Micro-grid energy storage discharge depth control method and system based on multi-objective optimization

Abstract

The invention relates to the technical field of micro-grid energy storage, and discloses a micro-grid energy storage discharge depth control method and system based on multi-objective optimization, and the method comprises the steps: determining an application scene type according to market demand data and operation mode data, determining an initial discharge interval based on the application scene type, the method comprises the steps of calculating discharge data, calculating a discharge influence coefficient according to the discharge data, adjusting an initial discharge interval according to the discharge influence coefficient, obtaining a target discharge interval, and finally optimizing a pre-constructed adjustment cost function based on a natural heuristic optimization algorithm and the target discharge interval to obtain the optimal discharge depth. The energy consumption of the energy storage system in a specific time span can be reduced, so that the overall energy consumption cost is reduced, the maximum utilization of electric energy can be ensured by the optimal discharge depth, the discharge depth with the lowest comprehensive cost is found by optimizing the energy consumption cost and the service life loss cost at the same time, and the energy utilization rate is improved. Therefore, the maximization of economic benefits is realized in the whole operation period.

Description

Technical Field

[0001] The present invention relates to the field of microgrid energy storage technology, and more specifically, to a microgrid energy storage discharge depth control method and system based on multi-objective optimization. Background Art

[0002] A microgrid refers to a small power system that can operate independently or in parallel with the main power grid. It usually includes distributed power generation resources (such as solar energy, wind energy, small gas turbines, etc.), energy storage devices, loads and control systems. Energy storage devices usually refer to energy storage power stations. Energy storage power stations store excess electricity from distributed power generation (such as solar energy, wind energy, etc.) and release it when needed to balance supply and demand and maintain the stable operation of the microgrid. The discharge of energy storage power stations is related to multiple factors, among which the discharge depth is one of the important factors. The discharge depth refers to the percentage of energy released from the full charge state to the stop of discharge in the total capacity of the energy storage system. This factor directly affects the performance and economy of the energy storage system. In the existing technology, there are related contents about the discharge depth, but there are still certain problems.

[0003] For example, the Chinese patent with publication number CN110504725A provides a method and device for fast balancing control of multiple battery stacks in an energy storage power station, which divides the battery stack state of charge into several areas, updates different state of charge and discharge depth in different areas, and gives active power instructions to each battery stack in sequence and proportion according to the updated state of charge and discharge depth. The Chinese patent with publication number CN114336694A provides an energy optimization control method for a hybrid energy storage power station, which obtains operating information of different types of energy storage systems in real time during the operation of the hybrid energy storage power station and calculates the optimal discharge depth and remaining operating life of each energy storage unit.

[0004] Although the above-mentioned patents disclose technical contents related to discharge depth, these patents mainly focus on how to determine the discharge depth according to the state of the energy storage system, but rarely optimize from the perspective of the discharge cost and market demand of the energy storage system. As a result, in actual applications, the adjustment of the discharge depth may not fully consider the economic efficiency, resulting in low economic benefits of the energy storage system.

[0005] In view of this, the present invention proposes a microgrid energy storage discharge depth control method and system based on multi-objective optimization to solve the above problems. Summary of the invention

[0006] In order to overcome the above-mentioned defects of the prior art, the present invention provides a microgrid energy storage discharge depth control method and system based on multi-objective optimization.

[0007] To achieve the above object, the present invention provides the following technical solutions: First, a microgrid energy storage discharge depth control method based on multi-objective optimization includes: Obtain market demand data and operation mode data of the target power station, determine the application scenario category according to the market demand data and operation mode data, and determine the initial discharge interval based on the application scenario category; Obtain the discharge data of the target power station, calculate the discharge influence coefficient according to the discharge data, adjust the initial discharge interval according to the discharge influence coefficient, and obtain the target discharge interval; Based on the nature-inspired optimization algorithm and the target discharge interval, the pre-constructed adjustment cost function is optimized to obtain the optimal discharge depth, the adjustment cost function at least includes an energy consumption cost function and a life depreciation function, and the optimal discharge depth is located in the target discharge interval.

[0008] Furthermore, the energy consumption cost function is obtained by quantifying the charging efficiency, discharging efficiency, charging power and discharging power within the time span, combining the adjustment weight of the target discharge depth, and dynamically calculating the energy consumption cost within the time span.

[0009] Furthermore, the life depreciation function is obtained by quantitatively calculating the charging efficiency, discharging efficiency, charging power, discharging power, and battery life loss cost caused by the grid-connected electricity price of the energy storage system during the charging and discharging process through a nonlinear amplification effect.

[0010] Furthermore, the adjustment cost function is the sum of the energy consumption cost function and the life loss function.

[0011] Furthermore, the nature-inspired optimization algorithm is a sparrow search algorithm, and the pre-constructed adjustment cost function is optimized based on the sparrow search algorithm and the target discharge interval to obtain the optimal discharge depth, including: A group of sparrow individuals are randomly generated, each of which represents a target discharge depth value; Randomly generate the initial position for each sparrow , that is, the target discharge depth value, so that it satisfies: ; In the formula, is the minimum value of the target discharge depth in the target discharge interval, is the maximum value of the target discharge depth in the target discharge interval.

[0012] Calculate the fitness function value of each sparrow individual, that is, adjust the cost function value ; Perform exploration operations on some sparrow individuals and update their target discharge depth by combining the dynamic attenuation of the number of iterations with random angle perturbations, that is, update the first A sparrow in the Generation Position , obtain the A sparrow in the Generation Position ; Check for updated Whether it is within the target discharge interval. If it is beyond the interval, boundary processing is performed; Performing a following operation on the remaining sparrow individuals, and updating the target discharge depths of the remaining sparrow individuals; Calculate the fitness function value of each sparrow individual after the update F , update the current optimal solution; When the preset maximum number of iterations is reached The search stops when the optimal solution is output, and the target discharge depth corresponding to the current optimal solution is the optimal discharge depth.

[0013] Furthermore, the method of performing boundary processing if the interval is exceeded includes: comparing the maximum value of the target discharge depth in the target discharge interval with The minimum value obtained by comparison is compared with the minimum value of the target discharge depth in the target discharge interval to obtain the maximum value after comparison, and the maximum value after comparison is used as The updated value.

[0014] Furthermore, the method for updating the target discharge depths of the remaining individual sparrows includes: updating the target discharge depth by combining a directional adjustment approaching a current optimal solution with a random step length control.

