A dynamic load regulation method for off-grid energy storage system

Abstract

The present application provides a dynamic load regulation method for an off-grid energy storage system, which relates to the technical field of off-grid energy storage management, including: performing energy supply adaptability calculation according to a first energy supply load parameter and predicted electricity demand to obtain a first energy supply adaptability parameter, performing aging analysis according to the first energy supply load parameter and predicted grid-connected energy storage data to obtain a first aging parameter; performing energy supply stability analysis to obtain an energy supply stability parameter, combining the first aging parameter and the first energy supply adaptability parameter to calculate and obtain a first load regulation adaptability, and continuing to perform energy supply load regulation optimization to obtain an optimal energy supply load parameter. The present application can solve the technical problem in the existing methods that the energy supply matching and energy supply stability are poor due to the inability to accurately adjust the energy supply output power according to energy storage data and energy demand, and can achieve the goal of accurately adjusting the energy supply output power, ensuring efficient and stable power supply of the energy storage system, while extending the service life of the system.

Description

Technical Field

[0001] The present application relates to the technical field of off-grid energy storage management, and in particular to a dynamic load regulation method for an off-grid energy storage system. Background Art

[0002] With the continuous growth of global energy demand and the increasing attention to environmental protection, the role of energy storage systems in the field of power supply is becoming increasingly important. Energy storage systems can not only balance the intermittent and unpredictable nature of renewable energy, but also improve the operating efficiency and reliability of the power grid. However, existing energy storage system load regulation methods often rely on simple control logic or preset regulation parameters. These methods cannot be accurately adjusted according to dynamically changing energy storage data and energy demand, resulting in a large deviation between the energy output power and actual demand.

[0003] At present, there are technical problems in the existing load regulation methods of energy storage systems, such as the inability to accurately adjust the energy output power according to energy storage data and energy demand, resulting in poor energy supply matching and stability, as well as affecting the service life of the energy storage system. Summary of the invention

[0004] The purpose of this application is to provide a dynamic load regulation method for an off-grid energy storage system, so as to solve the technical problems in the existing load regulation method of the energy storage system, which is that the energy supply output power cannot be accurately adjusted according to the energy storage data and energy demand, resulting in poor energy supply matching and energy supply stability, and affecting the service life of the energy storage system.

[0005] On the first aspect, in view of the above problems, the present application provides a dynamic load regulation method for an off-grid energy storage system, including: collecting off-grid energy supply data and grid-connected energy storage data of the off-grid energy storage system within a preset time in the past, and predicting and obtaining predicted electricity demand data and predicted grid-connected energy storage data within the next off-grid cycle; constructing a load regulation space for the energy storage system, and randomly generating a first energy supply load parameter; performing energy supply adaptability calculation according to the first energy supply load parameter and the predicted electricity demand data to obtain a first energy supply adaptability parameter, performing aging analysis according to the first energy supply load parameter and the predicted grid-connected energy storage data to obtain a first aging parameter; performing energy supply stability analysis according to the first energy supply adaptability parameter and the historical energy supply adaptability parameters within the preset time in the past to obtain an energy supply stability parameter, combining the first aging parameter and the first energy supply adaptability parameter to calculate and obtain a first load regulation fitness, and continuing to optimize the energy supply load regulation of the energy storage system to obtain the optimal energy supply load parameter, and performing energy supply load regulation in the next off-grid cycle.

[0006] In a second aspect, the present invention provides an electronic device, comprising: a memory for storing a computer software program; a processor for reading and executing the computer software program, thereby implementing the dynamic load regulation method of the off-grid energy storage system of the first aspect.

[0007] In a third aspect, the present invention provides a computer-readable storage medium having a computer software program stored therein, and when the computer software program is executed by a processor, the dynamic load regulation method for an off-grid energy storage system of the first aspect is implemented.

[0008] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0009] By collecting the off-grid energy supply data and the grid-connected energy storage data of the off-grid and grid-connected energy storage system in the past preset time, and predicting and obtaining the predicted electricity demand data and the predicted grid-connected energy storage data in the next off-grid and grid-connected cycle; then constructing the load regulation space of the energy storage system, and randomly selecting the first energy supply load parameter in the load regulation space; then performing energy supply adaptability calculation according to the first energy supply load parameter and the predicted electricity demand data to obtain the first energy supply adaptability parameter, performing aging analysis according to the first energy supply load parameter and the predicted grid-connected energy storage data to obtain the first aging parameter; further performing aging analysis according to the first energy supply adaptability parameter and the historical energy supply adaptability parameters in the past preset time Perform energy supply stability analysis to obtain energy supply stability parameters; then calculate and obtain a first load regulation fitness according to the energy supply stability parameters, the first aging parameters and the first energy supply adaptability parameters; finally, based on the first load regulation fitness, continue to optimize the energy supply load regulation of the energy storage system to obtain the optimal energy supply load parameters with the greatest fitness, and adjust the energy supply load of the off-grid energy storage system in the next off-grid cycle according to the optimal energy supply load parameters; by introducing an intelligent adjustment mechanism, the technical goal of accurately adjusting the energy supply output power can be achieved, ensuring efficient and stable power supply of the energy storage system, while slowing down the aging rate of the energy storage system and extending its service life.

