A method, device, system and medium for correcting an energy storage battery management system

Abstract

Embodiments of the present application disclose a kind of energy storage battery management system correction method, device, system and medium.The method comprises: generating prediction data based on historical battery data through twin model;When detecting model correction event, train general battery model according to prediction data using machine learning algorithm, obtain target battery model;Model update firmware or model update parameter of target battery model is issued to local energy storage battery manager, to indicate local energy storage battery manager based on model update firmware or model update parameter, carry out model firmware upgrade or model parameter update.The technical scheme of the present application optimizes battery model in cloud server, and synchronizes the optimized battery model to the local energy storage battery manager through firmware upgrade or parameter update, to update the battery model in local end with the optimized battery model, so as to improve the pain points existing in current electrochemical energy storage application.

Description

TECHNICAL FIELD

[0001] Embodiments of the present application relate to the field of electrochemical energy storage technology, and in particular to a correction method, device, system and medium for an energy storage battery management system. BACKGROUND

[0002] With the continuous development of energy technology and the vigorous support of the state for new energy technology, the development of the new energy industry is increasingly valued. As a strategic emerging industry, energy storage technology is an important link to enhance the safety, flexibility and comprehensive efficiency of new energy system supply, and is one of the key technologies to support new energy transformation.

[0003] As a branch of energy storage, electrochemical energy storage still has a relatively low overall proportion, but has great development potential and is widely used. However, the wide application of electrochemical energy storage has also brought about the following problems: the operating conditions of battery cells are not the same, the consistency of the electrochemical characteristics of battery cells produced by different manufacturers and production processes is difficult to guarantee, and the technical level of integrators based on battery cells is uneven. As a result, the current application of electrochemical energy storage generally has the following pain points: inconsistent cell balancing, inaccurate SOC (State of Charge) estimation of energy storage system, severe SOH (State of Health) degradation of battery, frequent safety accidents of electrochemical energy storage system, and short full life cycle of electrochemical energy storage application.

[0004] With the gradual popularization of cutting-edge technologies such as cloud big data, machine learning and AI, how to use these new technologies to effectively manage battery models in energy storage systems to improve the pain points in the current application of electrochemical energy storage has become an exploration direction in this field. SUMMARY

[0005] The present application provides a correction method, device, system and medium for an energy storage battery management system, which can improve the pain points in the current application of electrochemical energy storage, thereby improving the consistency of cell balancing, improving the accurate estimation of the state of charge SOC of the energy storage system, predicting the state of health SOH of the energy storage battery, providing safety warnings, fault tracking and fault analysis for the energy storage system, improving the performance of the energy storage battery, and prolonging the full life cycle of the energy storage system.

[0006] In a first aspect, embodiments of the present application provide a correction method for an energy storage battery management system, executed by a server, wherein the server synchronously deploys a twin model of a battery model in a local energy storage battery manager and a plurality of general battery models; the method comprises:

[0007] generating prediction data based on historical battery data by the twin model, wherein the historical battery data is battery data generated during the operation of an energy storage battery cluster reported by the local energy storage battery manager;

[0008] when a model correction event is detected, training the general battery model according to the prediction data by using a machine learning algorithm to obtain a target battery model;

[0009] issuing a model update firmware or a model update parameter of the target battery model to the local energy storage battery manager to instruct the local energy storage battery manager to perform model firmware upgrading or model parameter updating based on the model update firmware or the model update parameter.

[0010] In a second aspect, the embodiments of the present application also provide a correction method of an energy storage battery management system, which is executed by a local energy storage battery manager, a battery model in the local energy storage battery manager is deployed in a server to run a twin model of the battery model in the server, and the method comprises:

[0011] acquiring battery data generated by an energy storage battery cluster during operation, and reporting the battery data to the server at a preset time interval to instruct the server to trigger a model correction event based on the battery data, generating prediction data based on historical battery data by using the twin model, training a general battery model according to the prediction data by using a machine learning algorithm to obtain a target battery model, and issuing a model update firmware or a model update parameter of the target battery model to the local energy storage battery manager;

[0012] receiving the model update firmware or the model update parameter issued by the server, and verifying the model update firmware or the model update parameter;

[0013] based on the model update firmware or the model update parameter that passes the verification, performing model firmware upgrading or model parameter updating on the battery model to obtain a new battery model, and using the new battery model to manage an operation state of the energy storage battery cluster.