[0015] Furthermore, the method for determining the application scenario category according to the market demand data and the operation mode data includes: Input market demand data and operation mode data into the pre-built scenario classification model to obtain application scenario categories; The construction method of the scene classification model includes: Obtain W groups of training data, where W is a positive integer greater than 1, and the training data includes historical market demand data, historical operation mode data, and historical application scenario categories. The historical market demand data, historical operation mode data, and historical application scenario categories are used as sample sets, and the sample sets are divided into training sets and test sets. A classifier is constructed, and the historical market demand data and historical operation mode data in the training set are used as input data, and the historical application scenario categories in the training set are used as output data. The classifier is trained to obtain an initial classifier, and the initial classifier is tested using the test set, and a classifier that meets a preset accuracy is output as a scenario classification model.

[0016] Furthermore, the method of adjusting the initial discharge interval according to the discharge influence coefficient to obtain the target discharge interval includes: Input the discharge influence coefficient and the initial discharge interval into a pre-built interval adjustment model to obtain a target discharge interval; The method for constructing the interval adjustment model includes: A sample data set is obtained, wherein the sample data set includes a historical discharge influence coefficient, a historical initial discharge interval, and a historical target discharge interval. The sample data set is divided into a sample training set and a sample test set, and a regression network is constructed. The historical discharge influence coefficient and the historical initial discharge interval in the sample training set are used as input data of the regression network, and the historical target discharge interval in the sample training set is used as output data of the regression network. The regression network is trained to obtain an initial regression network for predicting the real-time target discharge interval. The initial regression network is tested using the sample test set, and the initial regression network that satisfies a preset error value is output as an interval adjustment model.

[0017] In the second aspect, a microgrid energy storage discharge depth control system based on multi-objective optimization is used to implement the above-mentioned microgrid energy storage discharge depth control method based on multi-objective optimization, including: The first processing module is used to obtain market demand data and operation mode data of the target power station, determine the application scenario category according to the market demand data and the operation mode data, and determine the initial discharge interval based on the application scenario category; The second processing module is used to obtain the discharge data of the target power station, calculate the discharge influence coefficient according to the discharge data, adjust the initial discharge interval according to the discharge influence coefficient, and obtain the target discharge interval; Optimization module: Based on the sparrow search algorithm and the target discharge interval, the pre-constructed adjustment cost function is optimized to obtain the optimal discharge depth. The adjustment cost function includes at least an energy consumption cost function and a life depreciation function. The optimal discharge depth is located in the target discharge interval.

[0018] Compared with the prior art, the present invention has the following beneficial effects: The present invention can determine the application scenario category according to market demand data and operation mode data, determine the initial discharge interval based on the application scenario category, calculate the discharge influence coefficient according to the discharge data, adjust the initial discharge interval according to the discharge influence coefficient, obtain the target discharge interval, and finally optimize the pre-constructed adjustment cost function based on the sparrow search algorithm and the target discharge interval to obtain the optimal discharge depth. By optimizing the discharge depth, the present invention can reduce the energy consumption of the energy storage system within a specific time span, thereby reducing the overall energy consumption cost. The optimal discharge depth can ensure the maximization of electric energy utilization. By simultaneously optimizing the energy consumption cost and the life depreciation cost, a discharge depth with the lowest comprehensive cost is found, thereby maximizing the economic benefit within the entire operation cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1It is a flow chart of a method for controlling the discharge depth of microgrid energy storage based on multi-objective optimization in the present invention; Figure 2 It is a structural schematic diagram of a microgrid energy storage discharge depth control system based on multi-objective optimization in the present invention; Figure 3 It is a schematic diagram of an optimization link of a microgrid energy storage discharge depth control system based on multi-objective optimization in the present invention; Figure 4 It is a flow chart of the method for optimizing and adjusting the cost function in the present invention. DETAILED DESCRIPTION

[0020] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0021] Example

[0022] See also Figure 1 As shown, this embodiment provides a microgrid energy storage discharge depth control method based on multi-objective optimization, including: S10: Obtain market demand data and operation mode data of the target power station, determine the application scenario category according to the market demand data and the operation mode data, and determine the initial discharge interval based on the application scenario category; In this embodiment, the market demand data at least includes future electricity prices and peak time lengths. The future electricity price refers to the average price of electricity at a future time. For example, the future electricity price may be the average price of electricity tomorrow, and the peak time length refers to the peak time length of electricity demand at a future time. It can be understood that both the future electricity price and the peak time length are obtained through corresponding historical data prediction. So, take the electricity price as an example: Methods for obtaining future electricity prices include: Get real-time electricity prices and input them into a pre-built electricity forecasting model to get future electricity prices.

[0023] The construction method of the power forecasting model includes: A historical data set, sliding step size and sliding window length are preset. The historical data set includes G groups of historical electricity price data. The historical data set is converted into multiple training samples using a sliding window method. The training samples are used as the input of the electricity prediction model. The historical electricity price data after the predicted sliding step size is used as the output. The subsequent historical electricity price data of each training sample is used as the prediction target. The prediction accuracy is used as the training target to train the electricity prediction model. A power prediction model is generated that predicts the electricity price at future moments based on the real-time electricity price. The power prediction model is an LSTM model.

[0024] It should be noted that the sliding window method is a conventional technical means of the LSTM model, and the present invention will not be further explained in principle here; however, in order to facilitate the implementation of the present invention, the present invention provides the following example of the sliding window method: Suppose we want to use the historical data [1,2,3,4,5,6] to train an LSTM model. In this embodiment, we take the prediction time span as 1 as an example, set the sliding step size to 1, and set the sliding window length to 3; then 3 groups of training samples and corresponding prediction target data are generated: [1,2,3], [2,3,4] and [3,4,5] as training samples, and [4], [5] and [6] as prediction targets respectively.

[0025] The prediction accuracy can be measured using mean square error or mean absolute error as the loss function, and the weights and biases of the model are updated through the back-propagation algorithm to generate a power prediction model.

[0026] It should be added that the above-mentioned real-time electricity price refers to the electricity price data of the previous period. For example, to predict the electricity price on August 1, the real-time electricity price can be the electricity price of the previous 30 days, that is, the electricity price in July. Similarly, the length of the peak period mentioned above is also predicted in the same way, and this embodiment will not go into details.