[0010] 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 according to the contents of the specification, and 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 specifically cited below. It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present application, nor is it intended to limit the scope of the present application. Other features of the present application will become easy to understand through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions in the present application or the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings in the following description are only exemplary, and for ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0012] Figure 1 A schematic flow chart of a dynamic load regulation method for an off-grid energy storage system for the present application;

[0013] Figure 2 A schematic diagram of a flow chart of obtaining predicted electricity demand data and predicted grid-connected energy storage data in a dynamic load regulation method for an off-grid energy storage system of the present application;

[0014] Figure 3 A functional block diagram of an electronic device provided by an embodiment of the present invention.

[0015] In the accompanying drawings, the components represented by the reference numerals are as follows:

[0016] Electronic device 100 , memory 110 , processor 120 , computer program 111 . DETAILED DESCRIPTION

[0017] This application provides a dynamic load regulation method for an off-grid energy storage system, which solves the technical problems in existing energy storage system load regulation methods, such as the inability to accurately adjust the energy supply output power according to energy storage data and energy demand, resulting in poor energy supply matching and stability, and affecting the service life of the energy storage system. It can improve the intelligence level of the energy storage system, achieve the technical goal of accurately adjusting the energy supply output power, ensure efficient and stable power supply of the energy storage system, and at the same time slow down the aging rate of the energy storage system and extend its service life.

[0018] Below, the technical solutions in the present application will be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described herein. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0019] For examples, please see the attached Figure 1 The present application provides a method for dynamic load regulation of an off-grid energy storage system, which specifically includes the following steps:

[0020] S100: Collect off-grid energy supply data and grid-connected energy storage data of the off-grid and grid-connected energy storage system within a preset period of time in the past, and predict and obtain predicted power demand data and predicted grid-connected energy storage data within the next off-grid and grid-connected cycle.

[0021] In one embodiment, as shown in the attached Figure 2 As shown, step S100 of the present application also includes:

[0022] S110: pre-training an off-grid and on-grid data predictor according to the operation data of the energy storage system in historical time;

[0023] In one embodiment, step S110 of the present application further includes:

[0024] S111: Based on the energy supply data and energy storage data of all off-grid and grid-connected cycles of the energy storage system in historical time, a sample historical off-grid energy supply data sequence set and a sample historical grid-connected energy storage data sequence set are collected as input supervised training data; S112: The electricity demand data and grid-connected energy storage data in the off-grid and grid-connected cycle after each sample historical off-grid energy supply data sequence and sample historical grid-connected energy storage data sequence are identified as a sample predicted electricity demand data set and a sample predicted grid-connected energy storage data set, as output supervised training data; S113: Using the input supervised training data and output supervised training data, using machine learning, supervised training of an off-grid and grid-connected data predictor is performed until convergence conditions are met.

[0025] Specifically, first, the energy supply data and energy storage data of all off-grid and grid-connected cycles of the energy storage system in the historical time are obtained, wherein the historical time refers to the time period from the start of the operation of the energy storage system to the current time node, that is, the historical operation time period of the energy storage system; the historical time includes multiple off-grid and grid-connected cycles, wherein the off-grid and grid-connected cycle refers to the complete operation cycle of the energy storage system switching from off-grid mode to grid-connected mode within a period of time, which can be set according to the actual scenario. For example, the energy storage system is in off-grid mode during the day, mainly responsible for powering the load, and switches to grid-connected mode at night for energy storage charging. The time period of one day and one night is a complete off-grid and grid-connected cycle. Next, the energy supply data and energy storage data in each off-grid cycle are extracted. The energy supply data refers to the output power of the energy storage system at multiple time nodes in the off-grid cycle, that is, the power when the system supplies power to the load, for example, the output power is collected every one minute in the off-grid cycle; the energy storage data refers to the charging power of the energy storage system at multiple time nodes in the off-grid cycle, for example, the charging power of the energy storage device is collected at the same time node (such as every minute); then the energy supply data and energy storage data in each off-grid cycle are arranged in the order of the time nodes to obtain a sample historical off-grid energy supply data sequence set and a sample historical grid-connected energy storage data sequence set, and the sample historical off-grid energy supply data sequence set and the sample historical grid-connected energy storage data sequence set are used as the supervised training data of the input layer of the subsequent off-grid and grid-connected data predictor.