[0014] In a third aspect, the embodiments of the present application also provide a correction device of an energy storage battery management system, which is deployed in a server, the server is deployed with a twin model of a battery model in a local energy storage battery manager and a plurality of general battery models, and the correction device comprises:

[0015] a data prediction module, configured to generate prediction data based on historical battery data by using the twin model, wherein the historical battery data is battery data generated by an energy storage battery cluster during operation and reported by the local energy storage battery manager;

[0016] a model training module, configured to, when a model correction event is detected, train the general battery model according to the prediction data by using a machine learning algorithm to obtain a target battery model;

[0017] The model issuing module is configured to issue model update firmware or model update parameters of the target battery model to the local energy storage battery manager, so as to instruct the local energy storage battery manager to perform model firmware upgrading or model parameter updating based on the model update firmware or model update parameters.

[0018] In a fourth aspect, the embodiments of the present application further provide a correction device of an energy storage battery management system, which is configured in a local energy storage battery manager, wherein a battery model in the local energy storage battery manager is deployed synchronously in a server to run a twin model of the battery model in the server; the correction device comprises:

[0019] The data reporting module is configured to acquire battery data generated by the energy storage battery cluster during operation, and report the battery data to the server at a preset time interval, so as to instruct the server to trigger a model correction event based on the battery data, generate prediction data based on historical battery data through the twin model, train a general battery model to obtain a target battery model according to the prediction data by using a machine learning algorithm, and issue model update firmware or model update parameters of the target battery model to the local energy storage battery manager.

[0020] The model verification module is configured to receive the model update firmware or model update parameters issued by the server, and verify the model update firmware or model update parameters.

[0021] The model updating module is configured to perform model firmware upgrading or model parameter updating on the battery model based on the model update firmware or model update parameters that pass the verification, obtain a new battery model, and manage the operating state of the energy storage battery cluster by using the new battery model.

[0022] In a fifth aspect, the embodiments of the present application further provide a correction system of an energy storage battery management system, which comprises at least two servers, a plurality of local energy storage battery managers and a plurality of energy storage battery clusters.

[0023] The at least two servers comprise one main server and a remaining number of standby servers, the main server and the standby servers are synchronously operated, and the standby servers are configured to backup data of the main server, and replace the main server to interact with the local energy storage battery managers when the main server is down.

[0024] The main server is in communication connection with the plurality of local energy storage battery managers, and is configured to perform the correction method of the energy storage battery management system as described in the first aspect.

[0025] The local energy storage battery managers are respectively in communication connection with the plurality of energy storage battery clusters, and are configured to perform the correction method of the energy storage battery management system as described in the second aspect.

[0026] The energy storage battery cluster is used for recording battery data generated in the running process and sending the battery data to the corresponding local energy storage battery manager.

[0027] In a sixth aspect, the application also provides a storage medium containing computer executable instructions for executing the modification method of the energy storage battery management system as claimed in any one of the first aspect and the second aspect when executed by a computer processor.

[0028] The embodiments of the application provide a modification method, device, system and medium of an energy storage battery management system. The battery model is optimized through a server, the optimized battery model is synchronized to a local energy storage battery manager through firmware upgrade or parameter update, the local battery model is updated by using the optimized battery model, the pain points existing in current electrochemical energy storage applications can be improved, and therefore the balancing consistency of the battery cell is improved, the state of charge SOC of the energy storage system is accurately evaluated, the health degree SOH of the energy storage battery is predicted, the fault safety early warning, fault tracking, fault analysis of the energy storage system are improved, the performance of the energy storage battery is improved, and the battery life of the energy storage system in the whole life cycle is prolonged. BRIEF DESCRIPTION OF DRAWINGS

[0029] Other features, objects and advantages of the application will become more apparent from the following detailed description of non-limiting embodiments made with reference to the accompanying drawings:

[0030] Figure 1 A system structure diagram of a modification system of an energy storage battery management system is provided for an embodiment of the application;

[0031] Figure 2 A flowchart of a modification method of an energy storage battery management system is provided for another embodiment of the application;

[0032] Figure 3 A flowchart of a modification method of an energy storage battery management system is provided for another embodiment of the application;

[0033] Figure 4 A flowchart of a modification method of an energy storage battery management system is provided for another embodiment of the application;

[0034] Figure 5 A flowchart of a modification method of an energy storage battery management system is provided for another embodiment of the application;

[0035] Figure 6 A flowchart of a modification method of an energy storage battery management system is provided for another embodiment of the application;

[0036] Figure 7A structure block diagram of a correction device of an energy storage battery management system according to an embodiment of the present application is provided.

