Battery pack charge and discharge regulation method and device based on capacity verification

Abstract

The present invention discloses a battery pack charge and discharge regulation method and device based on capacity verification, which relates to the technical field of battery charge and discharge regulation. The method includes: when the energy storage battery pack is discharging, obtaining a load data set, performing load prediction, and obtaining a predicted load within a predetermined period; predetermining a discharge strategy, performing discharge control on the energy storage battery pack, and collecting battery discharge data in real time; obtaining a three-dimensional comparison scheme; performing abnormality identification, constructing a battery discharge abnormality distribution, performing parameter optimization analysis, and performing discharge rectification regulation. The present invention solves the technical problem that the existing energy storage battery pack charge and discharge control method cannot accurately identify battery charge and discharge abnormality data, resulting in the inability to set the corresponding battery rectification parameters in a targeted manner, resulting in poor accuracy and reliability of battery pack charge and discharge regulation, and achieves the technical effect of realizing comprehensive and accurate management of the charge and discharge of the energy storage battery pack, ensuring safe and stable operation of the battery and extending its service life.

Description

Technical Field

[0001] The present invention relates to the technical field of battery charge and discharge regulation, and in particular to a battery pack charge and discharge regulation method and device based on capacity verification. Background Art

[0002] With the widespread application of renewable energy and the development of smart grids, energy storage technology is becoming increasingly important in power systems. As an efficient energy storage method, energy storage battery packs are widely used in various scenarios. However, in practical applications, the charge and discharge management of energy storage battery packs faces many challenges. On the one hand, accurate prediction of load demand is crucial for the reasonable configuration of the discharge strategy of energy storage battery packs. Traditional load prediction methods are often not very accurate and difficult to meet actual needs. At the same time, different load ends have different power consumption characteristics. How to comprehensively consider the historical load logs of multiple load ends for accurate load prediction is an urgent problem to be solved. On the other hand, battery packs are prone to various abnormal conditions during the charging and discharging process, such as excessive voltage differences between single cells, excessive temperatures, abnormal discharge trends, etc. Traditional abnormality detection methods are usually relatively simple and it is difficult to fully and accurately identify various abnormal conditions. In addition, there is a lack of effective means for handling abnormal conditions. Often, only simple alarms can be made, and parameter optimization and rectification adjustment cannot be carried out in time to ensure the safe and stable operation of the battery pack.

[0003] There is a technical problem in the existing energy storage battery pack charge and discharge control method that it is impossible to accurately identify abnormal battery charge and discharge data, resulting in the inability to set appropriate battery rectification parameters in a targeted manner, causing poor accuracy and reliability of battery pack charge and discharge regulation. Summary of the invention

[0004] The present application provides a battery pack charge and discharge regulation method and device based on capacity verification, which is used to solve the technical problem that the existing energy storage battery pack charge and discharge control method cannot accurately identify abnormal battery charge and discharge data, resulting in the inability to set appropriate battery rectification parameters in a targeted manner, resulting in poor accuracy and reliability of battery pack charge and discharge regulation.

[0005] In view of the above problems, the present application provides a battery pack charge and discharge regulation method and device based on capacity verification.

[0006] In a first aspect of the present application, a battery pack charge and discharge regulation method based on capacity verification is provided, the method comprising:

[0007] When the energy storage battery pack is discharging, a load data set is obtained at a fixed point by using the first monitoring unit, and a load prediction is performed according to the load data set to obtain a predicted load within a predetermined period; a predetermined discharge strategy is configured based on the predicted load, and the discharge of the energy storage battery pack is controlled, and the battery discharge data is collected in real time by using the second monitoring unit during the discharge cycle; a three-dimensional comparison scheme is obtained, wherein the three-dimensional comparison scheme includes a single deviation comparison strategy, a trend deviation comparison strategy, and a twin deviation comparison strategy; based on the single deviation comparison strategy, the trend deviation comparison strategy, and the twin deviation comparison strategy, anomalies are identified on the battery discharge data, and a battery discharge abnormality distribution is constructed; a parameter optimization analysis is performed on the battery discharge abnormality distribution by using a first rectifier corrector, and a first rectification parameter is output to perform discharge rectification adjustment on the energy storage battery pack.

[0008] In a second aspect of the present application, a battery pack charge and discharge regulation device based on capacity verification is provided, the device comprising:

[0009] A predicted load acquisition module, which is used to use the first monitoring unit to obtain a load data set at a fixed point when the energy storage battery group is discharging, perform load prediction based on the load data set, and obtain the predicted load within a predetermined period; a discharge control module, which configures a predetermined discharge strategy based on the predicted load, performs discharge control on the energy storage battery group, and uses the second monitoring unit to collect battery discharge data in real time during the discharge cycle; a three-dimensional comparison scheme acquisition module, which is used to obtain a three-dimensional comparison scheme, wherein the three-dimensional comparison scheme includes a single deviation comparison strategy, a trend deviation comparison strategy, and a twin deviation comparison strategy; a battery discharge abnormality distribution construction module, which is based on the single deviation comparison strategy, the trend deviation comparison strategy, and the twin deviation comparison strategy, to perform abnormal identification on the battery discharge data and construct a battery discharge abnormality distribution; a discharge rectification adjustment module, which is used to perform parameter optimization analysis on the battery discharge abnormality distribution through a first rectifier corrector, and output a first rectification parameter to perform discharge rectification adjustment on the energy storage battery group.

[0010] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0011] When the energy storage battery pack is discharging, the first monitoring unit is used to obtain the load data set at a fixed point, perform load prediction, and obtain the predicted load within a predetermined period; based on the predicted load, a predetermined discharge strategy is configured to control the discharge of the energy storage battery pack, and collect battery discharge data in real time; a three-dimensional comparison scheme is obtained; abnormalities are identified on the battery discharge data, and abnormal battery discharge distribution is constructed; the first rectifier corrector is used to perform parameter optimization analysis on the abnormal battery discharge distribution, and the first rectification parameter is output to adjust the discharge rectification of the energy storage battery pack. The technical effect of achieving comprehensive and accurate management of the charging and discharging of the energy storage battery pack, ensuring the safe and stable operation of the battery and extending the service life is achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0013] Figure 1 A schematic flow chart of a battery pack charge and discharge regulation method based on capacity verification provided in an embodiment of the present application;

[0014] Figure 2 A schematic diagram of the structure of a battery pack charge and discharge regulation device based on capacity verification provided in an embodiment of the present application.

[0015] Explanation of the reference numerals: predicted load acquisition module 10 , discharge control module 20 , three-dimensional comparison scheme acquisition module 30 , battery discharge abnormal distribution construction module 40 , discharge rectification adjustment module 50 . DETAILED DESCRIPTION

[0016] The present application provides a battery pack charge and discharge regulation method and device based on capacity verification, which is used to solve the technical problem that the existing energy storage battery pack charge and discharge control method cannot accurately identify abnormal battery charge and discharge data, resulting in the inability to set appropriate battery rectification parameters in a targeted manner, resulting in poor accuracy and reliability of battery pack charge and discharge regulation.

[0017] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0018] Embodiment 1, as Figure 1As shown, the present application provides a battery pack charge and discharge regulation method based on capacity verification, the method comprising:

[0019] Step S100: When the energy storage battery pack is discharging, a load data set is obtained at a fixed point using a first monitoring unit, and load prediction is performed according to the load data set to obtain a predicted load within a predetermined period.

