interval is set based on the duration of the scheme optimization; the starting node is advanced according to the tolerance interval, a collection node is determined, and the energy storage capacity of the collection node is monitored and obtained and set as the current energy storage capacity.

[0008] In a possible implementation, the energy balance optimization method of the wind power off-grid energy storage and hydrogen production system also performs the following processing: the optimization constraints include the energy storage capacity threshold value, the hydrogen production efficiency threshold value and the off-grid balance constraints, wherein the off-grid balance constraints include the frequency constraints and voltage constraints of the off-grid system.

[0009] In a possible implementation, with the optimization restriction as a constraint, with the maximum hydrogen production, minimum hydrogen production fluctuation and minimum energy storage aging as the optimization purpose, according to the predicted output power and the current energy storage capacity, multi-stage optimization of the charging and discharging strategy and the electrolyzer workload is performed, and the optimal energy balance control scheme is output, and the following processing is also performed: based on the maximum hydrogen production, minimum hydrogen production fluctuation and minimum energy storage aging, the optimization evaluation function is configured; in a three-dimensional simulation space, the wind power off-grid energy storage hydrogen production system is modeled to generate an energy storage hydrogen production simulation space, and the predicted output power and the current energy storage capacity are rendered to the energy storage hydrogen production simulation space; based on the optimization evaluation function, with the optimization restriction as a constraint, the energy storage hydrogen production simulation space is used to perform multi-stage optimization of the charging and discharging strategy and the electrolyzer workload, and multiple control time zones, multiple optimal charging and discharging strategies and multiple optimal hydrogen production currents are output; according to the order of the control time zones, the multiple control time zones, multiple optimal charging and discharging strategies and multiple optimal hydrogen production currents are integrated to generate the optimal energy balance control scheme.

[0010] In a possible implementation, the energy balance optimization method of the wind power off-grid energy storage hydrogen production system further performs the following processing: the expression of the optimization evaluation function is:

[0011] ;

[0012] in, To evaluate fitness, is the hydrogen production weight, is the hydrogen production fluctuation weight, is the energy storage aging weight, is the hydrogen production, is the hydrogen production fluctuation coefficient, is the charge and discharge rate weight, is the weight of the number of charge and discharge times, S is the mean of the charge and discharge rates, and C is the total number of charge and discharge times; wherein, the hydrogen production rate deviations of adjacent time zones within multiple hydrogen production time zones are calculated, and the mean of multiple hydrogen production rate deviations is calculated to obtain the hydrogen production fluctuation coefficient.

[0013] In a possible implementation, based on the optimization evaluation function and with the optimization restriction as a constraint, the energy storage hydrogen production simulation space is used to perform multi-stage optimization of the charging and discharging strategy and the electrolyzer workload, and multiple control time zones, multiple optimal charging and discharging strategies and multiple optimal hydrogen production currents are output. The following processing is also performed: the charging and discharging regulation space and the workload regulation space of the wind power off-grid energy storage hydrogen production system are obtained, wherein the charging and discharging regulation is the charging power / discharging power, and the workload is the hydrogen production current; according to the predicted output power configuration stage time interval, the predetermined time zone is divided into multiple control time zones according to the stage time interval, wherein the stage time interval is negatively correlated with the predicted output power, and any parameter is randomly selected in the charging and discharging regulation space and the workload regulation space for combination to obtain multiple Initial solutions, wherein each initial solution includes a charging and discharging strategy and a hydrogen production current; the multiple initial solutions are input into the energy storage and hydrogen production simulation space for simulation, and the multiple simulation results are evaluated according to the optimization evaluation function, and multiple fitness levels are output; with the optimization restriction conditions as constraints, based on the multiple fitness levels, stage optimization in the first control time zone is performed until convergence, and a first optimal charging and discharging strategy and a first hydrogen production current are output; a first energy storage capacity is determined based on the first optimal charging and discharging strategy and the first hydrogen production current, and the energy storage and hydrogen production simulation space is rendered and updated according to the first energy storage capacity, stage optimization in the second control time zone is performed, and iterative optimization is performed until the multi-stage optimization in the predetermined time zone is completed, and multiple control time zones, multiple optimal charging and discharging strategies and multiple optimal hydrogen production currents are output.

[0014] In a possible implementation, with the optimization restriction condition as a constraint, based on the multiple fitnesses, the stage optimization of the first control time zone is performed until convergence, and the first optimal charging and discharging strategy and the first hydrogen production current are output, and the following processing is also performed: based on the multiple fitnesses, the multiple initial solutions are arranged in descending order of fitness to generate an initial solution sequence, and the first Q solutions of the initial solution sequence are set as optimal solutions, and the last T solutions are set as inferior solutions, where T is an integer multiple of Q; with the Q optimal solutions as the center, the T inferior solutions are randomly clustered to obtain Q solution thresholds, and the solution with the smallest fitness within the Q solution thresholds is set as a differential solution, where the number of inferior solutions within each solution threshold is The quantity is the same; within the Q solution thresholds, the optimization direction is to approach the optimal solution and to move away from the poor solution, and the inferior solution and the poor solution within the solution threshold are adjusted according to a predetermined adjustment step to obtain Q updated solution thresholds; a penalty mechanism is set based on the optimization constraint condition, wherein, if the updated inferior solution or poor solution does not meet the optimization constraint condition, its fitness is halved; the Q updated solution thresholds are optimized according to the penalty mechanism, and the iterative optimization is continued until the predetermined number of optimization times is met, Q current solution thresholds are output, and the solution threshold with the largest sum of fitness is set as the optimal solution threshold, and the optimal solution within the optimal solution threshold is output to obtain the first optimal charging and discharging strategy and the first hydrogen production current.

[0015] In a possible implementation, a predetermined adjustment step is configured, and the following processing is performed: taking the fitness of the best solution within the solution threshold as a benchmark, fitness deviations are calculated for multiple inferior solutions to determine multiple fitness deviations; multiple adjustment compensation coefficients are determined based on the multiple fitness deviations, wherein the adjustment compensation coefficient and the fitness deviation are positively correlated; the initial adjustment step is optimized based on the multiple adjustment compensation coefficients to obtain multiple optimized adjustment steps to construct the predetermined adjustment step.