[0037] Figure 8 A structure block diagram of a correction device of an energy storage battery management system according to another embodiment of the present application is provided. DETAILED DESCRIPTION

[0038] The present application will be further described below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are intended for explanation only and are not limiting of the present application. In addition, it should be noted that only the parts related to the present application are shown in the accompanying drawings for the convenience of description.

[0039] Embodiment One

[0040] Figure 1 A system structure diagram of a correction system of an energy storage battery management system according to an embodiment of the present application is provided. As shown in the figure, the system includes at least two servers 110, a plurality of local energy storage battery managers 120 and a plurality of energy storage battery clusters 130. Figure 1

[0041] The at least two servers 110 include one main server 110 and a remaining number of standby servers 110. The main server 110 and the standby servers 110 are synchronously operated. The standby servers 110 are used to backup the data of the main server 110 and replace the main server 110 to interact with the local energy storage battery managers 120 when the main server 110 is down.

[0042] The main server 110 is in communication connection with the plurality of local energy storage battery managers 120 and is used to execute the correction method of the energy storage battery management system as described in any embodiment of the present application.

[0043] The local energy storage battery managers 120 are respectively in communication connection with the plurality of energy storage battery clusters 130 and are used to execute the correction method of the energy storage battery management system as described in any embodiment of the present application.

[0044] The energy storage battery clusters 130 are used to record the battery data generated during the operation and send the battery data to the corresponding local energy storage battery managers 120.

[0045] Specifically, the correction system of the energy storage battery management system further includes a communication interaction device. The local energy storage battery managers 120 and the servers 110 build a stable and fast communication link for bidirectional communication interaction through the communication interaction device. The communication interaction device includes switches, routers, intelligent communication gateways, etc.

[0046] As​Figure 1 As shown, the plurality of local energy storage battery managers 120 are respectively in the intermediate layer, each local energy storage battery manager 120 corresponds to a plurality of energy storage battery clusters 130, and the local energy storage battery manager 120 exchanges information with the plurality of energy storage battery clusters 130 through a CAN communication link. The local energy storage battery manager 120 has a wired Ethernet interface and a wireless network interface. The local energy storage battery manager 120 builds a wired network and a wireless network link dual-redundancy communication link through the wired Ethernet interface, the wireless network interface, the switch, the intelligent communication gateway and the wireless communication module, and the local energy storage battery manager 120 and the server 110 exchange information bidirectionally through the dual-redundancy communication link. The wireless network link includes WIFI, 4G and 5G, etc.

[0047] In an embodiment of the present application, the server and the local energy storage battery manager both have a real-time clock, and perform timing and time setting processing during system online and normal daily interaction. The server and the local energy storage battery manager record the current interaction action events in real time according to their own conditions. For example, the interaction action events include: cloud and local end synchronization model events, machine learning prediction model training events, server generated and determined prediction battery model events, cloud issued upgrade local end battery model events, cloud issued upgrade local end model parameter events, local end received cloud upgrade model firmware events, local end received cloud upgrade local end battery model parameter events, local end battery model upgrade success or failure events, local end model parameter adjustment success or failure events, cloud issued model upgrade firmware success / failure events, cloud issued model parameter adjustment success / failure events, local end received model upgrade firmware success / failure events, local end received model parameter adjustment success / failure events, local end start new model running events, etc.

[0048] In an embodiment of the present application, the plurality of local energy storage battery managers 120 and the corresponding plurality of energy storage battery clusters 130 are deployed in the energy storage container, and the server can be a separate server host or a distributed server cluster.

[0049] In an embodiment of the present application, two sets of servers 110 are deployed in the cloud, one of which is a standby server, and the two servers run synchronously. When one of the servers is down, the other server is notified to immediately establish bidirectional information interaction with the local energy storage battery manager 120, so as to prevent communication interaction data loss and ensure data transmission safety and reliability.