[0020] Specifically, when the energy storage battery pack is discharged, the first monitoring unit starts working. The first monitoring unit collects fixed-point data from multiple load ends (residential areas, electric vehicle charging stations, industrial equipment, etc.) to obtain a load data set. The load data set is preprocessed, the data of different load ends are sorted in chronological order, the integrity and accuracy of the data are checked, and missing values ​​and outliers are processed. Then, the ARIMA model is used for load forecasting. The specific steps are as follows: Stationarity test, the stationarity test of the load time series data can be carried out, and the unit root test (such as ADF test) can be used. If the data is not stationary, differential processing is performed until the data is stationary. Model order determination, determine the order (p, d, q) of the ARIMA model. Among them, p is the order of the autoregressive term, d is the number of differences, and q is the order of the moving average term. By observing the graphs of the autocorrelation function (ACF) and the partial autocorrelation function (PACF), combined with information criteria (such as AIC, BIC), the appropriate order is determined. Parameter estimation, the parameters of the selected ARIMA model are estimated using methods such as maximum likelihood estimation. Model diagnosis: diagnose the estimated model and check whether the residual is white noise. If the residual is not white noise, it is necessary to readjust the model order or make other improvements. Finally, use the trained ARIMA model to predict the load within the predetermined period to obtain the predicted load.

[0021] In this way, by using the load data set obtained by the first monitoring unit and combining it with the ARIMA model for load forecasting, a relatively accurate predicted load within a predetermined period can be obtained, providing strong support for the discharge strategy of the energy storage battery group.

[0022] Step S200: configuring a predetermined discharge strategy based on the predicted load, performing discharge control on the energy storage battery group, and collecting battery discharge data in real time using a second monitoring unit during a discharge cycle.

[0023] Specifically, first, the predetermined discharge strategy is configured according to the predicted load obtained above. If the predicted load is high, in order to meet the power demand, the following discharge strategy can be adopted: increase the discharge current of the energy storage battery group to release electric energy at a faster rate. At the same time, pay close attention to the temperature, voltage and other parameters of the battery to ensure operation within a safe range. For example, a higher upper limit of the discharge current is set, but when the battery temperature exceeds a certain threshold, the discharge current is automatically reduced to prevent overheating and damage to the battery. If the predicted load is low, a more conservative discharge strategy is adopted, such as reducing the discharge current to extend the service life of the battery. At the same time, the discharge voltage of the battery group is adjusted according to the actual situation to minimize energy waste while meeting the load demand. When the energy storage battery group is discharged and controlled, it can be achieved through an intelligent control system, which can monitor various parameters of the battery in real time, such as current, voltage, temperature, etc., and adjust according to the predetermined discharge strategy. For example, when it is detected that the discharge current exceeds the set upper limit, the current output is automatically reduced; when the battery temperature is too high, the heat dissipation system is started or the discharge power is reduced. During the discharge cycle, the second monitoring unit is used to collect battery discharge data in real time. The second monitoring unit includes various sensors, such as current sensors, voltage sensors, temperature sensors, etc. These sensors monitor the discharge status of the battery in real time and transmit the data to the control system for analysis. The collected battery discharge data include discharge voltage, discharge current, battery temperature, discharge rate and capacity ratio, etc. These data are very important for evaluating the performance of the battery, judging whether there are abnormal conditions, and optimizing the discharge strategy. For example, by real-time monitoring of the discharge current and voltage, the real-time power output of the battery can be calculated; by monitoring the battery temperature, overheating can be detected in time and corresponding measures can be taken; by monitoring the capacity ratio, the remaining power of the battery can be understood in order to arrange the discharge plan reasonably. Through the above steps, effective discharge control of the energy storage battery pack is achieved, and the battery discharge data is collected in real time to provide a basis for subsequent analysis and optimization.

[0024] Step S300: Obtain a three-dimensional comparison scheme, wherein the three-dimensional comparison scheme includes a monomer deviation comparison strategy, a trend deviation comparison strategy, and a twin deviation comparison strategy.

[0025] Specifically, the three-dimensional comparison scheme obtained is an important means to conduct a comprehensive analysis of the energy storage battery pack. This scheme includes a single deviation comparison strategy, a trend deviation comparison strategy, and a twin deviation comparison strategy. The single deviation comparison strategy mainly focuses on the difference between a single battery and other batteries. In the energy storage battery pack, due to the influence of factors such as the manufacturing process and the use environment, the performance of each single battery will be different. By comparing the discharge voltage, discharge current, battery temperature, discharge rate, and capacity ratio of each single battery with other single batteries, single batteries with abnormal performance are found. For example, if the voltage of a single battery is significantly lower than that of other batteries, or the temperature is too high, there is a problem and further inspection and processing is required.

[0026] The trend deviation comparison strategy focuses on analyzing the historical trend of battery discharge data. By collecting battery discharge data over a period of time, including the changes in parameters such as voltage, current, and temperature over time, a trend model of battery performance is established. Then, the current discharge data is compared with the historical trend. If the current data is found to deviate significantly from the historical trend, it means that the battery pack has an abnormality. For example, if the battery discharge rate suddenly increases, or the capacity ratio decreases faster, it is due to battery aging, changes in the external environment, or other factors.

[0027] The twin deviation comparison strategy uses the battery's attribute characteristics to build a twin model. First, the battery attribute characteristics of the energy storage battery pack are obtained, such as battery type, structure, electrical parameters, and distribution topology. Then, a virtual twin model is built based on these characteristics. The model can simulate the performance of the battery pack under different conditions. The actual battery discharge data is input into the twin model. By comparing the difference between the model prediction results and the actual data, abnormal situations can be identified. For example, if the voltage predicted by the twin model is significantly different from the actual voltage, it means that there is an internal fault or external interference in the battery.

[0028] By comprehensively applying these three comparison strategies, the energy storage battery pack can be fully monitored and analyzed, abnormal situations can be discovered in time, and the safety and reliability of the battery pack can be improved.

[0029] Step S400: Based on the single-cell deviation comparison strategy, the trend deviation comparison strategy and the twin deviation comparison strategy, anomaly identification is performed on the battery discharge data to construct a battery discharge anomaly distribution.

[0030] Specifically, the single-cell deviation comparison strategy, trend deviation comparison strategy and twin deviation comparison strategy are used to identify abnormalities in battery discharge data, and then the abnormal distribution of battery discharge is constructed. First, the single-cell deviation comparison strategy is used to compare the discharge voltage, discharge current, battery temperature, discharge rate and capacity ratio of each single cell in the battery discharge data with the corresponding parameters of other single cells. For example, the mean and standard deviation of each parameter are calculated. If the parameter value of a single cell exceeds the range of the mean value plus or minus a certain multiple of the standard deviation, it is considered that the single cell has an abnormality in this parameter. By comparing each parameter one by one, the single cells with abnormalities can be preliminarily screened out to form the first abnormal discharge data set.

[0031] Next, a trend deviation comparison strategy is adopted. The historical battery discharge data is read, and the discharge trend is analyzed. Through time series analysis and other methods, the change trend of the battery discharge data in different time periods is extracted. Then, the current battery discharge data is compared with the historical discharge trend. If the current data deviates significantly from the historical trend, such as a sudden and significant change in the discharge rate or an abnormally accelerated rate of decrease in the capacity ratio, the data point is considered to be abnormal and added to the second abnormal discharge data set.

[0032] Finally, the twin deviation comparison strategy is used. The battery property characteristics of the energy storage battery pack are obtained, including battery type, battery structure, electrical parameters, and battery distribution topology. According to these characteristics, a battery twin model is simulated to build a battery that can simulate the normal discharge behavior of the battery under different working conditions. The predicted load is rendered into the battery twin model to make it closer to the actual operating conditions. Then, the actual battery discharge data is input into the twin model to compare the difference between the model prediction results and the actual data. If the difference is large, for example, the voltage predicted by the twin model differs from the actual voltage by more than a certain threshold, the single cell is considered to be abnormal, and its data is added to the third abnormal discharge data set.

[0033] The above three abnormal discharge data sets are combined to obtain a comprehensive abnormal discharge data set. Then, this data set is arranged based on the abnormal battery number and battery distribution topology to construct the battery discharge abnormality distribution. This distribution clearly shows which single cells have abnormalities in which parameters, as well as the distribution of abnormalities in the entire battery pack, providing an important basis for subsequent parameter optimization analysis and discharge rectification regulation.

[0034] Step S500: performing parameter optimization analysis on the abnormal battery discharge distribution through a first rectifier corrector, and outputting a first rectification parameter to perform discharge rectification regulation on the energy storage battery group.