[0026] Then, the electricity demand data in the off-grid cycle after each sample historical off-grid energy supply data sequence in the sample historical off-grid energy supply data sequence set is obtained, that is, the electricity demand data in the next adjacent off-grid cycle, wherein the electricity demand data refers to the electricity demand power, and the electricity demand data is sampled and identified as sample predicted electricity demand data, wherein the sample predicted electricity demand data and the sample historical off-grid energy supply data sequence correspond one-to-one to obtain the sample predicted electricity demand data set. On the other hand, the grid-connected energy storage data in the off-grid cycle after each sample historical grid-connected energy storage data sequence in the sample historical grid-connected energy storage data sequence set is obtained, that is, the grid-connected energy storage data in the next adjacent off-grid cycle, wherein the grid-connected energy storage data refers to the grid-connected charging power, and the grid-connected energy storage data is sampled and identified as sample predicted grid-connected energy storage data, wherein the sample predicted grid-connected energy storage data and the sample historical grid-connected energy storage data sequence correspond one-to-one to obtain the sample predicted grid-connected energy storage data set. Then, the sample predicted electricity demand data set and the sample predicted grid-connected energy storage data set are combined as the supervised training data of the output layer of the subsequent off-grid data predictor.

[0027] An off-grid and grid-connected data predictor is constructed based on machine learning. For example, an off-grid and grid-connected data predictor is constructed based on a BP neural network. The off-grid and grid-connected data predictor is a feedforward neural network model that can be iteratively optimized in machine learning. Technical personnel in this field can also construct other machine learning models (such as regression models, long short-term memory networks, etc.) in combination with actual application scenarios. In this application, a BP neural network is taken as an example, wherein the off-grid and grid-connected data predictor includes an off-grid data prediction branch and a grid-connected data prediction branch. The off-grid data prediction branch includes an input layer, multiple hidden layers and an output layer, wherein the input data of the input layer is an off-grid energy supply data sequence, and the output data of the output layer is electricity demand data; then, the sample historical off-grid energy supply data sequence is used as input, the sample predicted electricity demand data is used as output, and the sample historical off-grid energy supply data sequence set and the sample predicted electricity demand data set are used as supervised training. Data, supervised training of the off-grid data prediction branch. First, divide the data set into multiple small batches, and perform forward propagation and back propagation on each batch of data. Processing a small portion of data each time helps to improve training speed and reduce memory consumption; then for each batch of data, the model calculates through the input layer, hidden layer and output layer to obtain the predicted electricity demand data; then the error between the predicted value and the actual electricity demand is calculated through the loss function, and the gradient of each parameter is calculated according to the size of the loss using the chain rule, and the weights and biases are updated according to these gradients; further use the optimization algorithm to adjust the model parameters, gradually reduce the value of the loss function, and continuously perform iterative training. When the value of the loss function on the training set and the validation set no longer decreases significantly, the model training ends, indicating that the model has achieved good prediction performance, and the off-grid data prediction branch that has been trained is obtained. Through the above steps, the off-grid data prediction branch can learn the patterns in historical data through supervised training and realize accurate prediction of future electricity demand.

[0028] The grid-connected data prediction branch also includes an input layer, multiple hidden layers and an output layer. The input data of the input layer is the grid-connected energy storage data sequence, and the output data of the output layer is the grid-connected energy storage data. Then, the sample historical grid-connected energy storage data sequence is used as input, the sample predicted grid-connected energy storage data is used as output, and the sample historical grid-connected energy storage data sequence set and the sample predicted grid-connected energy storage data set are used as supervised training data to supervise the grid-connected data prediction branch. First, the sample historical grid-connected energy storage data sequence is input into the input layer of the model. After the nonlinear transformation of the hidden layer, the output layer predicts the charging power of the next grid-connected cycle. Then, the error between the predicted value and the true value is calculated using the defined loss function, and the gradient of each parameter is calculated according to the value of the loss function. The weight and bias of the model are updated using the optimization algorithm to reduce the prediction error. Iterative training is continuously performed until the predetermined convergence condition is met (such as the value of the loss function reaches the convergence standard), and the trained grid-connected data prediction branch is obtained. Through the above steps, the grid-connected data prediction branch learns the energy storage behavior in the grid-connected cycle through supervised training, and can accurately predict the energy storage demand in the future grid-connected cycle.

[0029] Finally, an off-grid and grid-connected data predictor is formed based on the trained off-grid data prediction branch and the grid-connected data prediction branch. By constructing an off-grid and grid-connected data predictor based on machine learning, the efficiency and accuracy of electricity demand data and grid-connected energy storage data prediction can be improved.

[0030] S120: Collect the average power demand power and average grid-connected energy storage power of the off-grid and grid-connected energy storage system in multiple off-grid and grid-connected cycles within a preset time in the past as a historical off-grid energy supply data sequence and a historical grid-connected energy storage data sequence; S130: Input the historical off-grid energy supply data sequence and the historical grid-connected energy storage data sequence into the off-grid and grid-connected data predictor, obtain the output result, and obtain the predicted power demand data and the predicted grid-connected energy storage data.