[0050] Embodiment two

[0051] Figure 2A flowchart of a modification method of an energy storage battery management system is provided for another embodiment of the present application, which can be applicable to a scenario of managing the operating conditions of an energy storage battery by using a battery model. The method can be performed by a modification device of the energy storage battery management system, which can be implemented by software and / or hardware and is usually configured in a server. The server is deployed with a twin model of the battery model in the local energy storage battery manager and a plurality of general battery models in a synchronous manner. The method specifically includes the following steps:

[0052] In S210, the twin model is used to generate prediction data based on historical battery data. The historical battery data is the battery data generated during the operation of the energy storage battery cluster reported by the local energy storage battery manager.

[0053] The twin model is a virtual entity running on the server, which is constructed in the same way as the battery model in the local battery manager and initialized by the same data. Specifically, the method of synchronously deploying the twin model of the battery model in the local energy storage battery manager in the server in the embodiment of the present application can be as follows: the local energy storage battery manager constructs an initial battery model generation code by using desktop simulation software, simulates cell parameters by using a hardware-in-the-loop (HIL) device, trains the initial battery model by using the cell parameters simulated by the HIL device to obtain an initialized battery model, synchronously arranges the generated initialized battery model and code in the server and the local energy storage battery manager, and configures the corresponding model parameters to the local energy storage battery manager and the server for model and model parameter synchronization. After synchronization, the system can be powered on and operated, so as to ensure that the battery model in the local energy storage battery manager and the twin model in the server are consistent during initialization, and improve the effectiveness of subsequent updates of the battery model.

[0054] The plurality of general battery models are battery models of batteries constructed based on different principles in an ideal state. The battery model can include an internal resistance equivalent model Rint, a Theveini equivalent circuit model, a second-order RC equivalent circuit model, a PNGV equivalent circuit model, a GNL equivalent circuit model, and an improved hybrid circuit model.

[0055] The historical battery data is battery data of a long-term running of a battery cluster reported by a local energy storage battery manager to a server. Specifically, the local energy storage battery manager is internally deployed with a battery management system (BMS), and the BMS runs for a long time according to a built-in battery module and a model default parameter. In the long-term running process, the BMS records battery data of the battery cluster, and reports the battery data to a cloud server through a double-redundancy communication link. The operation server stores the battery data in the long-term running process as historical battery data. Optionally, the BMS reports the battery data to the cloud server according to different collection intervals. The cloud server classifies the battery data in the long-term running process according to different data types, and stores the historical data of each type of battery data by using different storage periods.

[0056] The prediction data is possible working state data of the battery cluster in a future period of time predicted by the server based on the historical battery data by using a prediction algorithm. Optionally, the prediction algorithm can include a linear regression algorithm, a logistic regression algorithm, a support vector machine algorithm, a random forest algorithm, and the like.

[0057] Specifically, after the server obtains the data generated in the running process of the battery cluster uploaded by the local energy storage battery manager, the server separates the battery data according to different data types, stores the historical battery data by using different storage periods, and uses a twin model corresponding to a battery model in the originally deployed local energy storage battery manager to predict the running state of the battery cluster at a future time, so as to obtain the prediction data. In the embodiment of the application, the twin model is used to predict the running data of the battery cluster in a second time period based on historical battery data in a first time period, as the prediction data. The first time period and the second time period are both time set according to actual application scenarios. For example, the first time period can be one week, and the second time period can be one day, that is, the twin model can be used to predict the running data of tomorrow based on the historical battery data of the past one week. It can be understood that the first time period and the second time period can be configured by a user, or be system default values.

[0058] S220, when a model correction event is detected, a machine learning algorithm is used to train the general battery model according to the prediction data, to obtain a target battery model.

[0059] The model correction event is an event triggering the server to perform a model optimization iteration. The condition triggering the model correction event can be that when it is determined that the battery model at the local end needs to be optimized, the model correction event is triggered. Specifically, when the deviation of the battery operating condition at the local end from the battery operating condition predicted at the server end is large, the model correction event is triggered. The battery operating condition deviation threshold can be set so that when the battery operating condition deviation is greater than the battery operating condition deviation threshold, it is determined that the battery operating condition deviation meets the set condition, and the model correction event is triggered.