[0035] Specifically, the first rectifier plays a key role. First, the first rectifier takes the battery discharge abnormal distribution as input data. This abnormal distribution shows in detail the abnormal conditions of each single cell in the energy storage battery pack during the discharge process, including the number of the abnormal battery and the specific abnormal state. When the first rectifier starts working, it will conduct an in-depth parameter optimization analysis of the battery discharge abnormal distribution. It is based on the property characteristics of the battery, considers factors such as battery type, battery structure, electrical parameters and battery distribution topology, and uses algorithms and models to determine the optimal rectification parameters. By analyzing the abnormal conditions of parameters such as voltage, current, and temperature in the abnormal distribution, combined with the characteristics of the battery, the optimal rectification parameter values ​​that can correct these abnormalities are calculated. After a series of analysis and calculation, the first rectifier outputs the first rectification parameters, which will be used to adjust the discharge rectification of the energy storage battery pack. When the energy storage battery pack is abnormal during the discharge process, the first rectification parameters adjust the output voltage, current and other parameters of the battery pack to return it to a normal working state. In this way, the first rectifier can effectively correct the battery discharge abnormality and ensure the safe, stable and efficient operation of the energy storage battery pack.

[0036] In a possible implementation, step S100 further includes:

[0037] Step S110: read the load monitoring node, call multiple historical load logs of multiple load ends based on the load monitoring node and the predetermined period matching, and calculate multiple historical starting loads and multiple historical predicted loads.

[0038] Step S120: Based on the multiple historical load logs, perform load trend analysis within a predetermined window to obtain multiple load trends.

[0039] Step S130: performing deviation mapping calculation on the load data set and the multiple historical starting point loads to obtain multiple load deviations.

[0040] Step S140: mapping and compensating the multiple historical predicted loads according to the multiple load trends and the multiple load deviations to obtain multiple standard predicted loads, and summing them up to obtain the predicted load.

[0041] Specifically, read the load monitoring node, which is the key entry point for obtaining load information, through which the relevant data of each load end can be accessed. Based on the load monitoring node and the predetermined period, start matching and calling multiple historical load logs of multiple load ends, which include different scenarios such as residential areas, electric vehicle charging stations, and industrial equipment. The historical load logs record the power usage of each load end in different time periods in the past. For the obtained historical load logs, further calculate multiple historical starting loads, where the historical starting load refers to the load mean at the starting moment. Select a specific starting moment in the historical load log, count the load values ​​of each load end at that moment, and then calculate its mean to obtain the historical starting load, which can reflect the average power demand of each load end at a specific starting moment. At the same time, multiple historical predicted loads are also calculated. This process first removes abnormal loads in multiple loads in the same historical time period. For example, if a load value that is significantly higher or lower than the normal range appears in a certain historical time period, it is caused by an abnormal event and is removed to ensure the accuracy of the data. Then, the load value after removing the anomalies is averaged to obtain the historical predicted load. This value is a predictive estimate of the load in the same historical time period in the past, providing a reference for subsequent load forecasts.

[0042] First, determine the size of the predetermined window, such as setting it to one week. From multiple historical load logs, select historical data of a period of time, and arrange the historical load data of each load end in chronological order. Take one week as the window, calculate the average value of the load data in the first window, and this average value represents a preliminary estimate of the load trend in this window. Then, move the window back by one time unit (such as one day), and calculate the load average value in the new window again. In this way, the window is continuously moved and the average value is calculated. In this way, a series of moving average values ​​are obtained. If these moving average values ​​show a gradual upward trend over time, it can be judged that the load trend of the load end during this period is increasing; if the moving average value gradually decreases, the load trend is decreasing; if the moving average value fluctuates within a smaller range, the load trend is relatively stable. Repeat the above process for multiple load ends to obtain multiple load trends. These load trends provide important basis for the subsequent discharge control of energy storage battery groups. For example, if the load trend of a certain load end is increasing, when configuring the discharge strategy of the energy storage battery group, the discharge power of the load end is appropriately increased to meet the possible increase in power demand in the future. On the contrary, if the load trend is decreasing, the discharge power can be appropriately reduced to save energy and extend battery life.

[0043] The two key elements of the load data set and multiple historical starting loads are clarified. The load data set is the set of load information currently obtained, and the historical starting load is the average power demand of each load end at a specific starting time, which is calculated from the historical load log. When performing deviation mapping calculations, the difference between the two is calculated. For each load end, the corresponding load value in the current load data set is subtracted from the corresponding historical starting load value to obtain a deviation value. This deviation value reflects the degree of difference between the current load and the historical starting load. These load deviations help understand the deviation between the current load and the historical load at the starting time, and provide an important basis for the subsequent mapping compensation of the historical predicted load. If the deviation is large, it means that the current load is significantly different from the historical situation, and the historical predicted load needs to be adjusted to a large extent; if the deviation is small, a small adjustment can be made or no adjustment can be made.

[0044] According to multiple load trends and multiple load deviations, multiple historical forecast loads are mapped and compensated to obtain multiple standard forecast loads, and finally the forecast load is obtained by summing up. Linear analysis compensation is a more intuitive method. First, determine the weight of the load trend and load deviation on the historical forecast load. For example, according to experience or data analysis, when the load trend is rising, a positive weight coefficient is given, indicating that a certain proportion of the load value needs to be increased on the basis of the historical forecast load; when the load trend is falling, a negative weight coefficient is given, indicating that the historical forecast load needs to be reduced. For the load deviation, another weight coefficient is determined according to the size of the deviation. The larger the deviation, the larger the weight coefficient, indicating that a larger adjustment is required to the historical forecast load. Then, for each historical forecast load, a linear calculation is performed according to the corresponding load trend and load deviation weight coefficient. For example, if a historical forecast load is 100 kilowatts, the load trend rising weight coefficient is 0.2, the load deviation weight coefficient is 0.1, and the load deviation is a positive deviation, indicating that the current load is higher than the historical average level. Then the load value after linear compensation is 100+100×0.2+100×0.1=130 kilowatts. In this way, each historical forecast load is calculated to obtain multiple standard forecast loads. Finally, these standard forecast loads are summed to obtain the forecast load. The historical forecast load is mapped and compensated more accurately, the accuracy of the forecast load is improved, and a more reliable basis is provided for the discharge strategy of the energy storage battery pack.

[0045] In a possible implementation, step S200 further includes:

[0046] Step S210: reading the real-time remaining capacity of the energy storage battery pack.

[0047] Step S220: configuring over-discharge constraints of the energy storage battery pack.

[0048] Step S230: subjecting the over-discharge constraint and the predicted load to the limitation and aiming at minimizing energy consumption, a linear optimization of discharge parameters is performed based on the real-time remaining capacity, and the predetermined discharge strategy is output.

[0049] Specifically, in order to accurately read the real-time remaining capacity of the energy storage battery pack, it is necessary to use advanced battery management systems and sensor technologies. These devices can monitor the voltage, current, temperature and other parameters of each single cell in the battery pack in real time, and calculate the real-time remaining capacity of the battery pack. In terms of capacity verification, the real-time remaining capacity read is compared with the preset standard capacity. The standard capacity is the rated capacity of the battery pack. By comparison, it can be determined whether the current capacity status of the battery pack is normal. If there is a large deviation between the real-time remaining capacity and the standard capacity, the system will issue an alarm to remind relevant personnel to check and handle it. Standard matching involves matching the real-time remaining capacity read with different usage scenarios and needs. For example, in some specific application scenarios, it is necessary to determine the minimum remaining capacity requirements of the battery pack based on the size and duration of the load, and to determine whether the current real-time remaining capacity can meet the actual needs based on these standards. If not, it is necessary to adjust the battery pack usage strategy, such as reducing the load and charging. In addition, capacity verification and standard matching can also help optimize the charge and discharge management of the battery pack. By accurately grasping the real-time remaining capacity, a more reasonable charging and discharging strategy can be formulated to avoid overcharging and overdischarging and extend the service life of the battery pack.