[0031] Specifically, first, the average power demand of the off-grid energy storage system in multiple off-grid cycles within the past preset time is collected. The preset time can be set according to the actual scenario, such as in the last month, and multiple average power demand powers are arranged in the order of the cycles to obtain a historical off-grid energy supply data sequence. On the other hand, the average grid-connected energy storage power of the off-grid energy storage system in multiple off-grid cycles within the past preset time is collected, and the multiple average grid-connected energy storage powers are arranged in the order of the cycles to obtain a historical grid-connected energy storage data sequence.

[0032] Then, the historical off-grid energy supply data sequence is input into the off-grid data prediction branch of the off-grid and grid-connected data predictor for prediction, and the power demand data in the next off-grid cycle is predicted to obtain predicted power demand data, wherein the predicted power demand data reflects the load power that the energy storage system needs to provide in the next off-grid cycle; the historical grid-connected energy storage data sequence is input into the grid-connected data prediction branch of the off-grid and grid-connected data predictor for prediction, and the energy storage data in the next grid-connected cycle is predicted to obtain predicted grid-connected energy storage data, wherein the predicted grid-connected energy storage data reflects the charging power obtained by the system from the power grid in the next grid-connected cycle.

[0033] S200: constructing a load adjustment space for the energy storage system and randomly generating a first energy supply load parameter;

[0034] In one embodiment, step S200 of the present application further includes:

[0035] S210: Obtaining an energy supply power range of the energy storage system and constructing it into a load regulation space; S220: Randomly generating a first energy supply load parameter in the load regulation space.

[0036] Specifically, first, the energy supply power range of the energy storage system is obtained, where the energy supply power range refers to the output power range of the energy storage system, which is determined by the minimum output power and maximum output power of the system. The minimum output power refers to the output power of the system under the lowest load condition, in kilowatts (kW), and the maximum output power refers to the output power of the energy storage system under the highest load condition. For example, the energy supply power range is 500 kW to 2000 kW. Then, the energy supply power range is defined as a load adjustment space, in which the output power of the energy storage system can be flexibly adjusted under different requirements to ensure that the load is stably powered. Then, any energy supply load parameter is randomly selected in the constructed load adjustment space as the first energy supply load parameter to obtain the first energy supply load parameter.

[0037] S300: performing energy supply adaptability calculation according to the first energy supply load parameter and the predicted electricity demand data to obtain a first energy supply adaptability parameter, and performing aging analysis according to the first energy supply load parameter and the predicted grid-connected energy storage data to obtain a first aging parameter;

[0038] In one embodiment, step S300 of the present application further includes:

[0039] S310: Calculate the ratio of the first energy supply load parameter and the predicted electricity demand data to obtain a first energy supply adaptability parameter;

[0040] Specifically, an energy supply adaptability calculation is performed based on the first energy supply load parameter and the predicted electricity demand data, that is, the ratio of the first energy supply load parameter to the predicted electricity demand data is calculated, and the ratio is set as the first energy supply adaptability parameter, wherein the energy supply adaptability parameter is used to characterize the degree of adaptability between the energy supply capacity of the energy storage system and the load demand. The larger the energy supply adaptability parameter, the more the energy storage system can meet the external electricity demand in the next off-grid cycle, and the better the adaptability. The smaller the energy supply adaptability parameter, the weaker the ability to meet the electricity demand, and the worse the adaptability.

[0041] S320: Inputting the first energy supply load parameter and the predicted grid-connected energy storage data into a pre-trained aging analyzer for aging analysis to obtain a first aging rate as a first aging parameter.

[0042] In one embodiment, step S320 of the present application further includes:

[0043] S321: Based on the aging test data of the off-grid energy storage system, a sample energy supply load parameter set and a sample grid-connected energy storage data set are collected, and the aging failure time of the energy storage system under different sample energy supply load parameters and sample grid-connected energy storage data are collected, and a sample aging rate set is calculated; S322: The sample energy supply load parameter set and the sample grid-connected energy storage data set are used as input supervised training data, and the sample aging rate set is used as output supervised training data, and machine learning is used to train the aging analyzer until the convergence conditions are met.