[0060] Exemplarily, the first operating curve is determined according to the prediction data. The battery data reported by the local energy storage battery manager in real time is obtained, and the second operating curve is determined according to the battery data in the second time period. The battery operating condition deviation is determined according to the second operating curve and the first operating curve. Specifically, the first operating curve corresponding to the prediction data can be determined by curve fitting, and the second operating curve corresponding to the battery data can be determined. The deviation of the first operating curve and the second operating curve at the same time is compared to obtain the battery operating condition deviation. For example, the deviation of the first operating curve and the second operating curve at the same time point (for example, the same hour or the same day, etc.) is compared as the battery operating condition deviation. When the battery operating condition deviation meets the set condition, the model correction event is triggered.

[0061] The machine learning algorithm can be least square regression, robust regression, local weighted least square, SVM, logistic regression, multi-class classification, and multi-feature optimal logistic regression, etc., which are existing algorithms that can be used to train the model.

[0062] Specifically, after the server predicts the operating state of the energy storage battery cluster in the future period of time based on the historical battery data, when the real-time state reported by the battery does not match the predicted operating state, it is determined that the model correction event is detected. The machine learning algorithm is used to train the plurality of general battery models deployed in the server based on the prediction data, to obtain a plurality of sub-models. The target sub-model used to update the local end battery model is selected from the plurality of sub-models according to the weight of the model, and the target battery module is formed by the target sub-model. It should be noted that the weight of the model can be determined by the data difference degree of the prediction data of each sub-model and the prediction data determined by the twin model based on the same historical battery data. The data difference degree can be determined by statistical methods. In the embodiment of the present application, the fitting curve is obtained by fitting the prediction data, and the data difference degree is determined by determining the deviation of the fitting curve. It can be understood that there are many ways to determine the data difference degree, and the embodiment of the present application does not make specific limitation. For example, the data difference degree can be determined by calculating the average, variance or standard deviation of each group of prediction data.

[0063] S230, issuing the model updating firmware or model updating parameter of the target battery model to the local energy storage battery manager, to instruct the local energy storage battery manager to perform model firmware upgrade or model parameter updating based on the model updating firmware or model updating parameter.

[0064] The model updating firmware includes model code, and the server issues the model code corresponding to the target battery model to the local energy storage battery manager through the reserved communication port, so that the local energy storage battery manager upgrades the firmware of the battery model.

[0065] Specifically, the server determines whether to update the battery model of the local energy storage battery manager or adjust the model parameter of the local energy storage battery manager according to the comparison result of the battery working condition deviation and the set threshold. When the battery working condition deviation is greater than the set threshold, it is determined that the model updating firmware of the target battery model needs to be issued to the local energy storage battery manager. The server determines the model updating firmware corresponding to the target battery model, issues the battery model updating command, and issues the model updating firmware corresponding to the target battery model to the local energy storage battery manager. When the battery working condition deviation is less than or equal to the set threshold, it is determined that the model updating parameter needs to be issued to the local energy storage battery manager. The server determines the model updating parameter corresponding to the target battery model, issues the battery model parameter updating command, and issues the model updating parameter corresponding to the target battery model to the local energy storage battery manager. The battery model of the local end is iteratively updated by the model updating estimation or model updating parameter issued by the server

[0066] The embodiment generates prediction data based on historical battery data by the twin model, wherein the historical battery data is the battery data generated during the operation of the energy storage battery cluster reported by the local energy storage battery manager; when a model correction event is detected, the general battery model is trained according to the prediction data by using a machine learning algorithm to obtain a target battery model; the model updating firmware or model updating parameter of the target battery model is issued to the local energy storage battery manager, to instruct the local energy storage battery manager to perform model firmware upgrade or model parameter updating based on the model updating firmware or model updating parameter, that is, the battery model is optimized by the server, and the optimized battery model is synchronized to the local energy storage battery manager by firmware upgrade or parameter updating, so as to update the battery model of the local end by using the optimized battery model, ensure the effectiveness of the battery model in the local energy storage battery management, and improve the balancing consistency of the battery cell, improve the accurate evaluation of the state of charge (SOC) of the energy storage system, predict the state of health (SOH) of the energy storage battery, and improve the performance of the energy storage battery, prolong the life cycle of the battery of the energy storage system.