[0050] Configuring over-discharge constraints for energy storage battery packs involves capacity verification and standard matching to ensure safe and stable operation of the battery pack. In terms of capacity verification, it is first necessary to clarify the rated capacity and current real-time remaining capacity of the energy storage battery pack. By comparing these two capacity values, the discharge degree of the battery pack is determined. When configuring over-discharge constraints, the performance changes and potential risks of the battery under different discharge degrees should be taken into account. For example, when the remaining capacity of the battery pack approaches a specific threshold, the discharge needs to be more strictly restricted to prevent irreversible damage to the battery caused by over-discharge. Standard matching also plays a key role in this process. There are standards and specifications for over-discharge of energy storage battery packs in the industry. These standards are based on a large number of experiments and practical application experience. When configuring over-discharge constraints, the characteristics of the energy storage battery pack need to be matched with these standards. For example, according to the type, specification and use environment of the battery, determine the appropriate over-discharge voltage threshold or the lower limit of the remaining capacity. At the same time, you can also refer to the successful application cases of similar battery packs and learn from their over-discharge constraint setting methods. In order to achieve effective capacity verification and standard matching, the following methods can be used. On the one hand, the advanced battery management system is used to monitor the capacity and discharge status of the battery pack in real time to ensure that relevant data can be accurately obtained. On the other hand, the battery pack is regularly tested for capacity and evaluated for performance to verify the rationality and effectiveness of the over-discharge constraint. If it is found that the actual operating conditions do not match the preset constraints, the constraint parameters should be adjusted in time to adapt to the changes in the battery pack.

[0051] In order to output the predetermined discharge strategy, it is necessary to comprehensively consider multiple factors to perform linear optimization of discharge parameters. First of all, it is important to meet the over-discharge constraint. This means that in the entire optimization process, it is necessary to ensure that the discharge strategy does not cause over-discharge of the energy storage battery pack, thereby protecting the life and performance of the battery pack. By monitoring and analyzing the real-time remaining capacity, combined with the preset over-discharge constraint, the lower limit of discharge is determined to prevent the battery pack from exceeding the safe range during discharge. At the same time, the predicted load is also a key limiting factor. According to the prediction of the load in the future, the discharge strategy needs to be able to meet these load requirements to ensure the normal operation of the system. This requires that the load changes be fully considered in the optimization process, and the discharge power and time of the battery pack be reasonably arranged to ensure that energy waste is reduced as much as possible while meeting the load requirements. Minimizing energy consumption is the core goal of the entire linear optimization of discharge parameters. In this process, the linear programming algorithm is used to find the optimal combination of discharge parameters. The linear programming algorithm can find a solution that minimizes energy consumption by optimizing the objective function under the premise of meeting various constraints. Specifically, it will adjust and optimize parameters such as discharge power and discharge time according to conditions such as real-time remaining capacity, over-discharge constraints and predicted load to minimize energy consumption. When performing linear optimization, a mathematical model must be established first. This model will include an objective function and various constraints. The objective function is an expression of energy consumption, such as the sum of the product of discharge power and discharge time. The constraints include over-discharge constraints, predicted load limits, and the physical characteristics of the battery pack. Then, this mathematical model is solved by a linear programming algorithm to obtain the optimal combination of discharge parameters, i.e., the predetermined discharge strategy. By using a linear programming algorithm to perform linear optimization of discharge parameters with the goal of minimizing energy consumption and satisfying over-discharge constraints and predicted loads, and outputting a predetermined discharge strategy, the efficiency and life of the energy storage battery pack can be effectively improved while ensuring the stable operation of the system.

[0052] In a possible implementation, step S200 further includes:

[0053] Step S240: the battery discharge data includes discharge data of a plurality of single cells, and the discharge data of the single cells includes discharge voltage, discharge current, battery temperature, discharge rate and capacity ratio.

[0054] Specifically, the specific composition of battery discharge data is clarified. The battery discharge data includes the discharge data of multiple single cells. For each single cell, its discharge data includes discharge voltage, discharge current, battery temperature, discharge rate and capacity ratio.

[0055] The discharge voltage refers to the output voltage value of a single cell during the discharge process. Different single cells have different discharge voltages due to their performance differences, usage time and other factors.

[0056] The discharge current is the current output by a single cell when discharging. The discharge current will affect the discharge power and discharge speed of the battery.

[0057] The battery temperature reflects the heat generated by the single cell during the discharge process. Excessive temperature will affect the performance and life of the battery, and even cause safety problems.

[0058] The discharge rate indicates how quickly a single cell discharges, and is usually measured by the amount of electricity discharged per unit time.

[0059] The capacity ratio is the actual battery capacity ratio, that is, the ratio of the actual capacity of the current single battery to the nominal capacity. This parameter can reflect the remaining capacity and usage status of the single battery.

[0060] By collecting and analyzing the discharge data of these single cells, we can fully understand the discharge conditions of each single cell in the energy storage battery pack, and provide detailed data support for subsequent abnormality identification and discharge control.

[0061] In a possible implementation, step S400 further includes:

[0062] Step S410: performing abnormal identification on the battery discharge data based on the cell deviation comparison strategy to obtain a first abnormal discharge data set, wherein the abnormal discharge data includes an abnormal battery number and an abnormal state.

[0063] Step S420: reading historical battery discharge data to perform discharge trend analysis, and performing abnormality identification on the battery discharge data based on a trend deviation comparison strategy and historical discharge trends to obtain a second abnormal discharge data set.

[0064] Step S430: Obtain battery attribute characteristics of the energy storage battery pack, simulate and build a battery twin model according to the battery attribute characteristics, and render the predicted load to the battery twin model, and identify abnormalities of the battery discharge data based on the battery twin model and the twin deviation comparison strategy to obtain a third abnormal discharge data set, wherein the battery attribute characteristics include battery type, battery structure, electrical parameters, battery distribution topology, and rectifier distribution topology.

[0065] Step S440: performing a union operation on the first abnormal discharge data set, the second abnormal discharge data set and the third abnormal discharge data set to obtain an abnormal discharge data set, and arranging the abnormal discharge data set based on the abnormal battery numbers and battery distribution topology to generate the battery discharge abnormality distribution.

[0066] Specifically, a single-cell deviation comparison strategy is used to identify abnormalities in battery discharge data. The discharge data of each single cell, including parameters such as discharge voltage, discharge current, battery temperature, discharge rate and capacity ratio, are compared with the corresponding parameters of other single cells. If the parameters of a single cell are significantly different from those of most other single cells, it is considered that the single cell is abnormal. For example, if the voltage of a single cell is significantly lower than that of other cells, or the temperature rises abnormally, it can be marked as abnormal. The information of all single cells identified as abnormal is organized into the first abnormal discharge data set, which includes the number of the abnormal battery and the description of the abnormal state, such as "voltage is too low" and "temperature is too high".

[0067] Read historical battery discharge data, analyze the discharge trend of these data, and find out the change trend of battery discharge data in the past period of time through time series analysis, such as the change trend of discharge voltage over time, the change trend of discharge current, etc. Then, based on the trend deviation comparison strategy and historical discharge trend, identify the abnormality of current battery discharge data. If the current data deviates significantly from the historical trend, it is considered to be abnormal. For example, if the discharge current has been relatively stable in the historical trend, but the discharge current of a single cell suddenly increases or decreases significantly, it is marked as abnormal, and the abnormal single cell information is organized into the second abnormal discharge data set.