[0016] The present application also provides an electronic device, comprising: a memory for storing executable instructions; and a processor for implementing an energy balance optimization method for an off-grid wind power energy storage and hydrogen production system when executing the executable instructions stored in the memory.

[0017] The energy balance optimization method and equipment of the wind power off-grid energy storage hydrogen production system proposed in this application is intended to obtain the wind power factor in the predetermined time zone, perform power generation prediction based on the wind power factor, and obtain the predicted output power; monitor and obtain the current energy storage capacity of the starting node in the predetermined time zone; with the optimization restriction conditions as constraints, the maximum hydrogen production, the minimum hydrogen production fluctuation and the minimum energy storage aging as the optimization purpose, according to the predicted output power and the current energy storage capacity, perform multi-stage optimization of the charging and discharging strategy and the electrolyzer workload, and output the optimal energy balance control plan; according to the optimal energy balance control plan, perform the energy storage hydrogen production operation in the predetermined time zone. It solves the technical problems existing in the prior art of large volatility of wind power generation, uneven utilization of energy storage capacity, and strong volatility of the hydrogen production process, and achieves the technical effects of increasing hydrogen production, reducing hydrogen production fluctuations, and increasing the service life of the energy storage hydrogen production system. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solution of the embodiment of the present disclosure, the accompanying drawings of the embodiment of the present disclosure will be briefly introduced below. A flow chart is used in the present application to illustrate the operations performed by the system according to the embodiment of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps can be processed in reverse order or simultaneously as needed. At the same time, other operations can also be added to these processes, or one or more operations can be removed from these processes.

[0019] Figure 1 A schematic flow chart of an energy balance optimization method for an off-grid wind power energy storage hydrogen production system provided in an embodiment of the present application;

[0020] Figure 2 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0021] Description of the reference numerals: input device 401 , processor 402 , memory 403 , output device 404 . DETAILED DESCRIPTION

[0022] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below.

[0023] In order to make the objectives, technical solutions and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings. The described embodiments should not be regarded as limiting the present application. All other embodiments obtained by ordinary technicians in the field without making creative work are within the scope of protection of this application.

[0024] In the following description, reference is made to "some embodiments", which describe a subset of all possible embodiments, but it is understood that "some embodiments" may be the same subset or different subsets of all possible embodiments, and may be combined with each other without conflict, and the terms "first\second" involved are merely to distinguish similar objects and do not represent a specific ordering of objects. The terms "including" and "having" and any variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or inherent to these processes, methods, products, or devices. Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those generally understood by technicians in the technical field of this application. The terms used herein are for the purpose of describing the embodiments of the present application only.

[0025] The present application embodiment provides an energy balance optimization method for a wind power off-grid energy storage hydrogen production system, such as Figure 1 As shown, the method includes:

[0026] Step S100, obtaining a wind power factor in a predetermined time zone, performing power generation prediction according to the wind power factor, and obtaining a predicted output power.

[0027] Preferably, in the wind power off-grid energy storage hydrogen production system, by analyzing and predicting the wind power factors in a specific time period, the power generation capacity of the wind power system in the time period is estimated, that is, the predicted output power is obtained, wherein the wind power factor refers to the key parameters that affect the power generation capacity of the wind turbine, which may include meteorological data such as wind speed, wind direction, temperature, air pressure, and even the geographical and environmental characteristics of the wind farm. The wind power factor is the main factor affecting the efficiency and output power of the wind turbine. The predetermined time zone refers to a time period, such as a period of several hours, a day or a longer period in the future. The wind power factor is used to predict the power generation capacity of the wind power system during the time period. Specifically, according to the obtained wind power factors (such as wind speed, wind direction, etc.), a wind power prediction model (such as regression analysis based on historical data, machine learning algorithm or other meteorological prediction model) is used to predict the wind power generation in the predetermined time zone, that is, by simulating the working condition of the wind turbine, combining meteorological forecasts and equipment characteristics, an estimated value of the wind power output power in the time period is obtained, and the predicted output power is obtained, that is, the power output that is expected to be obtained from the wind turbine in the predetermined time zone.

[0028] Further, step S100 also includes step S110, querying the wind power generation log, collecting a sample wind power factor set and a sample output power set; step S120, configuring N prediction operators, wherein the prediction operators include at least a long short-term memory network, a BP neural network and a random forest, and N is an integer greater than or equal to 3; step S130, based on historical data, counting the N prediction accuracies of the N prediction operators, and configuring N prediction scales according to the N prediction accuracies, wherein the prediction scale is the number of branches, and is positively correlated with the prediction accuracy; step S140, using the sample wind power factor set and the sample output power set as training data, and dividing them into N equal parts to obtain N training sets; step S150, using the N training sets, based on the N prediction scales, supervised training is performed on the N prediction operators until convergence, and N prediction branches are integrated to construct a wind power generation prediction plug-in; step S160, using the wind power generation prediction plug-in, power generation prediction is performed according to the wind power factors to obtain the predicted output power.

[0029] Preferably, the wind power generation log is queried to collect a sample wind power factor set and a sample output power set, wherein the wind power factor set includes various meteorological data that affect wind power generation (such as wind speed, wind direction, air pressure, temperature, etc.), and the output power set includes the actual output power data of the wind turbine under a given wind power factor, indicating the actual amount of electricity generated by the wind turbine in a specific period of time. N prediction operators are configured, and the prediction operators include at least a long short-term memory network (a deep learning algorithm suitable for processing time series data and capable of capturing the temporal variation of factors such as wind speed), a BP neural network (a back propagation neural network, a feedforward neural network trained by a back propagation algorithm for nonlinear prediction), and a random forest (an integrated learning method that performs prediction by constructing multiple decision trees and is suitable for processing complex, nonlinear regression problems). N is an integer greater than or equal to 3, indicating that at least three prediction operators are used for integrated prediction. Different algorithms have different advantages. By combining multiple models, the accuracy and robustness of the prediction can be improved.