[0067] Embodiment three

[0068] Figure 3 This is a flowchart illustrating a correction method for an energy storage battery management system according to another embodiment of the present invention. This embodiment further explains the correction method for the energy storage battery management system based on the above embodiments. Specifically, see [link to documentation]. Figure 3 The method may include:

[0069] S310. Based on historical battery data within the first time period, the twin model is used to predict the operating data of the energy storage battery cluster within the second time period, which is then used as the prediction data.

[0070] The first time period can be determined based on the time cycle corresponding to the data reported by the local energy storage battery manager. For example, if three days of battery data are reported, then the first time period is three days. Alternatively, the first time period can be determined by the system based on the historical battery model correction frequency, maximizing computational resource conservation while ensuring the battery model's effectiveness. It can also be determined based on the time input by the staff; however, this embodiment of the invention does not impose excessive limitations on this aspect.

[0071] The second time period indicates the length of time the forecast data will last. For example, forecast data for the next day can be predicted based on historical battery data from the past three days.

[0072] In this embodiment of the invention, preferably, the length of the first time period is longer than the length of the second time period. By acquiring as much data as possible, data within a shorter time period can be predicted, thereby improving the accuracy of the detection of the model correction event triggered by this application.

[0073] S320. Determine a first operating curve based on the predicted data; obtain the battery data reported in real time by the local energy storage battery manager, and determine a second operating curve based on the battery data within the second time period; determine the battery operating condition deviation based on the second operating curve and the first operating curve; trigger a model correction event when the battery operating condition deviation meets the set conditions.

[0074] Specifically, after obtaining the first running curve representing the predicted data and the second running curve representing the actual data, the two can be overlaid on a reference frame to visually determine their differences. When the deviation between the two exceeds a preset threshold, it indicates that the battery model in the local energy storage battery manager is not working effectively and needs correction, triggering a model correction event. When the deviation is less than the preset threshold, it indicates that the difference between the predicted and actual data is within an acceptable error range, and within the allowable error range, a model correction event will not be triggered.

[0075] S330, classifying the battery data based on the type of the battery data, determining a storage time of each type of the battery data, and storing the corresponding battery data as historical battery data according to the storage time, wherein the battery data includes single cell temperature data, single cell voltage data, charging and discharging event data, charging capacity energy data, discharging capacity energy data, OCV-SOC data, internal resistance data, SOP data, cycle life data, and self-discharge rate data.

[0076] The charging capacity energy data can include charging capacity energy data at different temperatures and charging capacity energy data at different rates. The discharging capacity energy data includes discharging capacity energy data at different temperatures and discharging capacity energy data at different rates. The OCV-SOC data includes discharging OCV-SOC data and charging OCV-SOC data. The internal resistance data includes internal resistance data at different temperatures, internal resistance data at different pulse currents, and internal resistance data at different pulse durations. The SOP data includes SOP data at different temperatures and SOP data at different pulse durations.

[0077] Specifically, for battery data, it can be divided into periodic data and non-periodic data. Periodic data refers to data that has no long-term reference significance as the battery operates, such as single cell temperature data, which is unstable and has too many influencing factors. As historical data, it has little reference significance for future prediction data, so it is classified as periodic data. Non-periodic data is related to the entire life cycle of the battery, has high reference significance, stable data, and small influencing factors, such as cycle life data. For different types of battery data, the storage time is pre-configured. After receiving the battery data, the server classifies the battery data according to the type of the battery data, and stores each battery data according to the pre-configured storage time for each type of battery data.

[0078] 340, for each general battery model, a machine learning algorithm is used to train according to the prediction data to obtain a plurality of candidate battery models; for each candidate battery model, the candidate battery model is used to predict the candidate operation data of the energy storage battery cluster in the second time period based on the historical battery data in the first time period, and a third operation curve is determined according to the candidate operation data; according to the deviation of each third operation curve from the first operation curve, the weight of each candidate battery model is determined, and the target battery model is generated according to the candidate battery model that satisfies the preset condition.