[0068] Obtain the battery property characteristics of the energy storage battery pack, which include several important aspects. The battery type determines the basic chemical properties and working principle of the battery. Different types of batteries, such as lithium-ion batteries and lead-acid batteries, have differences in performance, life, charge and discharge characteristics, etc. Understanding the battery type helps to better understand the behavior and problems of the battery. The battery structure includes the shape of the battery, the layout of the internal electrodes and diaphragms, etc. For example, cylindrical batteries, square batteries and soft-pack batteries have their own structural characteristics, which affect the heat dissipation performance, mechanical strength and installation method of the battery in the energy storage battery pack. Electrical parameters such as the rated voltage, capacity, internal resistance, etc. of the battery play a key role in the discharge performance and health status of the battery. The rated voltage determines the voltage range that the battery can output, the capacity indicates the amount of charge that the battery can store, and the internal resistance affects the charge and discharge efficiency and heat generation of the battery. The battery distribution topology involves the connection method and layout of the battery in the energy storage battery pack, which includes the series and parallel combinations between batteries, and the position relationship of the battery in the entire system. A reasonable battery distribution topology can improve the reliability and performance of the energy storage system, and also affect the methods of abnormal detection and fault diagnosis. The rectifier distribution topology involves the location and connection method of the rectifier in the energy storage system. The function of the rectifier is to convert AC power into DC power to charge the battery or power the load. A reasonable rectifier distribution topology can improve the efficiency and reliability of the system. According to these battery attribute characteristics, a battery twin model is simulated and built. This model is a digital simulation of the actual energy storage battery group, which can reflect the various characteristics and behaviors of the battery in a virtual environment. By rendering the predicted load to the battery twin model, the model is closer to the actual operation. The predicted load is the load demand within a predetermined period calculated according to the previous steps. Applying it to the battery twin model can simulate the response of the battery under different load conditions. Then, based on the battery twin model and the twin deviation comparison strategy, the actual battery discharge data is anomaly identified. The twin deviation comparison strategy compares the actual battery discharge data with the predicted results of the battery twin model to determine whether there is an anomaly. If the actual data differs greatly from the model prediction results, it is considered that the corresponding single battery has an anomaly. For example, if the twin model predicts that the voltage of a single cell under a specific load should be within a certain range, but the actual measured voltage is significantly beyond this range, or the battery temperature is significantly different from the temperature predicted by the model, the single cell is marked as abnormal. Finally, the information of all single cells identified as abnormal is organized into a third abnormal discharge data set, which includes the number of the abnormal battery and the description of the abnormal state, providing a basis for subsequent analysis and processing.

[0069] The first abnormal discharge data set, the second abnormal discharge data set and the third abnormal discharge data set are subjected to a union operation. These three data sets are abnormal data sets obtained by the single-cell deviation comparison strategy, the trend deviation comparison strategy and the twin deviation comparison strategy, respectively. Through the union operation, all the single-cell battery information identified as abnormal is integrated together to ensure that no battery with possible problems is missed. After obtaining the abnormal discharge data set, it is arranged based on the abnormal battery number and the battery distribution topology. The abnormal battery number accurately locates the specific problem battery, and the battery distribution topology reflects the connection method and position relationship of the battery in the energy storage battery pack. By combining these two factors for arrangement, the distribution of abnormal batteries in the entire energy storage battery pack is clearly displayed. For example, according to the battery distribution topology, it is determined which abnormal batteries are in the same series branch and which are in different parallel groups. This arrangement helps to quickly understand the severity and scope of the abnormal situation. If multiple abnormal batteries are concentrated in a specific area, it means that there are systemic problems in the area, such as poor heat dissipation or excessive local load. Ultimately, this arrangement generates an abnormal battery discharge distribution, which provides an intuitive reference for subsequent parameter optimization analysis and discharge rectification adjustment. According to the abnormal distribution, specific abnormal batteries or areas are further inspected and processed in a targeted manner to ensure the safe and stable operation of the energy storage battery pack.

[0070] In a possible implementation, step S410 further includes:

[0071] Step S411: randomly selecting first single-cell battery discharge data from the battery discharge data, wherein the first single-cell battery discharge data includes a first discharge voltage, a first discharge current, a first battery temperature, a first discharge rate, and a first capacity ratio.

[0072] Step S412: performing deviation calculation on the first discharge voltage and other discharge voltages in the battery discharge data, and counting the number of voltage deviations greater than a predetermined voltage threshold, which is recorded as a first voltage difference.

[0073] Step S413: based on the first discharge current, the first battery temperature, the first discharge rate and the first capacity ratio, a first current difference, a first temperature difference, a first rate difference and a first capacity ratio difference are calculated in sequence.

[0074] Step S414: performing abnormal association judgment on the first voltage difference, the first current difference, the first temperature difference, the first rate difference and the first capacity ratio difference according to a predetermined abnormal association index; if the judgment result is abnormal, adding the first single cell discharge data to the first abnormal discharge data set.

[0075] Specifically, a random selection operation is performed from the entire battery discharge data set, and the battery discharge data covers the discharge information of multiple single cells in the energy storage battery group. The goal of the selection is to determine a specific single cell, which is taken as the object of analysis and recorded as the first single cell. The discharge data of this first single cell contains multiple key parameters. Among them, the first discharge voltage reflects the output voltage value of the single cell during the discharge process. Different single cells have different discharge voltages due to their own performance, usage time, and the environment in which they are located. The first discharge current indicates the current output by the single cell during discharge. The discharge current directly affects the discharge power and discharge speed of the battery. The first battery temperature reflects the heat generation of the single cell during the discharge process. Excessive temperature will have an adverse effect on the performance and life of the battery, and even cause safety problems. The first discharge rate represents the speed of the discharge of the single cell, which is usually measured by the amount of electricity discharged per unit time. The first capacity ratio refers to the actual battery capacity ratio of the single cell, that is, the ratio of the current actual capacity to the nominal capacity. This parameter can reflect the remaining capacity and usage status of the single cell. By selecting such a first single cell and the multiple specific parameters it contains, basic data is provided for subsequent deviation calculation and abnormality judgment.

[0076] First, the operation is performed based on the first discharge voltage of the first single cell selected, and the deviation of the first discharge voltage and the discharge voltage of all other single cells in the battery discharge data is calculated respectively, and the relative deviation is calculated by calculating the absolute value of the difference between the voltage values ​​of the two. The purpose of this deviation calculation is to determine the degree of difference between the discharge voltage of the first single cell and the discharge voltage of other single cells. A predetermined voltage threshold is set. This threshold is a standard value predetermined according to the actual situation and is used to determine whether the voltage difference is significant. Then, the number of voltage deviations greater than the predetermined voltage threshold in all comparisons is counted. This number reflects the degree to which the discharge voltage of the first single cell exceeds the normal range compared with the discharge voltage of most other single cells. Finally, the number obtained by this statistics is recorded as the first voltage difference. The first voltage difference can be used as a measurement indicator to determine whether there is an abnormality in the voltage of the first single cell in the subsequent determination. If the first voltage difference is high, it means that the discharge voltage of the first single cell is greatly different from that of other single cells, and there is a problem, which needs further analysis and processing.

[0077] The calculations are performed sequentially based on multiple key parameters of the first single cell. First, the first discharge current is considered, and the deviations of the first discharge current of the first single cell and the discharge currents of other single cells in the battery discharge data are calculated respectively. The absolute value or relative deviation of the current difference is calculated, and a predetermined current threshold is set, and the number of current deviations greater than the threshold is counted, so as to obtain a first current difference, which reflects the degree of difference between the first single cell and other single cells in terms of discharge current. Then the first battery temperature is processed. Similarly, the first battery temperature of the first single cell is compared with the temperature of other single cells and the deviation is calculated, and a predetermined temperature threshold is set, and the number of temperature deviations greater than the threshold is counted to obtain a first temperature difference, which indicates the deviation of the first single cell from other single cells in terms of temperature. Then the first discharge rate is calculated, and the deviations of the first discharge rate of the first single cell and the discharge rates of other single cells are calculated. According to the predetermined rate threshold, the number of deviations greater than the threshold is counted to obtain a first rate difference, which is used to measure the difference between the first single cell and other batteries in terms of discharge rate. Finally, the first capacity ratio is compared with the capacity ratios of other single cells and the deviation is calculated. The deviation is statistically analyzed by the predetermined capacity ratio threshold to obtain the first capacity ratio difference, which reflects the difference between the first single cell and other cells in terms of capacity ratio. By calculating these differences in sequence, the differences between the first single cell and other single cells in different parameters can be fully understood, providing a richer basis for subsequent abnormality judgment.