[0044] Specifically, first, the aging test data of the off-grid energy storage system is obtained. These data are usually obtained through long-term experimental tests, and record the performance changes of the energy storage system under different load conditions, including energy supply load, grid-connected energy storage power, operating time, and aging failure time, etc. Then, sample energy supply load parameters and sample grid-connected energy storage data are collected from the aging test data, that is, the output power of the energy storage system under different test conditions is extracted from the aging test data. These parameters reflect the output load of the energy storage system in the off-grid mode, and sample grid-connected energy storage data is collected from the aging test data, that is, the energy storage data of the energy storage system in the grid-connected mode is extracted, indicating the charging power of the system under different test conditions. These data indicate the energy storage behavior of the system in the grid-connected state, and a set of sample energy supply load parameters and a set of sample grid-connected energy storage data are obtained. Next, the aging failure time of the energy storage system under different sample energy supply load parameters and sample grid-connected energy storage data is collected. The aging failure time refers to the time from the start of operation of the energy storage system to the occurrence of obvious aging or performance degradation, and the unit is usually months or years, and multiple aging failure times are obtained; then the aging rate is calculated based on the multiple aging failure times, where the aging rate is the reciprocal of the aging failure time. For example, assuming that the aging failure time of a sample is 5 years, the aging rate is 20% per year, and a sample aging rate set is obtained, in which the sample energy supply load parameters, sample grid-connected energy storage data and sample aging rates correspond one to one.

[0045] Then, an aging analyzer is constructed based on machine learning. Those skilled in the art can select an adaptive machine learning model (such as a regression model, a support vector machine, a BP neural network, a random decision forest, etc.) according to the scenario requirements. For example, an aging analyzer is constructed based on a random decision forest, wherein the aging analyzer includes an input layer, a hidden layer consisting of multiple random decision trees, and an output layer, wherein the input data of the input layer are energy supply load parameters and grid-connected energy storage data, and the output data of the output layer is an aging rate; then, sample energy supply load parameters and sample grid-connected energy storage data are used as input data, and sample aging rates are used as output data, and a set of sample energy supply load parameters, a set of sample grid-connected energy storage data, and a set of sample aging rates are used as output data. The aging analyzer is supervised and trained using the sample aging rate set as training data. First, the sample energy supply load parameter set and the sample grid-connected energy storage data set are used as input data and input into the input layer of the aging analyzer. These data will be passed to each decision tree of the random decision forest model. Then, each decision tree of the random decision forest is split by the features in the training data to learn how the energy supply load and grid-connected energy storage data affect the aging rate. The training of each tree uses different subsets of data and feature subsets to enhance the robustness and generalization ability of the model. Then, the decision tree is split according to specific conditions (such as the energy supply load parameter is greater than a certain threshold) to gradually build a tree structure. The leaf nodes of the tree give the predicted aging rate. At the same time, during the training process, the performance of the model can be optimized by adjusting the hyperparameters of the model (such as the number of decision trees, the depth of the tree, the maximum number of features, etc.) to avoid overfitting or underfitting. At the same time, the mean square error is used as the loss function to calculate the error between the predicted result of the model and the actual aging rate, and the error is minimized through continuous iteration. Iterative training is performed to obtain an aging analyzer that meets the expected convergence conditions (such as the value of the loss function reaches the convergence standard).

[0046] Then, the first energy supply load parameter and the predicted grid-connected energy storage data are input into the trained aging analyzer for aging analysis, and a first aging rate is output, and the first aging rate is used as the first aging parameter. By constructing an aging analyzer based on random decision forest, the aging rate of the energy storage system under different load and energy storage conditions can be effectively predicted, and the accuracy of the aging rate prediction can be significantly improved, thereby helping the subsequent energy storage system to optimize the operation strategy and extend the life of the energy storage system.

[0047] S400: Perform energy supply stability analysis based on the first energy supply adaptability parameter and the historical energy supply adaptability parameter within a preset time in the past to obtain energy supply stability parameters, calculate and obtain a first load regulation fitness by combining the first aging parameter and the first energy supply adaptability parameter, and continue to optimize the energy supply load regulation of the energy storage system to obtain the optimal energy supply load parameter, and perform energy supply load regulation in the next off-grid and on-grid cycle.

[0048] In one embodiment, step S400 of the present application further includes:

[0049] S410: Calculate and obtain a historical energy supply adaptability parameter sequence based on a historical energy supply load parameter sequence and a historical electricity demand data sequence of an off-grid energy storage system in multiple off-grid cycles within a preset time in the past; S420: Randomly extract a plurality of first historical energy supply adaptability parameters from the historical energy supply adaptability parameter sequence, calculate the mean to obtain a first average historical energy supply adaptability parameter, calculate the deviation between the first average historical energy supply adaptability parameter and the first energy supply adaptability parameter, and obtain a first stability parameter by subtracting the deviation from 1; S430: Randomly extract a plurality of historical energy supply adaptability parameters from the historical energy supply adaptability parameter sequence again, calculate and obtain a plurality of stability parameters in combination with the first energy supply adaptability parameter, and calculate the mean to obtain an energy supply stability parameter.