[0079] Specifically, since a plurality of different types of general battery models have been pre-deployed in the server, when it is determined that model correction is needed, a machine learning algorithm is used to train the general battery models based on historical battery data to obtain candidate battery models. For each candidate battery model, a prediction algorithm can be used to predict battery operation data in a second time period based on historical battery data in the first time period as candidate operation data. Based on the candidate operation data, a linear fitting method is used to obtain a third operation curve corresponding to the candidate operation data of each candidate battery model. The third curve is compared with a first operation curve corresponding to the prediction data predicted by the twin model based on the historical battery data in the same time period, so as to determine the optimal at least one candidate battery model according to the comparison result, and generate a target battery model according to the optimal candidate battery model.

[0080] In the implementation of the application, preferably, the weight of each candidate battery model is determined, which can be determined by the deviation of the operation curve. For example, the weight of the candidate battery model is positively correlated with the deviation of the third operation curve and the first operation curve. That is, the greater the deviation, the greater the weight of the candidate battery model, and the smaller the deviation, the smaller the weight of the candidate battery model. After determining the weight corresponding to each candidate battery model, the candidate battery models with a weight greater than a threshold value are selected to form the target battery model.

[0081] S350, obtain a battery working condition deviation, determine whether the battery working condition deviation is greater than a set threshold value, if yes, execute S360, otherwise, execute S370.

[0082] S360, issue a model update firmware of the target battery model to the local energy storage battery manager.

[0083] S370, issue a model update parameter of the target battery model to the local energy storage battery manager.

[0084] Specifically, when the battery working condition deviation is greater than the set threshold value, the server determines that the battery model running in the local energy storage battery manager has a large deviation, and the battery model of the local energy storage battery manager needs to be upgraded. When the battery working condition deviation is less than or equal to the set threshold value, the server determines that the battery model running in the local energy storage battery manager has a small deviation, which can be overcome by correcting the parameters. The application embodiment determines whether to perform firmware upgrade on the local battery model by comparing the battery working condition with the set threshold value, which can reasonably utilize the processing resources of the local energy storage battery manager and avoid occupying the processing resources for model firmware upgrade when it is not necessary.

[0085] In one specific embodiment, the model optimization step of the server end is described in detail. Figure 4A flow chart of a revision method of an energy storage battery management system according to another embodiment of the present application is shown in Figure 4 The method comprises the following steps:

[0086] S410, the local end BMS reports real-time operation data to the server, and subsequently performs S420.

[0087] S420, the server stores battery historical data, and subsequently performs S430.

[0088] S430, the server adopts a machine learning algorithm to perform multi-feature model training in combination with the battery historical data and a general battery model constructed by the server, and predicts whether the local end model is accurate based on the model obtained by training, and subsequently performs S440.

[0089] S440, the server judges whether the operation deviation of the local end battery model is within a normal range, if yes, returns to perform 430 again, and if no, subsequently performs S450.

[0090] S450, the server judges whether the local end battery model needs to update the model or adjust the model parameters, if it is judged that the model needs to be updated, S460 is performed, and if it is judged that the model parameters need to be adjusted, S470 is performed.

[0091] S460, the server issues model update firmware to the local end BMS through a communication link with the local end, for IAP self-upgrade of the battery model.

[0092] S470, the server issues model parameters to the local end BMS through a communication link with the local end, for adjustment of the local end battery model parameters.

[0093] It should be noted that after issuing the model update parameters or the model update firmware, the server needs to confirm whether the local energy storage battery manager receives the correct revision data and whether the data can be correctly processed to revise the local battery management system, and to start the new model or the new model parameter operation mode adjustment operation after the local energy storage battery manager model is upgraded or the model parameters are iteratively revised.

[0094] The embodiment of the application further generates a running curve by using historical data and prediction data for comparison, improves the accuracy and efficiency of the model correction time trigger, selects the best candidate battery model from multiple candidate battery models to form a target battery model based on weights, enriches the selection range of the target battery model, and ensures the effectiveness of the target battery model, and determines to issue model update firmware or model update parameters according to the battery working condition deviation, reasonably uses the processing resources of the local energy storage battery manager, and improves the efficiency of model updating.