[0030] Preferably, each prediction operator (such as LSTM, BP neural network, random forest) is trained and its prediction accuracy is counted, that is, the performance of each algorithm on historical data is measured, and then N prediction scales are configured according to N prediction accuracy. Specifically, each prediction operator will have a prediction scale, which is usually expressed as the number of branches or the number of trees, etc. The number of branches trained by each operator (including multiple prediction units, the number of prediction units) is greater, the more configurations are made, and the greater the weight of the operator when weighted. For example, if a model (such as LSTM) has a higher prediction accuracy, then the prediction scale (number of branches) of the model will be larger, otherwise, the scale will be smaller; then the sample data set is divided into N equal parts, and the sample wind power factor set and output power set are divided into N parts according to the N prediction operators. The number is equally divided into N training sets, each training set is used to train a prediction operator, and each training set is randomly extracted from the original data to ensure that each operator can learn different data features. Then, each prediction operator is trained with a given training set and prediction scale, so that each model predicts the output power according to the input wind power factor until the prediction error of the model converges, that is, a stable prediction accuracy is reached. After training, each prediction operator will obtain a prediction branch, that is, N prediction branches are obtained. These prediction branches are integrated into a wind power prediction plug-in, which can predict power generation according to wind power factors, that is, integrate the prediction results of multiple prediction branches to generate more accurate and stable wind power generation prediction values, that is, obtain predicted output power.

[0031] Step S200, monitoring and obtaining the current energy storage capacity of the starting node in the predetermined time zone.

[0032] Preferably, the starting node of the predetermined time zone refers to the starting moment of the time period, that is, the moment when monitoring starts. Specifically, the current energy storage capacity, that is, the amount of electrical energy stored in the energy storage system, is obtained through sensors, monitoring systems or intelligent battery management systems (BMS), usually in kilowatt-hours (kWh) or megawatt-hours (MWh), indicating the amount of electrical energy currently stored in the energy storage system and the remaining available capacity of the energy storage device, to ensure that the energy storage system can provide the required energy in the event of power shortage, thereby avoiding affecting the hydrogen production process due to insufficient energy storage.

[0033] Furthermore, step S200 also includes step S210, configuring a tolerance interval for energy storage capacity collection, wherein the tolerance interval is set based on a solution optimization duration; step S220, prepending the starting node according to the tolerance interval, determining a collection node, and monitoring and obtaining the energy storage capacity of the collection node as the current energy storage capacity.

[0034] Preferably, the tolerance interval for energy storage capacity acquisition is configured based on the duration of the optimal solution, that is, the allowable fluctuation range when the energy storage system capacity is adjusted is set, which reflects the range of variation of the energy storage capacity within a certain period of time. If the optimization calculation duration is long, a larger energy storage capacity fluctuation may be allowed, because the system adjustment may be more flexible within a longer time span; and for a shorter duration, the tolerance interval may be set smaller to ensure the accurate scheduling of the energy storage system; before monitoring and scheduling the energy storage capacity, the initial state of the energy storage capacity is adjusted or calculated in advance according to the setting of the tolerance interval. For example, during the pre-operation, the energy storage system estimates and pre-positions the starting node capacity of the energy storage system according to the fluctuation range to ensure that the initial capacity of the energy storage system can meet the optimization goal, and in subsequent operations, the subsequent energy scheduling and hydrogen production will not be affected by insufficient or excessive capacity, and then the acquisition nodes are determined, that is, the specific moments within a certain period of time when the energy storage capacity needs to be monitored. Through these nodes, the capacity data of the energy storage system can be obtained in real time to ensure that the system can dynamically adjust the charging and discharging strategy of the energy storage, and the capacity of the energy storage system is monitored in real time at these acquisition nodes, that is, the current energy storage capacity (the current state of charge).

[0035] Step S300, based on the optimization restriction conditions configured for the wind power off-grid energy storage hydrogen production system, with the optimization restriction conditions as constraints, with maximum hydrogen production, minimum hydrogen production fluctuation and minimum energy storage aging as optimization objectives, according to the predicted output power and current energy storage capacity, multi-stage optimization of the charging and discharging strategy and the electrolyzer workload is performed, and the optimal energy balance control plan is output.

[0036] Preferably, the optimization constraint conditions for the configuration of the off-grid wind power energy storage hydrogen production system refer to the setting of various technical parameters and operating conditions when optimizing the system, mainly including the constraints on multiple links such as wind power output, energy storage equipment, and electrolyzer operation. Specifically, they may include wind power output restrictions, setting the maximum power generation capacity of the wind turbine according to actual wind speed, wind direction and other conditions; energy storage capacity restrictions, the energy storage system has a certain charging and discharging range, and these restrictions will affect the service life and efficiency of the energy storage system; electrolyzer load restrictions, the working load of the electrolyzer has a maximum power limit within a certain period of time, and too high or too low will affect the production or quality of hydrogen; energy storage equipment health and aging restrictions, the energy storage equipment will age during the charging and discharging process, and excessive charging and discharging will accelerate the attenuation of the battery, and such restrictions need to be added during the optimization process.

[0037] Preferably, the optimization objectives are maximum hydrogen output, minimum hydrogen production fluctuation and minimum energy storage aging, that is, the hydrogen production is maximized by optimizing the coordination between wind power output and energy storage system. The optimization process needs to balance the wind power output fluctuation in different time periods to minimize the fluctuation of hydrogen production, and take measures to avoid frequent charging and discharging of the energy storage system, reduce its aging rate, and thus extend the service life of the energy storage equipment. According to the obtained predicted output power and current energy storage capacity, multi-stage optimization of charging and discharging strategy and electrolyzer workload is performed. Among them, the wind power output power is not constant, and the energy storage system needs to adjust the energy by charging (storing excess electrical energy) and discharging (releasing electrical energy when wind power is insufficient). The electrolyzer is the core equipment for converting electrical energy into hydrogen, and its workload directly determines the hydrogen output.