[0050] Specifically, first, obtain the historical energy supply load parameter sequence and historical electricity demand data sequence of the off-grid energy storage system in multiple off-grid cycles within the preset time in the past; and perform mean calculation on multiple historical energy supply load parameter sequences to obtain multiple historical energy supply load parameter means, and perform mean calculation on multiple historical electricity demand data sequences to obtain multiple historical electricity demand data means. Further, perform energy supply adaptability calculation based on multiple historical energy supply load parameter means and multiple historical electricity demand data means, that is, calculate the energy supply adaptability parameter of each cycle by the ratio of the historical energy supply load parameter mean to the historical electricity demand data mean. The energy supply adaptability parameter represents the degree of matching between the energy supply load and electricity demand of the energy storage system, obtains multiple historical energy supply adaptability parameters, and arranges them in the order of multiple off-grid cycles to obtain a historical energy supply adaptability parameter sequence.

[0051] Then, multiple historical energy supply adaptability parameters are randomly selected from the historical energy supply adaptability parameter sequence and set as the first historical energy supply adaptability parameter, wherein the number of extractions can be set according to the number of off-grid cycles within a preset time. For example, the number of extractions is set to 20% of the number of off-grid cycles. Assuming that the number of off-grid cycles is 30, the number of extractions is 6; then, the mean of the multiple first historical energy supply adaptability parameters is calculated to obtain the first average historical energy supply adaptability parameter. Further calculate the deviation amplitude between the first average historical energy supply adaptability parameter and the first energy supply adaptability parameter, the deviation amplitude represents the degree of difference between the energy supply adaptability parameter of the current cycle and the historical average, expressed as a percentage, for example, the deviation amplitude is equal to the difference between the first energy supply adaptability parameter of the current cycle and the first average historical energy supply adaptability parameter, divided by the first average historical energy supply adaptability parameter, and then multiplied by 100%; further adopt 1 minus the said deviation amplitude, and take the difference between the two as the first stability parameter, wherein the stability coefficient is an indicator for measuring whether the energy supply adaptability of the energy storage system remains stable in different off-grid cycles, and its purpose is to evaluate the fluctuation of the energy supply capacity of the energy storage system in multiple cycles to ensure the continuity and consistency of power supply, that is, the energy supply capacity of the energy storage system cannot be too high today and insufficient tomorrow. Such fluctuations will affect the user experience and may cause some cycles to fail to meet electricity demand.

[0052] Then, multiple historical energy supply adaptability parameters are randomly extracted multiple times in the historical energy supply adaptability parameter sequence, where the number of extractions can be set according to demand, for example, the number of extractions is set to 10 times, and multiple average historical energy supply adaptability parameters are calculated; then, multiple stability parameters are calculated based on the multiple average historical energy supply adaptability parameters and the first energy supply adaptability parameter; finally, the mean of the multiple stability parameters is calculated, and the mean calculation result is set as the energy supply stability parameter, where when the energy supply stability parameter is large, it means that the fluctuation between the current energy supply adaptability parameter and the historical adaptability is small, the system energy supply stability is good, and the power supply capacity is stable; when the energy supply stability parameter is small, it means that there is a large fluctuation in the system energy supply adaptability, the current energy supply adaptability is inconsistent with the historical adaptability, and there may be unstable power supply.

[0053] In one embodiment, step S400 of the present application further includes:

[0054] S440: Calculate and obtain a first load regulation adaptability according to the energy supply stability parameter, the first aging parameter and the first energy supply adaptability parameter, as shown in the following formula: ; Among them, PTF is the load regulation adaptability, , and is the weight, S is the energy supply stability parameter, D is the aging parameter, and P is the energy supply adaptability parameter; S450: continue to optimize the energy supply load parameters in the load regulation space until the optimization converges, output the energy supply load parameters with the maximum load regulation adaptability, and obtain the optimal energy supply load parameters.

[0055] Specifically, first, construct the load regulation fitness calculation formula: ; Among them, PTF is the load regulation fitness. The higher the fitness, the better the overall load regulation effect. is the energy supply stability weight, is the system aging weight, is the energy supply adaptability weight, where , and The sum is 1. The specific weight value is set according to the influence of the indicator on the load regulation adaptability. The greater the influence, the greater the corresponding weight value. For example, the influence of energy supply stability is 3, the influence of system aging is 2, and the influence of energy supply adaptability is 5. is 0.3, is 0.2, is 0.5; S is the energy supply stability parameter, which is positively correlated with the load regulation fitness, that is, the larger the energy supply stability parameter, the greater the load regulation fitness; D is the aging parameter, which is negatively correlated with the load regulation fitness, that is, the larger the aging parameter, the smaller the load regulation fitness; P is the energy supply adaptability parameter, which is positively correlated with the load regulation fitness, that is, the larger the energy supply adaptability parameter, the greater the load regulation fitness.