[0027] Specifically, the operating mode data includes but is not limited to battery ratio, discharge ambient temperature and working mode category. The battery ratio represents the ratio between lithium-ion batteries and lead-acid batteries. The discharge depth of lithium batteries is usually larger, which can reach 80%~90% or even higher, while the discharge depth of lead-acid batteries is usually smaller, generally around 50%, to extend battery life. The discharge ambient temperature can be the air temperature on the day of discharge. The discharge ambient temperature has a great influence on the performance of the battery. Under high or low temperature conditions, the discharge depth of the battery may need to be reduced to a certain extent. The working mode category represents whether the target power station is in island mode or in grid-connected mode. In grid-connected mode, the discharge depth of the target power station can be adjusted according to the status of the main power grid. In island operation mode, the target power station needs to manage the discharge depth more carefully to ensure the power supply stability during independent operation.

[0028] The application scenario categories include a first application scenario, a second application scenario, and a third application scenario. The discharge depth corresponding to the first application scenario is greater than the discharge depth corresponding to the second application scenario and greater than the discharge depth corresponding to the third application scenario. The method for determining the application scenario category according to the market demand data and the operation mode data includes: The market demand data and operation mode data are input into the pre-built scenario classification model to obtain the application scenario category.

[0029] The construction method of the scene classification model includes: Obtain W groups of training data, where W is a positive integer greater than 1, and the training data includes historical market demand data, historical operation mode data, and historical application scenario categories. The historical market demand data, historical operation mode data, and historical application scenario categories are used as sample sets, and the sample sets are divided into training sets and test sets. A classifier is constructed, and the historical market demand data and historical operation mode data in the training set are used as input data, and the historical application scenario categories in the training set are used as output data. The classifier is trained to obtain an initial classifier, and the initial classifier is tested using the test set. A classifier that meets a preset accuracy is output as a scenario classification model, and the classifier is preferably a naive Bayes model or a support vector machine model.

[0030] In this embodiment, the method for determining the initial discharge interval based on the application scenario category includes: The corresponding initial discharge interval is determined according to the application scenario category and the preset corresponding relationship.

[0031] It should be noted that, from the above content, the application scenario categories include the first application scenario, the second application scenario and the third application scenario. Each application scenario corresponds to an initial discharge interval. The discharge interval refers to the value range of the discharge depth. The initial discharge interval corresponding to the first application scenario is also larger than the initial discharge interval corresponding to the second application scenario. In this way, by classifying the application scenarios and setting different initial discharge intervals, it can be ensured that when the power station is operating in different application scenarios, its discharge depth matches the actual demand. This matching helps to optimize the use of batteries and avoid unnecessary deep discharge or shallow discharge, thereby improving the efficiency of the system. By combining market demand data with operating mode data to determine the discharge interval, the system can flexibly respond to market changes.

[0032] S20: Obtain discharge data of the target power station, calculate a discharge influence coefficient according to the discharge data, adjust an initial discharge interval according to the discharge influence coefficient, and obtain a target discharge interval; In this embodiment, the discharge data includes but is not limited to the maximum capacity value, the rated power value and the number of charge and discharge cycles. The maximum capacity value refers to the maximum available capacity of the energy storage system (such as a battery system) of the target power station. The maximum capacity value is usually expressed in kilowatt-hours (kWh), which represents the maximum electrical energy that can be stored in the battery when it is fully charged. The rated power value refers to the designed output capacity of the target power station, that is, the maximum power that the target power station can continuously provide under normal operating conditions. The number of charge and discharge cycles refers to the number of complete charge and discharge cycles experienced by the energy storage system (such as a battery) of the target power station. A complete charge and discharge cycle refers to the battery being fully discharged from a fully charged state to the minimum available capacity, and then charged back to a fully charged state.

[0033] It should be added that as the battery is used for a longer time, the maximum capacity value will gradually decay. The greater the discharge depth, the more electricity the battery releases in each cycle, which will accelerate the decay of the battery capacity. Therefore, in order to extend the battery life, the discharge depth needs to be reduced. The rated power value determines the maximum power that the power station can safely provide at any time. The larger the rated power value, the higher the discharge depth can be. The number of battery charge and discharge cycles and the discharge depth show a nonlinear relationship. Generally, the greater the discharge depth, the greater the loss of the battery in each charge and discharge cycle. Therefore, when the number of charge and discharge cycles increases, the discharge depth needs to be reduced.

[0034] The method for calculating the discharge influence coefficient according to the discharge data includes: using the number of charge and discharge cycles and the rated power value as variables, forming a numerator through a combination of an inverse sine function, an inverse cotangent function and a logarithmic function, and then using the maximum capacity value as a variable through a hyperbolic cosine function to form a denominator, and calculating the discharge influence coefficient; ; In the formula, DIF is the discharge influence coefficient, DCT is the number of charge and discharge cycles, MCV is the maximum capacity value, RPV is the rated power value, is the inverse sine function, is the hyperbolic cosine function, is the inverse cotangent function, It is a logarithmic function with base 5, and e is a natural constant.

[0035] In the discharge influence coefficient, DCT reflects the battery usage frequency. The more times, the more obvious the battery aging. The inverse sine function By performing a nonlinear transformation on the number of cycles and combining it with a logarithmic function with a base of 5, the sensitivity of the number of cycles to discharge can be amplified or adjusted. For example, when the number of cycles increases, the trend of the numerical change in this part will affect the overall size of DIF, reflecting the cumulative impact of the number of cycles on the battery discharge performance; the battery discharge power characteristics are reflected through RPV, which The inverse cotangent function can be used to convert the power value into an influencing factor of a specific trend. When discharging at high power, the output change of the inverse cotangent function will adjust the DIF to reflect the impact of the rated power on the discharge process (such as heat generation and energy loss). For example, high power may aggravate discharge loss. This function is used to incorporate the power impact into the comprehensive coefficient. MCV represents the inherent capacity of the battery and is the basic parameter of battery performance. After processing, the capacity value can be converted into an "inhibition or amplification factor" on the discharge effect. If the MCV is large, the denominator value may increase significantly, causing the overall DIF to decrease, indicating that the higher the initial capacity of the battery, the smaller the impact on performance under the same discharge conditions, reflecting the buffering effect of capacity on the discharge effect. The discharge impact coefficient can evaluate the battery health status, life loss, or performance changes during discharge by quantifying the comprehensive impact of parameters such as the number of charge and discharge cycles, rated power, and maximum capacity on the battery discharge characteristics or performance.

[0036] It can be understood that, from the above content, when the rated power value is larger, the discharge depth can be correspondingly improved, and the rated power value is negatively correlated with the discharge influence coefficient, so the discharge influence coefficient is negatively correlated with the discharge depth, and the discharge interval refers to the value range of the discharge depth, so in this embodiment, when the discharge influence coefficient is larger, the value range of the discharge depth is smaller. It can be understood that the smaller the value range of the discharge depth here is, the smaller the two endpoint values ​​of the discharge interval are, rather than the length value of the discharge interval is smaller.