[0078] The multiple differences of the first single battery are judged for abnormal association according to the predetermined abnormal association index. The predetermined abnormal association index is a set of rules for determining whether the battery is abnormal, and its purpose is to avoid misjudgment of a single index. For example, when the temperature, current and voltage are abnormal at the same time, it is judged as abnormal; or when the rate and temperature are abnormal at the same time, it is also judged as abnormal. This multi-index association judgment method is more accurate and reliable, because the abnormality of a single indicator may be caused by accidental factors or short-term fluctuations, while the simultaneous abnormality of multiple indicators is more likely to indicate that there is indeed a problem with the battery. First, the first voltage difference, the first current difference, the first temperature difference, the first rate difference and the first capacity ratio difference are compared with the predetermined abnormal association index respectively. If the combination of these differences meets the predetermined abnormal conditions, then the judgment result is abnormal. For example, if the first temperature difference and the first current difference exceed the corresponding threshold at the same time and meet the index of "abnormal rate and temperature are abnormal", then it can be judged that the first single battery is abnormal. Once the judgment result is abnormal, the discharge data of the first single battery is added to the first abnormal discharge data set, which is a collection of all single battery data judged to be abnormal. By adding the abnormal first single cell discharge data to the data set, it is convenient to further analyze and process the abnormal battery, such as determining the location of the abnormal battery, analyzing the cause of the abnormality, and taking corresponding repair measures.

[0079] In a possible implementation, step S500 further includes:

[0080] Step S510: using the battery attribute characteristics as similarity constraints, obtaining a sample correction data set based on power big data retrieval, wherein the sample correction data includes a sample battery discharge abnormality distribution and a sample optimized rectification parameter.

[0081] Step S520: configure N correction analysis operators, divide the sample correction data set evenly into N parts, train and cross-validate the N correction analysis operators respectively, obtain N first rectification correction branches and N correction credibility, and construct the first rectification corrector based on the N first rectification correction branches and N correction credibility integration.

[0082] Specifically, the battery attribute characteristics are used as similarity constraints and the power big data is used for retrieval. The battery attribute characteristics include battery type, battery structure, electrical parameters and battery distribution topology, which determine the performance and behavior of the battery. By using these characteristics as similarity constraints, sample data with similar characteristics to the current energy storage battery group are screened out from the huge power big data. The retrieved sample correction data set contains sample battery discharge abnormality distribution and sample optimized rectification parameters. The sample battery discharge abnormality distribution can provide a reference for the analysis of the current battery discharge abnormality, while the sample optimized rectification parameters can be used as an example to optimize the discharge rectification of the current energy storage battery group.

[0083] N correction analysis operators are configured. These operators are various data analysis algorithms or models, such as neural networks, decision trees, etc. Then, the sample correction data set is evenly divided into N parts, and the N correction analysis operators are trained separately, that is, each operator is learned and adjusted using the divided sample data, so that it can analyze the abnormal distribution of battery discharge and output appropriate rectification parameters. At the same time, cross-validation is performed to evaluate the performance and reliability of each operator. Cross-validation divides the sample data into a training set and a validation set, and repeats the training and validation process many times to obtain a more accurate performance evaluation. After training and cross-validation, N first rectification correction branches and N correction credibility are obtained. Each first rectification correction branch corresponds to a correction analysis operator, which can analyze the abnormal distribution of battery discharge and output rectification parameters. The correction credibility indicates the reliability and accuracy of each branch. Based on these N first rectification correction branches and N correction credibility, the first rectification corrector is constructed by an integrated learning method. Ensemble learning can improve the accuracy and stability of prediction by combining multiple weak learners (correction analysis operators here) to construct a strong learner (first rectification corrector). Finally, the first rectification corrector performs parameter optimization analysis on the current battery discharge anomaly distribution. The first rectification corrector comprehensively considers the results of multiple correction analysis branches, and performs weighted averaging or other integration strategies based on the correction credibility to output the first rectification parameters. These first rectification parameters can be used to perform discharge rectification adjustment on the energy storage battery pack to correct the battery discharge anomaly and improve the performance and stability of the energy storage battery pack.

[0084] In a possible implementation, step S500 further includes:

[0085] Step S530: When the energy storage battery pack is charged, the battery charging data is collected by the second monitoring unit.

[0086] Step S540: Based on the single-cell deviation comparison strategy, the trend deviation comparison strategy and the twin deviation comparison strategy, anomaly identification is performed on the battery charging data to generate a battery charging anomaly distribution.

[0087] Step S550: performing parameter optimization analysis on the abnormal charging distribution of the battery through a second rectifier corrector, and outputting a second rectifier parameter to perform charging rectification regulation on the energy storage battery group.

[0088] Specifically, when the energy storage battery pack is in a charging state, the battery charging data is collected through the second monitoring unit. This second monitoring unit is specifically used to monitor various parameters of the battery during the charging process. The collected data includes information such as charging voltage, charging current, and battery temperature. These data can reflect the state and performance of the battery during the charging process.

[0089] Based on the single-cell deviation comparison strategy, trend deviation comparison strategy and twin deviation comparison strategy, the battery charging data is identified for anomalies, thereby generating the battery charging anomaly distribution. Similar to the anomaly identification of battery discharge data, the single-cell deviation comparison strategy compares the charging parameters of each single-cell battery with other single-cell batteries to find out the single-cell batteries with large differences; the trend deviation comparison strategy determines whether the current charging data deviates from the normal trend by analyzing the trend of historical charging data; the twin deviation comparison strategy uses the battery twin model to compare the actual charging data with the model prediction results to identify anomalies. Through the comprehensive application of these strategies, single-cell batteries with abnormalities in the charging process can be accurately found, and the abnormal state and distribution can be determined.

[0090] The second rectifier corrector performs parameter optimization analysis on the abnormal distribution of battery charging, and outputs the second rectifier parameters to adjust the charging and rectification of the energy storage battery pack. This process is similar to the parameter optimization analysis during discharge. The second rectifier corrector performs complex calculations and analyses based on the abnormal distribution of charging, combined with the battery's property characteristics and charging requirements, to determine the optimal rectification parameters. These parameters can adjust the charging current, voltage, etc., making the charging process more stable and efficient, avoiding damage to the battery due to abnormal conditions, and extending the battery's service life.

[0091] In summary, by monitoring and identifying abnormalities during the charging process of the energy storage battery pack, and using the second rectifier to optimize parameters and adjust the rectification, the safety and performance of the battery during the charging process can be effectively guaranteed.

[0092] Embodiment 2 is based on the same inventive concept as the battery pack charge and discharge regulation method based on capacity verification in the above embodiment. Figure 2 As shown, the present application provides a battery pack charge and discharge regulation device based on capacity verification, and the device and method embodiments in the present application embodiments are based on the same inventive concept. Among them, the device includes:

[0093] The predicted load acquisition module 10 is used to obtain a load data set at a fixed point using the first monitoring unit when the energy storage battery group is discharging, perform load prediction based on the load data set, and obtain a predicted load within a predetermined period.

[0094] The discharge control module 20 configures a predetermined discharge strategy based on the predicted load, performs discharge control on the energy storage battery group, and uses the second monitoring unit to collect battery discharge data in real time during the discharge cycle.

[0095] The three-dimensional comparison scheme acquisition module 30 is used to obtain a three-dimensional comparison scheme, wherein the three-dimensional comparison scheme includes a monomer deviation comparison strategy, a trend deviation comparison strategy and a twin deviation comparison strategy.

[0096] The battery discharge abnormality distribution construction module 40 performs abnormality identification on the battery discharge data based on the single-cell deviation comparison strategy, the trend deviation comparison strategy and the twin deviation comparison strategy to construct the battery discharge abnormality distribution.