[0038] Preferably, according to the wind power output forecast and the energy storage capacity monitoring results, a suitable charging and discharging strategy is set to ensure the energy balance of the energy storage system at different time points, for example, charging is performed when there is excess wind power, and discharging is performed when there is insufficient wind power to support hydrogen production, and the workload of the electrolyzer is reasonably adjusted according to the wind power forecast and the charging and discharging conditions of the energy storage system to maintain the stability and efficiency of hydrogen production; multi-stage optimization refers to the optimization process being divided into multiple stages, each stage being adjusted and optimized based on the results of the previous stage, for example, in the first stage, a rough charging and discharging strategy is determined based on the predicted wind power output and initial energy storage capacity, in the second stage, the charging and discharging strategy is adjusted according to the changes in the workload and energy storage capacity of the electrolyzer, and in the third stage, the previous strategy is adjusted according to the actual operating conditions (such as the actual situation of wind power generation), and finally the optimal energy balance control scheme is obtained to ensure the stability of hydrogen production, maximize output, and minimize system losses and aging, thereby achieving optimal energy utilization and minimum operating costs, so that the entire system operates efficiently and stably.

[0039] Furthermore, step S300 also includes that the optimization constraint conditions include a storage capacity threshold value, a hydrogen production efficiency threshold value and an off-grid balance constraint, wherein the off-grid balance constraint includes a frequency constraint and a voltage constraint of the off-grid system.

[0040] Preferably, the energy storage capacity threshold value refers to the maximum and minimum power capacities that the energy storage system can withstand during operation, that is, the charging state of the energy storage system cannot be lower than a certain minimum value, nor higher than a certain maximum value, so as to prevent the energy storage device from being overcharged or over-discharged, thereby ensuring the safety and long-term operation stability of the equipment. For example, some energy storage devices (such as batteries) have a certain power range. If this range is exceeded, the equipment may be damaged or aged; the hydrogen production efficiency threshold value refers to the minimum efficiency requirement of the water electrolysis hydrogen production technology in the hydrogen production process, which defines the minimum efficiency of the hydrogen production process. A hydrogen production process below this efficiency will be deemed to be inconsistent with the system optimization goal. Among them, the hydrogen production efficiency is usually determined by the operating efficiency of the electrolyzer. If the efficiency of the electrolyzer is too low (for example, due to excessive load or problems with the electrolyzer itself), the hydrogen output will be reduced and it will lead to excessive power consumption. Setting the hydrogen production efficiency threshold value can ensure that the system only produces hydrogen when the efficiency meets the requirements.

[0041] Preferably, the off-grid balance constraint is to ensure the stable operation of the system, especially in the absence of external grid support, the energy storage system, wind turbines and hydrogen production equipment can work independently and in a balanced manner, mainly including off-grid system frequency constraints and voltage constraints. Specifically, the grid frequency in the off-grid system needs to be maintained within the specified range, and the energy storage system can respond quickly to changes in grid frequency (for example, increasing grid power by discharging when the frequency drops, and reducing grid load by charging when the frequency rises); voltage stability is also an important indicator of grid stability. Voltage changes may come from fluctuations in wind power generation, charging and discharging of the energy storage system, or electricity demand for hydrogen production. The energy storage system and wind turbines need to work together to ensure that the grid voltage is within a reasonable range and that the voltage is within the specified range to ensure safe and efficient operation of the system.

[0042] Furthermore, step S300 also includes step S310, configuring an optimization evaluation function based on the maximum hydrogen production, minimum hydrogen production fluctuation and minimum energy storage aging; step S320, modeling the wind power off-grid energy storage hydrogen production system in a three-dimensional simulation space, generating an energy storage hydrogen production simulation space, and rendering the predicted output power and current energy storage capacity to the energy storage hydrogen production simulation space; step S330, based on the optimization evaluation function, with the optimization restriction conditions as constraints, using the energy storage hydrogen production simulation space, performing multi-stage optimization of charging and discharging strategies and electrolyzer workloads, and outputting multiple control time zones, multiple optimal charging and discharging strategies and multiple optimal hydrogen production currents; step S340, integrating the multiple control time zones, multiple optimal charging and discharging strategies and multiple optimal hydrogen production currents in order of the control time zones to generate the optimal energy balance control scheme.

[0043] Preferably, an optimization evaluation function is configured based on maximum hydrogen production, minimum hydrogen production fluctuation and minimum energy storage aging. Specifically, maximum hydrogen production refers to maximizing hydrogen production within a given time period by effectively scheduling wind power output, energy storage capacity and electrolyzer load; hydrogen production with large fluctuations will not only affect energy utilization efficiency, but also may lead to increased volatility of the energy storage system. The deviations of adjacent hydrogen production rates in multiple time periods (i.e., the degree of change in hydrogen production rate between different time periods) are calculated, and the fluctuation coefficient is determined based on the mean of these deviations. Hydrogen production with large fluctuations will not only affect energy utilization efficiency, but also may lead to increased volatility of the energy storage system; the aging of energy storage equipment will affect the life and efficiency of the energy storage equipment. By controlling the frequency and amplitude of charging and discharging, the operation mode of the energy storage equipment is optimized, the aging of the energy storage equipment is minimized, and its service life is extended; combining these three optimization objectives, an optimization evaluation function is established.

[0044] Preferably, the wind power off-grid energy storage and hydrogen production system is modeled in a three-dimensional simulation space. Specifically, by integrating the energy storage system, wind power output and hydrogen production models, a simulation space is generated that can reflect the system operation status in real time. The user can visualize the changes in wind power output power and energy storage capacity, and observe the coordinated work of various parts of the system. The predicted output power and current energy storage capacity are then rendered to the energy storage and hydrogen production simulation space. That is, in the simulation space, the real-time wind power output power and current energy storage capacity data are dynamically displayed, and the changes in various parameters of energy storage and hydrogen production can be clearly seen through the visualization effect of the simulation space.