[0037] The method of adjusting the initial discharge interval according to the discharge influence coefficient and obtaining the target discharge interval includes: The discharge influence coefficient and the initial discharge interval are input into the pre-built interval adjustment model to obtain the target discharge interval.

[0038] The construction methods of interval-adjusted models include: A sample data set is obtained, wherein the sample data set includes a historical discharge influence coefficient, a historical initial discharge interval, and a historical target discharge interval. The sample data set is divided into a sample training set and a sample test set, and a regression network is constructed. The historical discharge influence coefficient and the historical initial discharge interval in the sample training set are used as input data of the regression network, and the historical target discharge interval in the sample training set is used as output data of the regression network. The regression network is trained to obtain an initial regression network for predicting the real-time target discharge interval. The initial regression network is tested using the sample test set, and the initial regression network that satisfies a preset error value is output as an interval adjustment model. The initial regression network is preferably a deep neural network model.

[0039] It can be understood that the above-mentioned historical target discharge interval is pre-set based on expert experience. The setting logic is that the larger the discharge influence coefficient is, the smaller the initial discharge interval is adjusted to a smaller range. Conversely, the smaller the discharge influence coefficient is, the larger the initial discharge interval is adjusted to a larger range. The standard influence coefficient can be set in advance. When the discharge influence coefficient is greater than the standard influence coefficient, the initial discharge interval needs to be adjusted to a smaller range, otherwise it needs to be adjusted to a larger range.

[0040] S30: optimizing a pre-constructed adjustment cost function based on a nature-inspired optimization algorithm and a target discharge interval to obtain an optimal discharge depth, the adjustment cost function at least including an energy consumption cost function and a life depreciation function, and the optimal discharge depth is within the target discharge interval; In this implementation, the adjustment cost function includes at least an energy consumption cost function and a life depreciation function. The energy consumption cost function is obtained by quantifying the charging efficiency, discharging efficiency, charging power and discharging power within the time span, combining the adjustment weight of the target discharge depth, and dynamically calculating the energy consumption cost within the time span. For example, the expression of the energy consumption cost function is: ; In the formula, is the energy cost, is the unit power cost in the time span, is the average charging power within the time span, is the average discharge power within the time span, is the average charging efficiency within the time span, is the average discharge efficiency within the time span, is the time span, is the target discharge depth.

[0041] The numerator of the energy cost function is the power lost through efficiency Multiply by the unit power cost , can be directly converted into the ineffective energy consumption cost during the charging process; through the deviation power Multiply by the unit power cost , can quantify the loss of revenue caused by insufficient discharge efficiency; combined with the time span It can accumulate instantaneous power losses as the total cost within the time span, reflecting the economic impact of long-term operation; the denominator of the energy consumption cost function is combined with the target discharge depth. When the target discharge depth is large, the denominator value grows rapidly, which can reduce the impact weight of efficiency loss in the numerator and avoid cost surges due to the pursuit of high discharge depth; when the target discharge depth is small, the denominator approaches a constant, and the loss term in the numerator is significantly amplified, encouraging the algorithm to avoid overly conservative discharge strategies; through the quadratic function characteristics of the denominator, the benefits of increased discharge depth (more energy release) and the costs of efficiency loss (economic price) can be dynamically balanced to ensure that the optimization results strike a balance between technical feasibility and economy; the energy consumption cost function dynamically maps the technical operating parameters of the energy storage system into economic costs by quantifying efficiency losses and adjusting the impact of discharge depth through the denominator, providing economic indicators for multi-objective optimization and guiding the adjustment of discharge depth.

[0042] It should be noted that the time span refers to the charging and discharging time in the past period of time, because the ultimate goal of the present invention is to determine the discharge depth value from an economic perspective. For example, the present invention needs to determine the discharge depth on August 1, so the present invention needs to estimate the energy consumption cost and life depreciation cost on August 1, and determine the final discharge depth value from the perspective of reducing energy consumption cost and life depreciation cost. The time span can be the charging and discharging time in the past month, so the above-mentioned unit power cost refers to the cost corresponding to each unit power (such as per kilowatt-hour) within a certain time span, which usually includes various costs generated by the energy storage system during operation, such as operating costs, maintenance costs, etc.

[0043] It should be added that the above-mentioned average charging power is different from the average charging efficiency. The average charging power is the average power consumed during the charging process of the energy storage system within a time span. The average charging power reflects the charging intensity of the energy storage system within a certain period of time. The average charging efficiency refers to the efficiency of electric energy conversion during the charging process of the energy storage system within a time span. A higher charging efficiency means less energy loss, thereby reducing energy consumption costs. It can be understood that the average discharge power is the same as the average discharge efficiency, and this embodiment will not go into too much detail on this.

[0044] The life depreciation function is obtained by quantitatively calculating the charging efficiency, discharging efficiency, charging power, discharging power, and battery life loss cost caused by the grid-connected electricity price of the energy storage system during the charging and discharging process through the nonlinear amplification effect. For example, the expression of the life depreciation function is: ; In the formula, is the life loss cost, is the on-grid electricity price within the time span, is a constant greater than 1.

[0045] Charging side life loss term of life depreciation function The energy throughput during the charging process is reflected by charging power, time span and discharge depth, and normalized by the maximum capacity value to avoid evaluation deviations caused by capacity differences; average charging efficiency The lower the energy loss during charging, the higher the equivalent life loss. The life loss term on the charging side directly relates the charging loss cost to energy throughput and efficiency, among which high power, long-term charging or deep discharge will accelerate capacity decay. The life loss term on the discharge side The severity of the discharge operation is comprehensively reflected through the discharge power, discharge efficiency, and discharge depth. Normalization combined with the rated power value can ensure that the loss assessment is based on the rated capacity of the system; the economic benefits of the discharge behavior are coupled with the life loss through the grid-connected electricity price; the life loss term on the discharge side can increase the benefits during high electricity price periods, but it will accelerate the life decay, and a balance needs to be made between benefits and losses; the life loss term on the charging side and the life loss term on the discharge side of the life depreciation function are combined with the exponential Nonlinear amplification of loss costs can reflect the power-law decay characteristics of battery life and guide the life depreciation function to avoid high-loss operations (such as frequent deep discharge), thereby extending battery life, but this needs to be weighed against short-term economic benefits (such as discharge at high electricity prices). The life depreciation function converts battery life loss into economic costs, forcing the nature-inspired optimization algorithm to find a balance between "more discharge to make money" and "less discharge to maintain life", ultimately maximizing the benefits of the energy storage system over its entire life cycle.