[0097] The discharge rectification and adjustment module 50 is used to perform parameter optimization analysis on the abnormal discharge distribution of the battery through a first rectification corrector, and output a first rectification parameter to perform discharge rectification adjustment on the energy storage battery group.

[0098] Furthermore, the discharge rectification and regulation module 50 further includes:

[0099] A battery charging data acquisition unit, wherein the battery charging data acquisition unit is used to collect battery charging data through the second monitoring unit when the energy storage battery pack is charging.

[0100] A battery charging abnormality distribution generating unit is configured to perform abnormality identification on the battery charging data based on the single cell deviation comparison strategy, the trend deviation comparison strategy and the twin deviation comparison strategy, and generate a battery charging abnormality distribution.

[0101] A charging rectification regulating unit is used to perform parameter optimization analysis on the abnormal charging distribution of the battery through a second rectification corrector, and output a second rectification parameter to perform charging rectification regulation on the energy storage battery group.

[0102] Furthermore, the predicted load acquisition module 10 further includes:

[0103] The load calculation unit is used to read the load monitoring node, call multiple historical load logs of multiple load ends based on the load monitoring node and the predetermined period matching, and calculate multiple historical starting loads and multiple historical predicted loads.

[0104] A load trend acquisition unit is configured to perform load trend analysis within a predetermined window based on the multiple historical load logs to acquire multiple load trends.

[0105] A load deviation acquisition unit is used to perform deviation mapping calculation on the load data set and the multiple historical starting point loads to obtain multiple load deviations.

[0106] A predicted load summing unit is used to map and compensate the multiple historical predicted loads according to the multiple load trends and the multiple load deviations to obtain multiple standard predicted loads, and to sum up the predicted loads.

[0107] Furthermore, the discharge control module 20 also includes:

[0108] A remaining capacity reading unit is used to read the real-time remaining capacity of the energy storage battery pack.

[0109] A discharge constraint configuration unit, wherein the discharge constraint configuration unit is used to configure over-discharge constraints of the energy storage battery pack.

[0110] A predetermined discharge strategy output unit is used to perform linear optimization of discharge parameters based on the real-time remaining capacity to output the predetermined discharge strategy with the goal of satisfying the over-discharge constraint and the predicted load and minimizing energy consumption.

[0111] Furthermore, the discharge control module 20 also includes:

[0112] The battery discharge data includes discharge data of a plurality of single cells, and the discharge data of the single cells includes discharge voltage, discharge current, battery temperature, discharge rate and capacity ratio.

[0113] Furthermore, the battery discharge abnormal distribution construction module 40 also includes:

[0114] A first abnormal discharge data set acquisition unit is used to identify abnormalities in the battery discharge data based on the cell deviation comparison strategy to obtain a first abnormal discharge data set, wherein the abnormal discharge data includes an abnormal battery number and an abnormal state.

[0115] The second abnormal discharge data set acquisition unit is used to read historical battery discharge data for discharge trend analysis, identify abnormalities in the battery discharge data based on a trend deviation comparison strategy and historical discharge trends, and obtain a second abnormal discharge data set.

[0116] A third abnormal discharge data set acquisition unit is used to acquire battery attribute characteristics of the energy storage battery group, simulate and build a battery twin model according to the battery attribute characteristics, render the predicted load to the battery twin model, and identify the battery discharge data based on the battery twin model and the twin deviation comparison strategy to obtain a third abnormal discharge data set, wherein the battery attribute characteristics include battery type, battery structure, electrical parameters, battery distribution topology, and rectifier distribution topology.

[0117] An abnormal discharge data set acquisition unit is used to perform a union operation on the first abnormal discharge data set, the second abnormal discharge data set and the third abnormal discharge data set to obtain an abnormal discharge data set, and arrange the abnormal discharge data set based on the abnormal battery number and battery distribution topology to generate the battery discharge abnormality distribution.

[0118] Furthermore, the first abnormal discharge data set acquisition unit further includes:

[0119] The first single-cell battery discharge data selection unit is used to select random first single-cell battery discharge data from the battery discharge data, wherein the first single-cell battery discharge data includes a first discharge voltage, a first discharge current, a first battery temperature, a first discharge rate and a first capacity ratio.

[0120] A first voltage difference acquisition unit is used to calculate the deviation between the first discharge voltage and other discharge voltages in the battery discharge data, and count the number of voltage deviations greater than a predetermined voltage threshold, which is recorded as a first voltage difference.

[0121] A difference calculation unit is configured to calculate a first current difference, a first temperature difference, a first rate difference, and a first capacity ratio difference in sequence based on the first discharge current, the first battery temperature, the first discharge rate, and the first capacity ratio.

[0122] An abnormal association judgment unit is used to perform abnormal association judgment on the first voltage difference, the first current difference, the first temperature difference, the first rate difference and the first capacity ratio difference according to a predetermined abnormal association index, and if the judgment result is abnormal, the first single battery discharge data is added to the first abnormal discharge data set.

[0123] Furthermore, the discharge rectification and regulation module 50 further includes:

[0124] A sample correction data set acquisition unit is used to acquire a sample correction data set based on power big data retrieval with the battery attribute characteristics as similarity constraints, wherein the sample correction data includes a sample battery discharge abnormality distribution and a sample optimized rectification parameter.

[0125] The first rectifier corrector construction unit is used to configure N correction analysis operators, evenly divide the sample correction data set into N parts, train and cross-validate the N correction analysis operators respectively, obtain N first rectifier correction branches and N correction credibility, and construct the first rectifier corrector based on the N first rectifier correction branches and N correction credibility integration.

[0126] It should be noted that the above-mentioned sequence of the embodiments of the present application is only for description and does not represent the advantages and disadvantages of the embodiments. And the above-mentioned specific embodiments of this specification are described. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0127] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the protection scope of the present application.

[0128] This specification and the drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A battery pack charge and discharge regulation method based on capacity verification, characterized in that: include: When the energy storage battery pack is discharging, a load data set is obtained at a fixed point using the first monitoring unit, and a load prediction is performed according to the load data set to obtain a predicted load within a predetermined period; Based on the predicted load configuration, a predetermined discharge strategy is configured to control the discharge of the energy storage battery group, and the battery discharge data is collected in real time by using the second monitoring unit during the discharge cycle; Obtaining a three-dimensional comparison scheme, wherein the three-dimensional comparison scheme includes a monomer deviation comparison strategy, a trend deviation comparison strategy, and a twin deviation comparison strategy; Based on the single-cell deviation comparison strategy, the trend deviation comparison strategy, and the twin deviation comparison strategy, anomaly identification is performed on the battery discharge data to construct a battery discharge anomaly distribution; Performing parameter optimization analysis on the abnormal distribution of battery discharge by a first rectifier corrector, and outputting a first rectifier parameter to perform discharge rectification regulation on the energy storage battery group; The battery discharge data includes discharge data of a plurality of single cells, and the discharge data of the single cells includes discharge voltage, discharge current, battery temperature, discharge rate and capacity ratio; Based on the single cell deviation comparison strategy, the trend deviation comparison strategy and the twin deviation comparison strategy, the battery discharge data is anomaly identified to construct a battery discharge anomaly distribution, including: Based on the cell deviation comparison strategy, the battery discharge data is subjected to abnormal identification to obtain a first abnormal discharge data set, wherein the abnormal discharge data includes an abnormal battery number and an abnormal state; Read historical battery discharge data to perform discharge trend analysis, and identify abnormalities in the battery discharge data based on a trend deviation comparison strategy and historical discharge trends to obtain a second abnormal discharge data set; Acquire battery attribute characteristics of an energy storage battery pack, simulate and build a battery twin model according to the battery attribute characteristics, render the predicted load to the battery twin model, and identify abnormalities of the battery discharge data based on the battery twin model and the twin deviation comparison strategy to obtain a third abnormal discharge data set, wherein the battery attribute characteristics include battery type, battery structure, electrical parameters, battery distribution topology, and rectifier distribution topology; Performing a union operation on the first abnormal discharge data set, the second abnormal discharge data set and the third abnormal discharge data set to obtain an abnormal discharge data set, and arranging the abnormal discharge data set based on the abnormal battery numbers and the battery distribution topology to generate the battery discharge abnormality distribution; A first abnormal discharge data set is obtained, including: Selecting random first single-cell battery discharge data from the battery discharge data, wherein the first single-cell battery discharge data includes a first discharge voltage, a first discharge current, a first battery temperature, a first discharge rate, and a first capacity ratio; Calculating deviations between the first discharge voltage and other discharge voltages in the battery discharge data, and counting the number of voltage deviations greater than a predetermined voltage threshold, which is recorded as a first voltage difference; Based on the first discharge current, the first battery temperature, the first discharge rate and the first capacity ratio, sequentially calculating a first current difference, a first temperature difference, a first rate difference and a first capacity ratio difference; Performing abnormal correlation judgment on the first voltage difference, the first current difference, the first temperature difference, the first rate difference and the first capacity ratio difference according to a predetermined abnormal correlation index, and if the judgment result is abnormal, adding the first single battery discharge data to the first abnormal discharge data set; Construct the first rectifier, including: Taking the battery attribute characteristics as similarity constraints, a sample correction data set is obtained based on power big data retrieval, wherein the sample correction data includes sample battery discharge abnormality distribution and sample optimized rectification parameters; Configure N correction analysis operators, divide the sample correction data set evenly into N parts, train and cross-validate the N correction analysis operators respectively, obtain N first rectification correction branches and N correction credibility, and construct the first rectification corrector based on the N first rectification correction branches and N correction credibility integration.