[0056] According to the load regulation fitness calculation formula, the fitness calculation is performed on the energy supply stability parameter, the first aging parameter and the first energy supply adaptability parameter to obtain the first load regulation fitness. Then, the second energy supply load parameter is randomly selected in the load regulation space, wherein the second energy supply load parameter is different from the first energy supply load parameter, and the second load regulation fitness of the second energy supply load parameter is calculated; the same method is used for iterative optimization selection until the predetermined number of optimizations is met, and the predetermined number of optimizations can be set according to the optimization accuracy requirements. For example, the predetermined number of optimizations can be set to 100 times. After the predetermined number of optimizations is met, the energy supply load parameter with the largest current load regulation fitness is output as the optimal energy supply load parameter. Finally, according to the optimal energy supply load parameter, the off-grid energy storage system is subjected to energy supply load adjustment in the next off-grid cycle. By optimizing the energy supply load parameter to obtain the optimal energy supply load parameter, and adjusting the energy supply load according to the optimal energy supply load parameter, the energy storage system can optimize its power supply adaptability, reduce the aging rate and improve the power supply stability in different operation cycles, which not only improves the performance of the system, but also prolongs the life of the equipment.

[0057] In summary, the dynamic load regulation method of an off-grid energy storage system provided in this application has the following technical effects:

[0058] By collecting the off-grid energy supply data and the grid-connected energy storage data of the off-grid and grid-connected energy storage system in the past preset time, and predicting and obtaining the predicted electricity demand data and the predicted grid-connected energy storage data in the next off-grid and grid-connected cycle; then constructing the load regulation space of the energy storage system, and randomly selecting the first energy supply load parameter in the load regulation space; then performing energy supply adaptability calculation according to the first energy supply load parameter and the predicted electricity demand data to obtain the first energy supply adaptability parameter, performing aging analysis according to the first energy supply load parameter and the predicted grid-connected energy storage data to obtain the first aging parameter; further performing aging analysis according to the first energy supply adaptability parameter and the historical energy supply adaptability parameters in the past preset time Perform energy supply stability analysis to obtain energy supply stability parameters; then calculate and obtain a first load regulation fitness according to the energy supply stability parameters, the first aging parameters and the first energy supply adaptability parameters; finally, based on the first load regulation fitness, continue to optimize the energy supply load regulation of the energy storage system to obtain the optimal energy supply load parameters with the greatest fitness, and adjust the energy supply load of the off-grid energy storage system in the next off-grid cycle according to the optimal energy supply load parameters; by introducing an intelligent adjustment mechanism, the technical goal of accurately adjusting the energy supply output power can be achieved, ensuring efficient and stable power supply of the energy storage system, while slowing down the aging rate of the energy storage system and extending its service life.

[0059] See also Figure 3 , Figure 3 Schematic diagram of an electronic device provided by an embodiment of the present invention. Figure 3 As shown, an embodiment of the present invention provides an electronic device 100, including a memory 110, a processor 120, and a computer program 111 stored in the memory 110 and executable on the processor 120. When the processor 120 executes the computer program 111, the steps of the method in Example 1 are implemented.

[0060] It should be noted that in the above embodiments, the description of each embodiment has its own emphasis, and for parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0061] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0062] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded computer, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0063] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0064] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0065] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0066] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A dynamic load regulation method for an off-grid energy storage system, characterized in that the method include: Collect off-grid energy supply data and grid-connected energy storage data of the off-grid and grid-connected energy storage system in the past preset time, and predict and obtain the predicted power demand data and predicted grid-connected energy storage data in the next off-grid and grid-connected cycle; Constructing a load regulation space of the energy storage system and randomly generating a first energy supply load parameter; Performing energy supply adaptability calculation according to the first energy supply load parameter and the predicted electricity demand data to obtain a first energy supply adaptability parameter, and performing aging analysis according to the first energy supply load parameter and the predicted grid-connected energy storage data to obtain a first aging parameter; According to the first energy supply adaptability parameter and the historical energy supply adaptability parameter within the preset time in the past, an energy supply stability analysis is performed to obtain an energy supply stability parameter, and the first load regulation fitness is calculated and obtained by combining the first aging parameter and the first energy supply adaptability parameter, and the energy supply load regulation optimization of the energy storage system is continued to obtain the optimal energy supply load parameter, and the energy supply load is regulated in the next off-grid cycle; Among them, performing energy supply stability analysis according to the first energy supply adaptability parameter and the historical energy supply adaptability parameter within the past preset time to obtain the energy supply stability parameter includes: According to the historical energy supply load parameter sequence and the historical electricity demand data sequence of the off-grid energy storage system in multiple off-grid cycles within a preset time in the past, a historical energy supply adaptability parameter sequence is calculated; Randomly extracting a plurality of first historical energy supply adaptability parameters from the historical energy supply adaptability parameter sequence, calculating the mean to obtain a first average historical energy supply adaptability parameter, calculating the deviation amplitude between the first average historical energy supply adaptability parameter and the first energy supply adaptability parameter, and subtracting the deviation amplitude from 1 to obtain a first stability parameter; Again, randomly extracting multiple historical energy supply adaptability parameters multiple times in the historical energy supply adaptability parameter sequence, combining the first energy supply adaptability parameter, calculating multiple stability parameters, and calculating the average to obtain an energy supply stability parameter; The first load regulation adaptability is calculated by combining the first aging parameter and the first energy supply adaptability parameter, and the energy supply load regulation optimization of the energy storage system is continued to obtain the optimal energy supply load parameter, including: According to the energy supply stability parameter, the first aging parameter and the first energy supply adaptability parameter, the first load regulation adaptability is calculated and obtained as follows: ； Among them, PTF is the load regulation fitness, , and is the weight, S is the energy supply stability parameter, D is the aging parameter, and P is the energy supply adaptability parameter; Continue to optimize the energy supply load parameters in the load regulation space until the optimization converges, output the energy supply load parameters with the maximum load regulation fitness, and obtain the optimal energy supply load parameters.