[0046] It should be noted that the grid-connected electricity price refers to the price that the grid company pays to the operator of the energy storage system when the energy storage system transmits electricity to the grid. It is understandable that the grid-connected electricity price also fluctuates. In this example, the grid-connected electricity price can be the average electricity price within a time span.

[0047] The adjustment cost function is the sum of the energy consumption cost function and the life loss function, such as the adjustment cost function The expression is: ; It can be understood that when optimizing the adjustment cost function based on the sparrow search algorithm and the target discharge interval, this embodiment only uses the target discharge depth as the independent variable and the adjustment cost function value as the dependent variable. The remaining values ​​can be regarded as constants, for example, the average charging power, average charging efficiency and maximum capacity value mentioned above. The average charging power reflects the charging intensity of the energy storage system within the time span. The time span here can be a period of time in the past, such as the charging and discharging time in the past month. Therefore, the average charging power, average charging efficiency, etc. mentioned above can be calculated in advance and stored in the system. Only the target discharge depth is determined at a future time. Therefore, the purpose of optimizing the adjustment cost function in this embodiment is to determine the optimal target discharge depth within the target discharge interval.

[0048] like Figure 4 As shown, the natural inspiration optimization algorithm can be a sparrow search algorithm. The pre-constructed adjustment cost function is optimized based on the sparrow search algorithm and the target discharge interval. The sparrow search algorithm has a strong global search capability by simulating the sparrow's foraging and anti-predation behaviors. When optimizing the adjustment cost function, the solution space of the target discharge depth can be effectively traversed to avoid falling into the local optimal solution, so as to more accurately find the optimal discharge depth that minimizes (or maximizes) the cost function. The method for obtaining the optimal discharge depth includes: A group of sparrow individuals are randomly generated, each of which represents a target discharge depth value; Randomly generate the initial position for each sparrow , that is, the target discharge depth value, so that it satisfies: ; In the formula, is the minimum value of the target discharge depth in the target discharge interval, is the maximum value of the target discharge depth in the target discharge interval.

[0049] Calculate the fitness function value of each sparrow individual, that is, adjust the cost function value ; Perform exploration operations on some sparrow individuals and update their target discharge depth by combining the dynamic attenuation of the number of iterations with random angle perturbations, that is, update the first A sparrow in the Generation Position , obtain the A sparrow in the Generation Position ;like ; In the formula, For the A sparrow in the The position of the generation, For the A sparrow in the The position of the generation, is the maximum number of iterations, is a random angle, So e The exploration operation combines the dynamic attenuation of the number of iterations with random angle perturbations, and can dynamically adjust the search strategy according to the iteration process. The early stage focuses on global exploration, and the later stage focuses on local development, adapting to the dual needs of extensive search and fine optimization in the cost function optimization process; Check for updated Whether it is within the target discharge interval. If it is beyond the interval, boundary processing is performed; Performing a following operation on the remaining sparrow individuals, and updating the target discharge depths of the remaining sparrow individuals; Calculate the fitness function value of each sparrow individual after the update F , update the current optimal solution; When the preset maximum number of iterations is reached The search stops when the optimal solution is output, and the target discharge depth corresponding to the current optimal solution is the optimal discharge depth.

[0050] In this embodiment, by optimizing the discharge depth, the energy consumption of the energy storage system within a specific time span can be reduced, thereby reducing the overall energy consumption cost. The optimal discharge depth can ensure the maximum utilization of electric energy. By simultaneously optimizing the energy consumption cost and the life depreciation cost, a discharge depth with the lowest comprehensive cost is found, thereby maximizing economic benefits throughout the entire operation cycle.

[0051] If the interval is exceeded, the method for boundary processing includes: comparing the maximum value of the target discharge depth in the target discharge interval with The minimum value obtained by comparison is compared with the minimum value of the target discharge depth in the target discharge interval to obtain the maximum value after comparison, and the maximum value after comparison is used as Updated value ;like .

[0052] In this embodiment, boundary processing can ensure that the target discharge depth is always within the specified target discharge interval to prevent unreasonable results, which helps to maintain the practical feasibility of the optimization solution and avoid extreme discharge depths that damage the system.

[0053] Update the target discharge depth of the remaining sparrows The method includes: updating the target discharge depth by combining the directional adjustment approaching the current optimal solution with the random step control; ; In the formula, is a random number, is the location of the current optimal solution.

[0054] Methods for updating the current optimal solution include: If the new target discharge depth To make the fitness function value lower, the optimal solution is updated.

[0055] In this embodiment, by checking the fitness function value, it is ensured that only those solutions that can improve the optimization target (i.e., reduce the adjustment cost function value) are accepted. This method can ensure that the final optimal solution is indeed the optimal in terms of the cost function and meets actual needs.

[0056] Example 2 like Figure 2-Figure 3 As shown, this embodiment provides a microgrid energy storage discharge depth control system based on multi-objective optimization on the basis of embodiment 1, including: The first processing module is used to obtain market demand data and operation mode data of the target power station through smart meters and environmental sensors, determine the application scenario category according to the market demand data and operation mode data, and determine the initial discharge interval based on the application scenario category; In this embodiment, the market demand data at least includes future electricity prices and peak time lengths. The future electricity price refers to the average price of electricity at a future time. For example, the future electricity price may be the average price of electricity tomorrow, and the peak time length refers to the peak time length of electricity demand at a future time. It can be understood that both the future electricity price and the peak time length are obtained through corresponding historical data prediction. So, take the electricity price as an example: Methods for obtaining future electricity prices include: Get real-time electricity prices and input them into a pre-built electricity forecasting model to get future electricity prices.

[0057] The construction method of the power forecasting model includes: A historical data set, sliding step size and sliding window length are preset. The historical data set includes G groups of historical electricity price data. The historical data set is converted into multiple training samples using a sliding window method. The training samples are used as the input of the electricity prediction model. The historical electricity price data after the predicted sliding step size is used as the output. The subsequent historical electricity price data of each training sample is used as the prediction target. The prediction accuracy is used as the training target to train the electricity prediction model. A power prediction model is generated that predicts the electricity price at future moments based on the real-time electricity price. The power prediction model is an LSTM model.