[0045] Preferably, based on the optimization evaluation function and with optimization constraints as constraints, multi-stage optimization is performed. The optimization of each stage will be adjusted according to the results of the previous stage to ensure that the whole process gradually converges to the optimal solution. Specifically, the core of the optimization lies in the reasonable arrangement of the charging and discharging operations of the energy storage system and the workload of the electrolyzer. The charging and discharging strategy determines the power dispatch of the energy storage system, and the electrolyzer load directly affects the hydrogen production. Through multi-stage optimization, the charging and discharging strategy and the electrolyzer load are continuously adjusted to ensure that all strategies and control operations are carried out under the optimization constraints, to ensure the safe and stable operation of the system, to maximize the final hydrogen production, and to control volatility and energy storage aging; and then output multiple control time zones, multiple optimal charging and discharging strategies and multiple optimal hydrogen production currents, where The control time zone refers to the different time periods divided during the optimization process. Specific charging and discharging strategies and electrolyzer loads are implemented in each time zone to ensure that hydrogen production in this period is maximized and fluctuations are minimized. The optimal charging and discharging strategy is the strategy output for each control time zone. The optimal hydrogen production current refers to the optimal current level that the electrolyzer should maintain in each control time zone. The current directly affects the efficiency of the water electrolysis reaction, and thus affects the hydrogen production. Finally, multiple control time zones, multiple optimal charging and discharging strategies, and multiple optimal hydrogen production currents are integrated to obtain the optimal energy balance control solution, ensuring that the balance between the energy storage system and hydrogen production is best reflected in different time periods, thereby ensuring that the wind power off-grid energy storage hydrogen production system can achieve the best hydrogen production, stability, and equipment life in long-term operation.

[0046] Furthermore, step S310 also includes that the expression of the optimization evaluation function is:

[0047] ;

[0048] in, To evaluate fitness, is the hydrogen production weight, is the hydrogen production fluctuation weight, is the energy storage aging weight, is the hydrogen production, is the hydrogen production fluctuation coefficient, is the charge and discharge rate weight, is the weight of the number of charge and discharge times, S is the mean of the charge and discharge rates, and C is the total number of charge and discharge times; wherein, the hydrogen production rate deviations of adjacent time zones within multiple hydrogen production time zones are calculated, and the mean of multiple hydrogen production rate deviations is calculated to obtain the hydrogen production fluctuation coefficient.

[0049] Preferably, in the optimization process of the wind power off-grid energy storage hydrogen production system, in order to ensure the stability of the hydrogen production process, the hydrogen production rate deviations of adjacent time zones within multiple time periods (time zones) are calculated and their averages are calculated to obtain the hydrogen production fluctuation coefficient, wherein the hydrogen production rate refers to the rate at which hydrogen is produced when the electrolyzer performs water electrolysis reaction within a certain period of time, usually expressed as the hydrogen output per unit time, and the hydrogen production rate deviation refers to the amplitude of change of the hydrogen production rate between consecutive adjacent time zones. The hydrogen production rate deviations in multiple time zones are averaged to obtain the overall deviation as the hydrogen production fluctuation coefficient, which is used to quantify the volatility of the hydrogen production rate. If the fluctuation coefficient is low, it means that hydrogen production is more stable and the fluctuation is small; and if the fluctuation coefficient is high, it means that the production is more unstable, and it may be necessary to optimize the system scheduling to reduce the fluctuation.

[0050] Furthermore, step S330 also includes step S331, obtaining the charge and discharge regulation space and workload regulation space of the wind power off-grid energy storage hydrogen production system, wherein the charge and discharge regulation is charging power / discharging power, and the workload is the hydrogen production current; step S332, configuring the stage time interval according to the predicted output power, dividing the predetermined time zone into multiple control time zones according to the stage time interval, wherein the stage time interval is negatively correlated with the predicted output power, step S333, randomly selecting any parameter in the charge and discharge regulation space and the workload regulation space for combination, to obtain multiple initial solutions, wherein each initial solution includes a charge and discharge strategy and a hydrogen production current; step S334, inputting the multiple initial solutions into the storage The hydrogen production simulation space is simulated, and multiple simulation results are evaluated according to the optimization evaluation function, and multiple fitness levels are output; step S335, with the optimization restriction condition as a constraint, based on the multiple fitness levels, stage optimization of the first control time zone is performed until convergence, and a first optimal charging and discharging strategy and a first hydrogen production current are output; step S336, based on the first optimal charging and discharging strategy and the first hydrogen production current, a first energy storage capacity is determined, and the energy storage and hydrogen production simulation space is rendered and updated according to the first energy storage capacity, stage optimization of the second control time zone is performed, and iterative optimization is performed until the multi-stage optimization of the predetermined time zone is completed, and multiple control time zones, multiple optimal charging and discharging strategies and multiple optimal hydrogen production currents are output.

[0051] Preferably, the charge and discharge regulation space and the workload regulation space of the wind power off-grid energy storage hydrogen production system are obtained, wherein the charge and discharge regulation space refers to the range of adjustable charging power and discharging power of the energy storage system, the charging and discharging power is the key parameter for regulating the energy storage system, and the workload regulation space refers to the adjustable working current range of the electrolyzer during the hydrogen production process, and the current of the electrolyzer determines the rate of the water electrolysis reaction and directly affects the hydrogen production; then, according to the predicted output power configuration stage time interval, the predetermined time zone is divided into multiple control time zones, and in each control time zone, a specific charge and discharge strategy and hydrogen production current are executed, The size of the time interval depends on the accuracy and volatility of the wind power output forecast. The stage time interval is negatively correlated with the predicted output power, which means that if the predicted wind power output power is more volatile, a shorter time interval is selected for more frequent adjustments to cope with changes in wind power output; when the predicted output power is relatively stable, the stage time interval is relatively long; according to the setting of the stage time interval, the predetermined time zone (such as a day or a week) is divided into multiple smaller time zones. In each control time zone, the charging and discharging strategy and the selection of hydrogen production current will be determined based on the wind power forecast and energy storage capacity to achieve optimal hydrogen production and system efficiency.