2. The battery pack charge and discharge regulation method based on capacity verification according to claim 1, characterized in that: Also includes: When the energy storage battery pack is charged, the battery charging data is collected by the second monitoring unit; Based on the single deviation comparison strategy, the trend deviation comparison strategy and the twin deviation comparison strategy, anomaly identification is performed on the battery charging data to generate a battery charging anomaly distribution; The second rectifier corrector performs parameter optimization analysis on the abnormal distribution of battery charging, and outputs second rectification parameters to perform charging rectification regulation on the energy storage battery group.

3. The battery pack charge and discharge regulation method based on capacity verification according to claim 1, characterized in that: Get forecasted load, including: Read the load monitoring node, call multiple historical load logs of multiple load ends based on the load monitoring node and the predetermined period matching, and calculate multiple historical starting loads and multiple historical predicted loads; Based on the multiple historical load logs, perform load trend analysis within a predetermined window to obtain multiple load trends; Performing deviation mapping calculation on the load data set and the multiple historical starting point loads to obtain multiple load deviations; The multiple historical predicted loads are mapped and compensated according to the multiple load trends and the multiple load deviations to obtain multiple standard predicted loads, which are summed up to obtain the predicted load.

4. The battery pack charge and discharge regulation method based on capacity verification according to claim 1, characterized in that: Configuring a predetermined discharge strategy based on the predicted load includes: Reading the real-time remaining capacity of the energy storage battery pack; Configuring over-discharge constraints for the energy storage battery pack; With the over-discharge constraint and the predicted load as constraints and with the purpose of minimizing energy consumption, a linear optimization of discharge parameters is performed based on the real-time remaining capacity to output the predetermined discharge strategy.

5. A battery pack charge and discharge regulation device based on capacity verification, characterized in that: The device comprises: A predicted load acquisition module, wherein the predicted load acquisition module is used to obtain a load data set at a fixed point using the first monitoring unit when the energy storage battery pack is discharged, perform load prediction based on the load data set, and obtain a predicted load within a predetermined period; A discharge control module, wherein the discharge control module configures a predetermined discharge strategy based on the predicted load, performs discharge control on the energy storage battery group, and uses a second monitoring unit to collect battery discharge data in real time during a discharge cycle; A three-dimensional comparison scheme acquisition module, wherein the three-dimensional comparison scheme acquisition module is used to obtain a three-dimensional comparison scheme, wherein the three-dimensional comparison scheme includes a monomer deviation comparison strategy, a trend deviation comparison strategy, and a twin deviation comparison strategy; A battery discharge abnormality distribution construction module, wherein the battery discharge abnormality distribution construction module performs abnormality identification on the battery discharge data based on the single deviation comparison strategy, the trend deviation comparison strategy and the twin deviation comparison strategy, and constructs a battery discharge abnormality distribution; A discharge rectification and adjustment module, the discharge rectification and adjustment module is used to perform parameter optimization analysis on the abnormal distribution of battery discharge through a first rectification corrector, and output a first rectification parameter to perform discharge rectification and adjustment on the energy storage battery group; The battery discharge data includes discharge data of a plurality of single cells, and the discharge data of the single cells includes discharge voltage, discharge current, battery temperature, discharge rate and capacity ratio; The battery discharge abnormal distribution building module also includes: a first abnormal discharge data set acquisition unit, which performs abnormal identification on the battery discharge data based on the single cell deviation comparison strategy to obtain a first abnormal discharge data set, wherein the abnormal discharge data includes an abnormal battery number and an abnormal state; a second abnormal discharge data set acquisition unit, the second abnormal discharge data set acquisition unit being used to read historical battery discharge data for discharge trend analysis, and to identify abnormalities in the battery discharge data based on a trend deviation comparison strategy and historical discharge trends, to obtain a second abnormal discharge data set; a third abnormal discharge data set acquisition unit, the third abnormal discharge data set acquisition unit is used to acquire battery attribute characteristics of the energy storage battery group, simulate and build a battery twin model according to the battery attribute characteristics, render the predicted load to the battery twin model, and perform abnormal identification on the battery discharge data based on the battery twin model and the twin deviation comparison strategy to obtain a third abnormal discharge data set, wherein the battery attribute characteristics include battery type, battery structure, electrical parameters, battery distribution topology, and rectifier distribution topology; an abnormal discharge data set acquisition unit, the abnormal discharge data set acquisition unit being used to perform a union operation on the first abnormal discharge data set, the second abnormal discharge data set and the third abnormal discharge data set to obtain an abnormal discharge data set, and to arrange the abnormal discharge data set based on the abnormal battery numbers and the battery distribution topology to generate the battery discharge abnormality distribution; The first abnormal discharge data set acquisition unit further includes: A first single-cell battery discharge data selection unit, the first single-cell battery discharge data selection unit is used to select a random first single-cell battery discharge data from the battery discharge data, wherein the first single-cell battery discharge data includes a first discharge voltage, a first discharge current, a first battery temperature, a first discharge rate and a first capacity ratio; a first voltage difference acquisition unit, the first voltage difference acquisition unit being used to calculate a deviation between the first discharge voltage and other discharge voltages in the battery discharge data, and to count the number of voltage deviations greater than a predetermined voltage threshold value, which is recorded as a first voltage difference; a difference calculation unit, wherein the difference calculation unit sequentially calculates a first current difference, a first temperature difference, a first rate difference, and a first capacity ratio difference based on the first discharge current, the first battery temperature, the first discharge rate, and the first capacity ratio; an abnormal correlation judgment unit, the abnormal correlation judgment unit being used to perform abnormal correlation judgment on the first voltage difference, the first current difference, the first temperature difference, the first rate difference and the first capacity ratio difference according to a predetermined abnormal correlation index, and if the judgment result is abnormal, add the first single battery discharge data to the first abnormal discharge data set; The discharge rectification and regulation module also includes: A sample correction data set acquisition unit, the sample correction data set acquisition unit is used to obtain a sample correction data set based on power big data retrieval with the battery attribute characteristics as similarity constraints, wherein the sample correction data includes a sample battery discharge abnormality distribution and a sample optimized rectification parameter; The first rectifier corrector construction unit is used to configure N correction analysis operators, evenly divide the sample correction data set into N parts, train and cross-validate the N correction analysis operators respectively, obtain N first rectifier correction branches and N correction credibility, and construct the first rectifier corrector based on the N first rectifier correction branches and N correction credibility integration.

Images