[0058] Specifically, the operating mode data includes but is not limited to battery ratio, discharge ambient temperature and working mode category. The battery ratio represents the ratio between lithium-ion batteries and lead-acid batteries. The discharge depth of lithium batteries is usually larger, which can reach 80%~90% or even higher, while the discharge depth of lead-acid batteries is usually smaller, generally around 50%, to extend battery life. The discharge ambient temperature can be the air temperature on the day of discharge. The discharge ambient temperature has a great influence on the performance of the battery. Under high or low temperature conditions, the discharge depth of the battery may need to be reduced to a certain extent. The working mode category represents whether the target power station is in island mode or in grid-connected mode. In grid-connected mode, the discharge depth of the target power station can be adjusted according to the status of the main power grid. In island operation mode, the target power station needs to manage the discharge depth more carefully to ensure the power supply stability during independent operation.

[0059] The application scenario categories include a first application scenario, a second application scenario, and a third application scenario. The discharge depth corresponding to the first application scenario is greater than the discharge depth corresponding to the second application scenario and greater than the discharge depth corresponding to the third application scenario. The method for determining the application scenario category according to the market demand data and the operation mode data includes: The market demand data and operation mode data are input into the pre-built scenario classification model to obtain the application scenario category.

[0060] The construction method of the scene classification model includes: Obtain W groups of training data, where W is a positive integer greater than 1, and the training data includes historical market demand data, historical operation mode data, and historical application scenario categories. The historical market demand data, historical operation mode data, and historical application scenario categories are used as sample sets, and the sample sets are divided into training sets and test sets. A classifier is constructed, and the historical market demand data and historical operation mode data in the training set are used as input data, and the historical application scenario categories in the training set are used as output data. The classifier is trained to obtain an initial classifier, and the initial classifier is tested using the test set. A classifier that meets a preset accuracy is output as a scenario classification model, and the classifier is preferably a naive Bayes model or a support vector machine model.

[0061] The second processing module is used to obtain the discharge data of the target power station through the state monitor, calculate the discharge influence coefficient according to the discharge data, adjust the initial discharge interval through the adaptive coordinator according to the discharge influence coefficient, and obtain the target discharge interval; In this embodiment, the discharge data includes but is not limited to the maximum capacity value, the rated power value and the number of charge and discharge cycles. The maximum capacity value refers to the maximum available capacity of the energy storage system (such as a battery system) of the target power station. The maximum capacity value is usually expressed in kilowatt-hours (kWh), which represents the maximum electrical energy that can be stored in the battery when it is fully charged. The rated power value refers to the designed output capacity of the target power station, that is, the maximum power that the target power station can continuously provide under normal operating conditions. The number of charge and discharge cycles refers to the number of complete charge and discharge cycles experienced by the energy storage system (such as a battery) of the target power station. A complete charge and discharge cycle refers to the battery being fully discharged from a fully charged state to the minimum available capacity, and then charged back to a fully charged state.

[0062] The method for calculating the discharge influence coefficient according to the discharge data includes: using the number of charge and discharge cycles and the rated power value as variables, forming a numerator through a combination of an inverse sine function, an inverse cotangent function and a logarithmic function, and then using the maximum capacity value as a variable through a hyperbolic cosine function to form a denominator, and calculating the discharge influence coefficient; ; In the formula, DIF is the discharge influence coefficient, DCT is the number of charge and discharge cycles, MCV is the maximum capacity value, RPV is the rated power value, is the inverse sine function, is the hyperbolic cosine function, is the inverse cotangent function, It is a logarithmic function with base 5, and e is a natural constant.

[0063] In the discharge influence coefficient, DCT reflects the battery usage frequency. The more times, the more obvious the battery aging. The inverse sine function By performing a nonlinear transformation on the number of cycles and combining it with a logarithmic function with a base of 5, the sensitivity of the number of cycles to discharge can be amplified or adjusted. For example, when the number of cycles increases, the trend of the numerical change in this part will affect the overall size of DIF, reflecting the cumulative impact of the number of cycles on the battery discharge performance; the battery discharge power characteristics are reflected through RPV, which The inverse cotangent function can be used to convert the power value into an influencing factor of a specific trend. When discharging at high power, the output change of the inverse cotangent function will adjust the DIF to reflect the impact of the rated power on the discharge process (such as heat generation and energy loss). For example, high power may aggravate discharge loss. This function is used to incorporate the power impact into the comprehensive coefficient. MCV represents the inherent capacity of the battery and is the basic parameter of battery performance. After processing, the capacity value can be converted into an "inhibition or amplification factor" on the discharge effect. If the MCV is large, the denominator value may increase significantly, causing the overall DIF to decrease, indicating that the higher the initial capacity of the battery, the smaller the impact on performance under the same discharge conditions, reflecting the buffering effect of capacity on the discharge effect. The discharge impact coefficient can evaluate the battery health status, life loss, or performance changes during discharge by quantifying the comprehensive impact of parameters such as the number of charge and discharge cycles, rated power, and maximum capacity on the battery discharge characteristics or performance.

[0064] The method of adjusting the initial discharge interval according to the discharge influence coefficient and obtaining the target discharge interval includes: The discharge influence coefficient and the initial discharge interval are input into the pre-built interval adjustment model to obtain the target discharge interval.

[0065] The construction methods of interval-adjusted models include: A sample data set is obtained, wherein the sample data set includes a historical discharge influence coefficient, a historical initial discharge interval, and a historical target discharge interval. The sample data set is divided into a sample training set and a sample test set, and a regression network is constructed. The historical discharge influence coefficient and the historical initial discharge interval in the sample training set are used as input data of the regression network, and the historical target discharge interval in the sample training set is used as output data of the regression network. The regression network is trained to obtain an initial regression network for predicting the real-time target discharge interval. The initial regression network is tested using the sample test set, and the initial regression network that satisfies a preset error value is output as an interval adjustment model. The initial regression network is preferably a deep neural network model.

[0066] Optimization module: optimizes the pre-built adjustment cost function based on the nature-inspired optimization algorithm and the target discharge interval to obtain the optimal discharge depth, and adjusts the dual-mode inverter and the resonant converter according to the optimal discharge depth. The adjustment cost function includes at least an energy consumption cost function and a life depreciation function. The optimal discharge depth is located in the target discharge interval; In this embodiment, the adjustment cost function includes at least an energy consumption cost function and a life depreciation function. Then the energy consumption cost function is obtained by quantifying the charging efficiency, discharging efficiency, charging power and discharging power within the time span, combining the adjustment weight of the target discharge depth, and dynamically calculating the energy consumption cost within the time span; for example, the expression of the energy consumption cost function is: ; In the formula, is the energy cost, is the unit power cost in the time span, is the average charging power within the time span, is the average discharge power within the time span, is the average charging efficiency within the time span, is the average discharge efficiency within the time span, is the time span, is the target discharge depth.