[0052] Preferably, a combination of charge and discharge strategies (charge and discharge power) and hydrogen production current is randomly selected in the charge and discharge regulation space and the workload regulation space to generate multiple initial solutions, each of which includes a charge and discharge strategy and a hydrogen production current. Then, the multiple initial solutions are input into the energy storage hydrogen production simulation space for simulation. The simulation space simulates the operating state of the entire wind power off-grid energy storage hydrogen production system, and is evaluated using an optimization evaluation function, that is, the pros and cons of each initial solution are evaluated according to the optimization objective, and the fitness of each initial solution is output to obtain multiple fitnesses. The higher the fitness, the better the performance of the solution. Then, with the optimization restriction condition as a constraint, based on multiple fitnesses, the stage optimization in the first control time zone is executed until convergence, that is, based on the evaluation of the initial solution, the charge and discharge strategy and the hydrogen production current are adjusted to find the optimal solution, for example, find When the optimal strategy is reached and there is no longer any significant improvement, the first optimal solution is output, that is, the optimal charging and discharging strategy and hydrogen production current in the first control time zone; based on the determined first optimal charging and discharging strategy and hydrogen production current, the required energy storage capacity is calculated, and the energy storage and hydrogen production simulation space is updated according to the first energy storage capacity, including adjusting the energy storage state and parameters in the simulation model to reflect the latest energy storage capacity and operating conditions; similarly, the staged optimization of the second control time zone is executed, and iterative optimization is performed until all control time zones are optimized, and the final charging and discharging strategy and hydrogen production current are output, and multiple control time zones, multiple optimal charging and discharging strategies and multiple optimal hydrogen production currents are obtained to guide actual operations, so that the entire wind power off-grid energy storage and hydrogen production system can operate efficiently and stably in multiple time periods.

[0053] Further, step S335 also includes step S3351, based on the multiple fitnesses, arranging the multiple initial solutions in descending order of fitness, generating an initial solution sequence, and setting the first Q solutions of the initial solution sequence as optimal solutions, and setting the last T solutions as inferior solutions, wherein T is an integer multiple of Q; step S3352, taking the Q optimal solutions as the center, randomly clustering the T inferior solutions, obtaining Q solution thresholds, and setting the solution with the smallest fitness within the Q solution thresholds as a poor solution, wherein the number of poor solutions within each solution threshold is the same; step S3353, within the Q solution thresholds, taking approaching the optimal solution and moving away from the poor solution as the optimization direction, The inferior solution and the difference solution within the solution threshold are adjusted according to a predetermined adjustment step to obtain Q updated solution thresholds; step S3354, setting a penalty mechanism based on the optimization constraint condition, wherein if the updated inferior solution or the difference solution does not satisfy the optimization constraint condition, its fitness is halved; step S3355, optimizing the Q updated solution thresholds according to the penalty mechanism, continuing to iterate the optimization until the predetermined number of optimization times is met, outputting Q current solution thresholds, and setting the solution threshold with the largest sum of fitness as the optimal solution threshold, outputting the optimal solution within the optimal solution threshold, and obtaining the first optimal charging and discharging strategy and the first hydrogen production current.

[0054] Preferably, multiple initial solutions are sorted according to their fitness. The larger the fitness value, the better the quality of the solution. An initial solution sequence is generated and arranged in order from the best solution to the worst solution according to the fitness. The initial solution sequence after fitness sorting is divided into two parts. The first Q solutions are selected as excellent solutions, and the last T solutions are selected as inferior solutions. The sum of Q and T is the total number of initial solutions, and T is an integer multiple of Q, that is, the number of inferior solutions is an integer multiple of excellent solutions; then the inferior solutions are randomly clustered, that is, with the Q excellent solutions as the center, the T inferior solutions are randomly clustered, and the inferior solutions are assigned to different areas, and a number of solution thresholds are formed, each solution threshold contains a certain number of Inferior solutions, and the number of inferior solutions in each solution threshold is the same, where the solution threshold refers to a set obtained by clustering in the solution space, and the solution with the smallest fitness in each solution threshold is taken as the differential solution. For the inferior solutions in each solution threshold, in the optimization process, the adjustment direction is to approach the optimal solution (adjust the inferior solution to a better optimal solution) and to move away from the differential solution (avoid the inferior solution from continuing to develop in the direction of the differential solution). The inferior solutions and differential solutions in the solution threshold are adjusted according to the predetermined adjustment step size to obtain Q updated solution thresholds, where the predetermined adjustment step size determines the amplitude of the adjustment. A step size that is too large may cause the quality of the solution to fluctuate, and a step size that is too small may make the optimization process too slow.

[0055] Preferably, a penalty mechanism is set based on the optimization constraints. Specifically, when the updated inferior solution or poor solution does not meet the optimization constraints (such as energy storage capacity, hydrogen production efficiency, stability, etc.), their fitness is halved, so that the solutions that do not meet the constraints are penalized in fitness, forcing these solutions to move away from the optimal solution, thereby guiding the optimization process to develop in the direction of meeting the constraints. After each update, the solution threshold continues to be optimized, that is, the solutions that do not meet the constraints are adjusted based on the penalty mechanism, and the quality of the solution is gradually improved in each round of iteration until the predetermined number of optimization times is reached. After multiple rounds of iterations, the solution threshold with the largest sum of fitness is selected as the final optimal solution threshold, which includes the optimal charging and discharging strategy and hydrogen production current. Specifically, within the optimal solution threshold, the best performing solution is selected and output as the optimal charging and discharging strategy and hydrogen production current, which becomes the final control strategy of the wind power off-grid energy storage hydrogen production system, ensuring the stability of the system in operation, the efficiency of hydrogen production and the balance of the energy storage system.

[0056] Furthermore, step S3353 also includes step S3353a, calculating the fitness deviation of multiple inferior solutions based on the fitness of the best solution within the solution threshold, and determining multiple fitness deviations; step S3353b, determining multiple adjustment compensation coefficients based on the multiple fitness deviations, wherein the adjustment compensation coefficient and the fitness deviation are positively correlated; step S3353c, optimizing the initial adjustment step size based on the multiple adjustment compensation coefficients, and obtaining multiple optimized adjustment step sizes to construct the predetermined adjustment step size.