2. The dynamic load regulation method of the off-grid energy storage system according to claim 1, characterized in that: Collect off-grid energy supply data and grid-connected energy storage data of the off-grid and grid-connected energy storage system in the past preset time, and predict and obtain the predicted power demand data and predicted grid-connected energy storage data in the next off-grid and grid-connected cycle, including: Pre-training an off-grid and on-grid data predictor based on the operation data of the energy storage system in historical time; Collect the average power demand power and average grid-connected energy storage power of the off-grid energy storage system in multiple off-grid cycles within a preset time in the past as a historical off-grid energy supply data series and a historical grid-connected energy storage data series; The historical off-grid energy supply data sequence and the historical grid-connected energy storage data sequence are input into the off-grid and grid-connected data predictor, and the output results are obtained to obtain the predicted electricity demand data and the predicted grid-connected energy storage data.

3. The dynamic load regulation method of the off-grid energy storage system according to claim 2, characterized in that: According to the operation data of the energy storage system in the historical time, the off-grid data predictor is pre-trained, including: According to the energy supply data and energy storage data of all off-grid and grid-connected cycles of the energy storage system in historical time, a sample historical off-grid energy supply data sequence set and a sample historical grid-connected energy storage data sequence set are collected as input supervised training data; Identify the electricity demand data and grid-connected energy storage data in the off-grid and grid-connected period after each sample historical off-grid energy supply data sequence and sample historical grid-connected energy storage data sequence as a sample predicted electricity demand data set and a sample predicted grid-connected energy storage data set, as output supervised training data; The input supervised training data and the output supervised training data are used to supervise the training of the off-grid data predictor by using machine learning until the convergence condition is met.

4. The dynamic load regulation method of the off-grid energy storage system according to claim 1, characterized in that: Constructing the load regulation space of the energy storage system and randomly generating the first energy supply load parameter includes: Obtaining an energy supply power range of the energy storage system and constructing it as a load adjustment space; A first energy supply load parameter is randomly generated in the load regulation space.

5. The dynamic load regulation method of the off-grid energy storage system according to claim 1, characterized in that: Performing energy supply adaptability calculation according to the first energy supply load parameter and the predicted electricity demand data to obtain a first energy supply adaptability parameter, and performing aging analysis according to the first energy supply load parameter and the predicted grid-connected energy storage data to obtain a first aging parameter, including: Calculating a ratio of the first energy supply load parameter to the predicted electricity demand data to obtain a first energy supply adaptability parameter; The first energy supply load parameter and the predicted grid-connected energy storage data are input into a pre-trained aging analyzer for aging analysis to obtain a first aging rate as a first aging parameter.

6. The dynamic load regulation method of the off-grid energy storage system according to claim 5, characterized in that: The pre-training process of the aging analyzer includes: According to the aging test data of the off-grid energy storage system, a sample energy supply load parameter set and a sample grid-connected energy storage data set are collected, and the aging failure time of the energy storage system under different sample energy supply load parameters and sample grid-connected energy storage data is collected to calculate and obtain a sample aging rate set; The sample energy supply load parameter set and the sample grid-connected energy storage data set are used as input supervised training data, and the sample aging rate set is used as output supervised training data. The aging analyzer is trained using machine learning until a convergence condition is met.

7. An electronic device, characterized in that: include: Memory for storing computer software programs; A processor is used to read and execute the computer software program, thereby implementing the dynamic load regulation method of the off-grid energy storage system as described in any one of claims 1-6.

8. A computer-readable storage medium, characterized in that: The storage medium stores a computer software program, and when the computer software program is executed by the processor, the dynamic load regulation method for the off-grid energy storage system according to any one of claims 1 to 6 is implemented.

Images