[0067] The numerator of the energy cost function is the power lost through efficiency Multiply by the unit power cost , can be directly converted into the ineffective energy consumption cost during the charging process; through the deviation power Multiply by the unit power cost , can quantify the loss of revenue caused by insufficient discharge efficiency; combined with the time span It can accumulate instantaneous power losses as the total cost within the time span, reflecting the economic impact of long-term operation; the denominator of the energy consumption cost function is combined with the target discharge depth. When the target discharge depth is large, the denominator value grows rapidly, which can reduce the impact weight of efficiency loss in the numerator and avoid cost surges due to the pursuit of high discharge depth; when the target discharge depth is small, the denominator approaches a constant, and the loss term in the numerator is significantly amplified, encouraging the algorithm to avoid overly conservative discharge strategies; through the quadratic function characteristics of the denominator, the benefits of increased discharge depth (more energy release) and the costs of efficiency loss (economic price) can be dynamically balanced to ensure that the optimization results strike a balance between technical feasibility and economy; the energy consumption cost function dynamically maps the technical operating parameters of the energy storage system into economic costs by quantifying efficiency losses and adjusting the impact of discharge depth through the denominator, providing economic indicators for multi-objective optimization and guiding the adjustment of discharge depth.

[0068] The life depreciation function is obtained by quantifying the battery life loss cost caused by charging power, discharging power, charging power, discharging power, and grid-connected electricity price during the charging and discharging process of the energy storage system through the nonlinear amplification effect. For example, the expression of the life depreciation function is: ; In the formula, is the life loss cost, is the on-grid electricity price within the time span, is a constant greater than 1.

[0069] Charging side life loss term of life depreciation function The energy throughput during the charging process is reflected by charging power, time span and discharge depth, and normalized by the maximum capacity value to avoid evaluation deviations caused by capacity differences; average charging efficiency The lower the energy loss during charging, the higher the equivalent life loss. The life loss term on the charging side directly relates the charging loss cost to energy throughput and efficiency, among which high power, long-term charging or deep discharge will accelerate capacity decay. The life loss term on the discharge side The severity of the discharge operation is comprehensively reflected through the discharge power, discharge efficiency, and discharge depth. Normalization combined with the rated power value can ensure that the loss assessment is based on the rated capacity of the system; the economic benefits of the discharge behavior are coupled with the life loss through the grid-connected electricity price; the life loss term on the discharge side can increase the benefits during high electricity price periods, but it will accelerate the life decay, and a balance needs to be made between benefits and losses; the life loss term on the charging side and the life loss term on the discharge side of the life depreciation function are combined with the exponential Nonlinear amplification of loss costs can reflect the power-law decay characteristics of battery life and guide the life depreciation function to avoid high-loss operations (such as frequent deep discharge), thereby extending battery life, but this needs to be weighed against short-term economic benefits (such as discharge at high electricity prices). The life depreciation function converts battery life loss into economic costs, forcing the nature-inspired optimization algorithm to find a balance between "more discharge to make money" and "less discharge to maintain life", ultimately maximizing the benefits of the energy storage system over its entire life cycle.

[0070] It should be noted that the on-grid electricity price refers to the price that the grid company pays to the operator of the energy storage system when the energy storage system transmits electricity to the grid. It is understandable that the on-grid electricity price also fluctuates, so in this example, the on-grid electricity price can be the average electricity price within a time span; The adjustment cost function is the sum of the energy consumption cost function and the life loss function, such as the adjustment cost function The expression is: ; The above-mentioned nature-inspired optimization algorithm may be a sparrow search algorithm. Based on the sparrow search algorithm and the target discharge interval, the pre-constructed adjustment cost function is optimized to obtain the optimal discharge depth, which includes: A group of sparrow individuals are randomly generated, each of which represents a target discharge depth value; Randomly generate the initial position for each sparrow , that is, the target discharge depth value, so that it satisfies: ; In the formula, is the minimum value of the target discharge depth in the target discharge interval, is the maximum value of the target discharge depth in the target discharge interval.

[0071] Calculate the fitness function value of each sparrow individual, that is, adjust the cost function value ; Perform exploration operations on some sparrow individuals and update their target discharge depth by combining the dynamic attenuation of the number of iterations with random angle perturbations, that is, update the first A sparrow in the Generation Position , obtain the A sparrow in the Generation Position ;like ; In the formula, For the A sparrow in the The position of the generation, For the A sparrow in the The position of the generation, is the maximum number of iterations, is a random angle, So e The exponential function with base , e is a natural constant; Check for updated Whether it is within the target discharge interval. If it is beyond the interval, boundary processing is performed; Performing a following operation on the remaining sparrow individuals, and updating the target discharge depths of the remaining sparrow individuals; Calculate the fitness function value of each sparrow individual after the update F , update the current optimal solution; When the preset maximum number of iterations is reached The search stops when the optimal solution is output, and the target discharge depth corresponding to the current optimal solution is the optimal discharge depth.

[0072] In this embodiment, by optimizing the discharge depth, the energy consumption of the energy storage system within a specific time span can be reduced, thereby reducing the overall energy consumption cost. The optimal discharge depth can ensure the maximum utilization of electric energy. By simultaneously optimizing the energy consumption cost and the life depreciation cost, a discharge depth with the lowest comprehensive cost is found, thereby maximizing economic benefits throughout the entire operation cycle.

[0073] If the interval is exceeded, the method for boundary processing includes: comparing the maximum value of the target discharge depth in the target discharge interval with The minimum value obtained by comparison is compared with the minimum value of the target discharge depth in the target discharge interval to obtain the maximum value after comparison, and the maximum value after comparison is used as Updated value ;like ; In this embodiment, boundary processing can ensure that the target discharge depth is always within the specified target discharge interval to prevent unreasonable results, which helps to maintain the practical feasibility of the optimization solution and avoid extreme discharge depths that damage the system.

[0074] Update the target discharge depth of the remaining sparrows The method includes: updating the target discharge depth by combining the directional adjustment approaching the current optimal solution with the random step control; ; Where S is a random number, is the location of the current optimal solution.

[0075] Methods for updating the current optimal solution include: If the new target discharge depth To make the fitness function value lower, the optimal solution is updated.

[0076] In this embodiment, by checking the fitness function value, it is ensured that only those solutions that can improve the optimization target (i.e., reduce the adjustment cost function value) are accepted. This method can ensure that the final optimal solution is indeed the optimal in terms of the cost function and meets actual needs.