[0057] Preferably, the fitness of the optimal solution within the solution threshold is used as a benchmark to calculate the fitness deviation of multiple inferior solutions, that is, the difference between the fitness value of each inferior solution and the fitness value of the optimal solution within the solution threshold is calculated to obtain multiple fitness deviations, which reflect the gap between the inferior solution and the optimal solution. The larger the gap, the farther the inferior solution is from the optimal solution. Then, multiple adjustment compensation coefficients are determined according to the multiple fitness deviations, that is, the fitness deviation of each inferior solution is calculated, and a corresponding compensation coefficient is determined according to the deviation size. The fitness deviation is positively correlated with the compensation coefficient, which means that the larger the fitness deviation, the larger the compensation coefficient, and the inferior solution needs more adjustment. The adjustment compensation coefficient refers to a coefficient used to adjust the step size, which is mainly used to dynamically adjust the inferior solution. For example, For solutions with large fitness deviations, larger compensation coefficients are required to increase the adjustment step size and help the solution approach the optimal solution as quickly as possible. Finally, the initial adjustment step size is optimized according to multiple adjustment compensation coefficients. By using the adjustment compensation coefficients, the initial step size is dynamically optimized to construct a predetermined adjustment step size. Specifically, in the adjustment process of certain solutions, if the fitness deviation of the solution is large, a larger compensation coefficient can be used to increase the step size and accelerate the adjustment; if the fitness deviation of the solution is small, the step size is reduced by a smaller compensation coefficient, and detailed adjustments are made to ensure that each adjustment step size can effectively approach the optimal solution, avoid excessively large or small step sizes that lead to inefficient optimization or unstable results, and thus improve the overall optimization efficiency.

[0058] Step S400: executing the energy storage and hydrogen production operation in the predetermined time zone according to the optimal energy balance control scheme.

[0059] Preferably, in the wind power off-grid energy storage hydrogen production system, based on the optimal control scheme obtained in the optimization process, specific operation steps are actually performed to realize the energy storage hydrogen production operation. Specifically, according to the optimal energy balance control scheme, the energy storage equipment (such as batteries, supercapacitors, etc.) is controlled to perform reasonable charging and discharging operations within a predetermined time zone. For example, when the wind power output power is higher than the demand, the energy storage equipment will charge and store the excess electric energy. When the wind power generation is insufficient or the electrolyzer needs to be powered, the energy storage equipment will discharge and release the stored electric energy for use by the electrolyzer, ensuring that the energy storage system operates most efficiently within the predetermined time zone, while avoiding excessive charging and discharging, and extending the service life of the energy storage equipment; according to the optimal energy balance control scheme, the electrolyzer (used for electrolysis of water to produce hydrogen) adjusts its workload according to the actual charging and discharging state of the energy storage system. For example, when the wind power output is strong and the energy storage system is fully charged, the electrolyzer may operate at full load to maximize hydrogen production; and when the wind power generation is insufficient, the electrolyzer may be adjusted to low load operation to ensure that there is no energy waste. Through the reasonable coordination of charging and discharging strategies and electrolyzer load, the instability of wind power generation can be smoothed as much as possible to ensure the stability and efficiency of hydrogen production.

[0060] Figure 2 1 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention, showing a block diagram of an exemplary electronic device suitable for implementing an implementation of the present invention. Figure 2 The electronic device shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention. The electronic device is in the form of a general computing device, and its components may include but are not limited to an input device 401, a processor 402, a memory 403, and an output device 404. Among them, the processor 402 may be one or more; the memory 403 may include a computer-readable medium and at least one program product, and the program product has a set (at least one) of program modules, which are configured to perform the functions of each embodiment of the present application.

[0061] The memory 403 shown in the embodiment of the present invention may adopt any combination of one or more computer-readable media; the computer-readable storage medium may be, but is not limited to, an infrared, semiconductor system, device or component, or any combination of the above, for storing software programs, computer executable programs and modules, such as the program instructions / modules corresponding to the energy balance optimization method of the wind power off-grid energy storage and hydrogen production system in the embodiment of the present invention. The processor 402 executes various functional applications and data processing of the computer device by running the software programs, instructions and modules stored in the memory 403, thereby realizing the energy balance optimization method of the wind power off-grid energy storage and hydrogen production system mentioned above.

[0062] The energy balance optimization device of the wind power off-grid energy storage and hydrogen production system provided in the embodiment of the present invention can execute the energy balance optimization method of the wind power off-grid energy storage and hydrogen production system provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0063] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules may be used and run on the user terminal and / or server, and the various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0064] The above specific implementations do not constitute a limitation on the protection scope of this application. It should be understood by those skilled in the art that various modifications, combinations and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions and improvements made within the spirit and principles of this application should be included in the protection scope of this application.

Claims

1. An energy balance optimization method for an off-grid wind power energy storage hydrogen production system, characterized in that: Methods include: Obtaining a wind power factor in a predetermined time zone, performing power generation prediction according to the wind power factor, and obtaining predicted output power, including: Query wind power generation logs and collect sample wind power factor sets and sample output power sets; Configure N prediction operators, where the prediction operators include at least a long short-term memory network, a BP neural network, and a random forest, and N is an integer greater than or equal to 3; Counting N prediction accuracies of the N prediction operators based on historical data, and configuring N prediction scales according to the N prediction accuracies, wherein the prediction scale is the number of branches and is positively correlated with the prediction accuracy; The sample wind power factor set and the sample output power set are used as training data, and are equally divided into N parts to obtain N training sets; Using the N training sets and based on the N prediction scales, the N prediction operators are supervised trained until convergence, and N prediction branches are integrated to construct a wind power generation prediction plug-in; Using the wind power generation prediction plug-in, power generation prediction is performed according to the wind power factor to obtain the predicted output power; Monitoring and obtaining the current energy storage capacity of the starting node in the predetermined time zone includes: Configure a tolerance interval for energy storage capacity acquisition, wherein the tolerance interval is set based on the duration of the solution optimization; The starting node is advanced according to the tolerance interval, a collection node is determined, and the energy storage capacity of the collection node is monitored and obtained and set as the current energy storage capacity; Based on the optimization restriction conditions of the wind power off-grid energy storage hydrogen production system configuration, with the optimization restriction conditions as constraints, with the maximum hydrogen production, minimum hydrogen production fluctuation and minimum energy storage aging as optimization purposes, according to the predicted output power and current energy storage capacity, multi-stage optimization of the charging and discharging strategy and the electrolyzer workload is performed to output the optimal energy balance control plan; According to the optimal energy balance control scheme, the energy storage and hydrogen production operation in the predetermined time zone is performed.

2. The energy balance optimization method of the wind power off-grid energy storage hydrogen production system according to claim 1 is characterized in that: The optimization constraints include a storage capacity threshold, a hydrogen production efficiency threshold, and an off-grid balance constraint, wherein the off-grid balance constraint includes a frequency constraint and a voltage constraint of the off-grid system.