[0077] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters, weights and thresholds in the formula are set by technicians in this field according to actual conditions.

[0078] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed in the present invention can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

[0079] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0080] In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of the units is only one, and there may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0081] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0082] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0083] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

[0084] Finally: The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for controlling the discharge depth of energy storage in a microgrid based on multi-objective optimization, characterized in that: include: Obtain market demand data and operation mode data of the target power station, determine the application scenario category according to the market demand data and operation mode data, and determine the initial discharge interval based on the application scenario category; Obtain the discharge data of the target power station, calculate the discharge influence coefficient according to the discharge data, adjust the initial discharge interval according to the discharge influence coefficient, and obtain the target discharge interval; Based on the nature-inspired optimization algorithm and the target discharge interval, the pre-constructed adjustment cost function is optimized to obtain the optimal discharge depth, the adjustment cost function at least includes an energy consumption cost function and a life depreciation function, and the optimal discharge depth is located in the target discharge interval.

2. The microgrid energy storage discharge depth control method based on multi-objective optimization according to claim 1 is characterized in that: The energy consumption cost function is obtained by quantifying the charging efficiency, discharging efficiency, charging power and discharging power within a time span, combining the adjustment weight of the target discharge depth, and dynamically calculating the energy consumption cost within the time span.

3. The microgrid energy storage discharge depth control method based on multi-objective optimization according to claim 2 is characterized in that: The life depreciation function is obtained by quantitatively calculating the charging efficiency, discharging efficiency, charging power, discharging power, and battery life loss cost caused by the grid-connected electricity price of the energy storage system during the charging and discharging process through the nonlinear amplification effect.

4. The microgrid energy storage discharge depth control method based on multi-objective optimization according to claim 3 is characterized in that: The adjustment cost function is the sum of the energy consumption cost function and the life loss function.

5. The microgrid energy storage discharge depth control method based on multi-objective optimization according to claim 4 is characterized in that: The nature-inspired optimization algorithm is a sparrow search algorithm. Based on the sparrow search algorithm and the target discharge interval, the pre-constructed adjustment cost function is optimized to obtain the optimal discharge depth, including: A group of sparrow individuals are randomly generated, each of which represents a target discharge depth value; Randomly generate the initial position for each sparrow , that is, the target discharge depth value, so that it satisfies: ； In the formula, is the minimum value of the target discharge depth in the target discharge interval, is the maximum value of the target discharge depth in the target discharge interval; Calculate the fitness function value of each sparrow individual, that is, adjust the cost function value ; Perform exploration operations on some sparrow individuals and update their target discharge depth by combining the dynamic attenuation of the number of iterations with random angle perturbations, that is, update the first A sparrow in the Generation Position , obtain the A sparrow in Generation Position ; Check for updated Whether it is within the target discharge interval. If it is beyond the interval, boundary processing is performed; Performing a following operation on the remaining sparrow individuals, and updating the target discharge depths of the remaining sparrow individuals; Calculate the fitness function value of each sparrow individual after the update F , update the current optimal solution; When the preset maximum number of iterations is reached The search stops when the optimal solution is output, and the target discharge depth corresponding to the current optimal solution is the optimal discharge depth.

6. The microgrid energy storage discharge depth control method based on multi-objective optimization according to claim 5 is characterized in that: The method for performing boundary processing if the interval is exceeded includes: comparing the maximum value of the target discharge depth in the target discharge interval with The minimum value obtained by comparison is compared with the minimum value of the target discharge depth in the target discharge interval to obtain the maximum value after comparison, and the maximum value after comparison is used as The updated value.

7. The microgrid energy storage discharge depth control method based on multi-objective optimization according to claim 5 is characterized in that: The method for updating the target discharge depths of the remaining sparrow individuals includes: updating the target discharge depth by combining a directional adjustment approaching a current optimal solution with a random step length control.

8. The microgrid energy storage discharge depth control method based on multi-objective optimization according to claim 1 is characterized in that: The method for determining the application scenario category according to market demand data and operation mode data includes: Input market demand data and operation mode data into the pre-built scenario classification model to obtain application scenario categories; The method for constructing the scene classification model includes: Obtain W groups of training data, where W is a positive integer greater than 1, and the training data includes historical market demand data, historical operation mode data, and historical application scenario categories. The historical market demand data, historical operation mode data, and historical application scenario categories are used as sample sets, and the sample sets are divided into training sets and test sets. A classifier is constructed, and the historical market demand data and historical operation mode data in the training set are used as input data, and the historical application scenario categories in the training set are used as output data. The classifier is trained to obtain an initial classifier, and the initial classifier is tested using the test set, and a classifier that meets a preset accuracy is output as a scenario classification model.

9. The microgrid energy storage discharge depth control method based on multi-objective optimization according to claim 1 is characterized in that: The method of adjusting the initial discharge interval according to the discharge influence coefficient to obtain the target discharge interval includes: Input the discharge influence coefficient and the initial discharge interval into a pre-built interval adjustment model to obtain a target discharge interval; The method for constructing the interval adjustment model includes: A sample data set is obtained, wherein the sample data set includes a historical discharge influence coefficient, a historical initial discharge interval, and a historical target discharge interval. The sample data set is divided into a sample training set and a sample test set, and a regression network is constructed. The historical discharge influence coefficient and the historical initial discharge interval in the sample training set are used as input data of the regression network, and the historical target discharge interval in the sample training set is used as output data of the regression network. The regression network is trained to obtain an initial regression network for predicting the real-time target discharge interval. The initial regression network is tested using the sample test set, and the initial regression network that satisfies a preset error value is output as an interval adjustment model.

10. A microgrid energy storage discharge depth control system based on multi-objective optimization, which is used to implement the microgrid energy storage discharge depth control method based on multi-objective optimization described in any one of claims 1-9, characterized in that: include: The first processing module is used to obtain market demand data and operation mode data of the target power station, determine the application scenario category according to the market demand data and the operation mode data, and determine the initial discharge interval based on the application scenario category; The second processing module is used to obtain the discharge data of the target power station, calculate the discharge influence coefficient according to the discharge data, adjust the initial discharge interval according to the discharge influence coefficient, and obtain the target discharge interval; Optimization module: Based on the sparrow search algorithm and the target discharge interval, the pre-constructed adjustment cost function is optimized to obtain the optimal discharge depth. The adjustment cost function includes at least an energy consumption cost function and a life depreciation function. The optimal discharge depth is located in the target discharge interval.

Images