3. The energy balance optimization method of the wind power off-grid energy storage hydrogen production system according to claim 2 is characterized in that: Taking the optimization restriction conditions as constraints, taking the maximum hydrogen production, the minimum hydrogen production fluctuation and the minimum energy storage aging as optimization purposes, and according to the predicted output power and the current energy storage capacity, multi-stage optimization of the charging and discharging strategy and the electrolyzer workload is performed to output the optimal energy balance control scheme, including: An optimization evaluation function is configured based on the maximum hydrogen production, minimum hydrogen production fluctuation and minimum energy storage aging; In a three-dimensional simulation space, the wind power off-grid energy storage hydrogen production system is modeled to generate an energy storage hydrogen production simulation space, and the predicted output power and current energy storage capacity are rendered in the energy storage hydrogen production simulation space; Based on the optimization evaluation function, with the optimization restriction condition as a constraint, the energy storage hydrogen production simulation space is used to perform multi-stage optimization of the charging and discharging strategy and the electrolyzer workload, and multiple control time zones, multiple optimal charging and discharging strategies, and multiple optimal hydrogen production currents are output; According to the sequence of the control time zones, the multiple control time zones, the multiple optimal charging and discharging strategies and the multiple optimal hydrogen production currents are integrated to generate the optimal energy balance control scheme.

4. The energy balance optimization method of the wind power off-grid energy storage hydrogen production system according to claim 3 is characterized in that: The expression of the optimization evaluation function is: ; in, To evaluate fitness, is the hydrogen production weight, is the hydrogen production fluctuation weight, is the energy storage aging weight, is the hydrogen production, is the hydrogen production fluctuation coefficient, is the charge and discharge rate weight, is the weight of charge and discharge times, S is the mean charge and discharge rate, and C is the total charge and discharge times; The hydrogen production rate deviations of adjacent time zones within a plurality of hydrogen production time zones are calculated, and the mean of the plurality of hydrogen production rate deviations is calculated to obtain the hydrogen production fluctuation coefficient.

5. The energy balance optimization method of the wind power off-grid energy storage hydrogen production system according to claim 3 is characterized in that: Based on the optimization evaluation function, with the optimization restriction condition as a constraint, the energy storage hydrogen production simulation space is used to perform multi-stage optimization of the charging and discharging strategy and the electrolyzer workload, and multiple control time zones, multiple optimal charging and discharging strategies and multiple optimal hydrogen production currents are output, including: Obtain the charge and discharge regulation space and workload regulation space of the wind power off-grid energy storage hydrogen production system, wherein the charge and discharge regulation is charging power / discharging power, and the workload is the hydrogen production current; The stage time interval is configured according to the predicted output power, and the predetermined time zone is divided into a plurality of control time zones according to the stage time interval, wherein the stage time interval is negatively correlated with the predicted output power, Randomly select any parameter in the charge-discharge regulation space and the workload regulation space to combine and obtain multiple initial solutions, wherein each initial solution includes a charge-discharge strategy and a hydrogen production current; Inputting the multiple initial solutions into the energy storage hydrogen production simulation space for simulation, and evaluating the multiple simulation results according to the optimization evaluation function, and outputting multiple fitness levels; Taking the optimization restriction condition as a constraint and based on the multiple fitnesses, performing stage optimization in the first control time zone until convergence, and outputting a first optimal charging and discharging strategy and a first hydrogen production current; Based on the first optimal charge and discharge strategy and the first hydrogen production current, the first energy storage capacity is determined, and the energy storage and hydrogen production simulation space is rendered and updated according to the first energy storage capacity, and stage optimization in the second control time zone is executed, and iterative optimization is performed until the multi-stage optimization in the predetermined time zone is completed, and multiple control time zones, multiple optimal charge and discharge strategies, and multiple optimal hydrogen production currents are output.

6. The energy balance optimization method of the wind power off-grid energy storage hydrogen production system according to claim 5 is characterized in that: Taking the optimization restriction condition as a constraint and based on the multiple fitnesses, performing stage optimization in the first control time zone until convergence, and outputting a first optimal charging and discharging strategy and a first hydrogen production current, including: Based on the multiple fitnesses, the multiple initial solutions are arranged in descending order of fitness to generate an initial solution sequence, and the first Q solutions of the initial solution sequence are set as optimal solutions, and the last T solutions are set as inferior solutions, where T is an integer multiple of Q; Taking Q optimal solutions as the center, randomly cluster T inferior solutions to obtain Q solution thresholds, and set the solution with the smallest fitness within the Q solution thresholds as the differential solution, where the number of inferior solutions within each solution threshold is the same; Within the Q solution thresholds, taking approaching the optimal solution and moving away from the poor solution as the optimization direction, adjusting the poor solution and the poor solution within the solution threshold according to a predetermined adjustment step length, and obtaining Q updated solution thresholds; A penalty mechanism is set based on the optimization constraint, wherein if the updated inferior solution or poor solution does not meet the optimization constraint, its fitness is halved; The Q updated solution thresholds are optimized according to the penalty mechanism, and the iterative optimization is continued until a predetermined number of optimization times is met, Q current solution thresholds are output, and the solution threshold with the largest sum of fitness is set as the optimal solution threshold, and the optimal solution within the optimal solution threshold is output to obtain the first optimal charging and discharging strategy and the first hydrogen production current.

7. The energy balance optimization method of the wind power off-grid energy storage hydrogen production system according to claim 6 is characterized in that: Configure the scheduled adjustment step size, including: Taking the fitness of the best solution within the solution threshold as a benchmark, the fitness deviation of multiple inferior solutions is calculated to determine multiple fitness deviations; Determining a plurality of adjustment compensation coefficients according to the plurality of fitness deviations, wherein the adjustment compensation coefficients are positively correlated with the fitness deviations; The initial adjustment step length is optimized according to the multiple adjustment compensation coefficients to obtain multiple optimized adjustment step lengths to construct the predetermined adjustment step length.

8. An electronic device, characterized in that: The electronic device comprises: A memory for storing executable instructions; The processor is used to implement the energy balance optimization method of the wind power off-grid energy storage hydrogen production system as described in any one of claims 1 to 7 when executing the executable instructions stored in the memory.

Images