[0095] Embodiment four

[0096] Figure 5 The flowchart of the correction method of the energy storage battery management system provided by another embodiment of the application can be applied to the scene of managing the running status of the energy storage battery by using the battery model. The method can be performed by a correction device of the energy storage battery management system, the device can be realized by software and / or hardware, and is usually configured in a local energy storage battery manager. The battery model in the local energy storage battery manager is deployed in a server to run a twin model of the battery model in the server. The specific steps include the following steps:

[0097] S510, obtain battery data generated by the energy storage battery cluster during running, and report the battery data to the server at a preset time interval to instruct the server to trigger a model correction event based on the battery data, generate prediction data based on historical battery data by using the twin model, train a general battery model to obtain a target battery model according to the prediction data by using a machine learning algorithm, and issue model update firmware or model update parameters of the target battery model to the local energy storage battery manager.

[0098] Specifically, the local energy storage battery manager can run according to the actual running condition and the original deployed battery default parameters. The local energy storage battery manager can perform information interaction with multiple energy storage battery clusters through a preset communication interaction link during long-term running to obtain battery data generated during the running of the energy storage battery cluster, and report the obtained battery data to the server through the preset communication interaction link. Optionally, the battery data is reported to the server deployed with the twin model and multiple general battery models at different preset upload time intervals. Due to different sampling accuracies, the time interval of the local energy storage battery manager for collecting battery data is different, resulting in different upload time intervals of the battery data reported to the server. For example, the local energy storage manager collects battery data according to a preset accuracy, and reports the battery data to the server at an accuracy interval.

[0099] In the embodiment of the present application, preferably, the preset communication link can be a dual-redundancy communication link. For example, the wired network link of the dual-redundancy communication link is constructed through a wired Ethernet interface, a switch, a router, an intelligent communication gateway and the like. The wireless network link of the dual-redundancy communication link is constructed through a wireless network interface, a wireless module, a switch, a router, an intelligent communication gateway and the like. The server adopts a host and standby machine mode to construct a stable dual-communication loop dual-cloud server redundancy communication link architecture.

[0100] S520, receiving the model update firmware or model update parameter issued by the server, and verifying the model update firmware or model update parameter.

[0101] Specifically, since the server issues the model update firmware or model update parameter to the local energy storage battery manager through the communication link, there may be a case that the model update firmware or model update parameter has error data due to network reasons. For example, packet loss or malicious tampering occurs during transmission. Therefore, in the embodiment of the present application, the model update firmware or model update parameter issued by the server is verified to ensure that the dependent data of the local energy storage battery manager for battery model update is correct.

[0102] In the embodiment of the present application, preferably, if the verification is incorrect, the local energy storage battery manager feeds back to the server to require reissuing of the model update firmware or model update parameter, so as to ensure that the local energy storage battery manager can complete the battery model update process.

[0103] S530, based on the verified model update firmware or model update parameter, performing model firmware upgrade or model parameter update on the battery model to obtain a new battery model, and using the new battery model to manage the operation state of the energy storage battery cluster.

[0104] Specifically, in the embodiment of the present application, after the model itself is updated by the model update firmware or model update parameter issued by the server and verified to be correct, the local energy storage battery manager feeds back information of successful update to the server. After receiving the start running new model command issued by the server, the local energy storage battery manager starts the new battery model and manages the operation state of the energy storage battery cluster according to the updated battery model.

[0105] In one specific embodiment, the model iteration update step of the local end is described in detail. Figure 6 A flowchart of a correction method of an energy storage battery management system according to another embodiment of the present application is shown in FIG. 5, which includes the following steps. Figure 6

[0106] ​S610, the server evaluates whether to update the local battery model or adjust the local battery model parameters according to the battery historical data and real-time battery data. If the local battery model parameters are adjusted, S621 is executed. If the local battery model is updated, S631 is executed.

[0107] S621, the server issues an update local battery model parameter command, and S622 is executed subsequently.

[0108] S622, the local end receives the update battery model parameter command issued by the server, and S623 is executed subsequently.

[0109] S623, the local end corrects the battery model parameters. If the correction is correct, S624 is executed subsequently. If the correction is incorrect, S621 is executed again.

[0110] S624, the local end feeds back the model parameter upgrade success result to the server, requires the server to issue a start running new model command, and S625 is executed subsequently.

[0111] S625, the server issues a start running new model command, and S626 is executed subsequently.

[0112] S626, the local end BMS starts running according to the new model parameters.

[0113] S631, the server issues an update local battery model command, and S632 is executed subsequently.

[0114] S632, the local end receives the update battery model command issued by the server, and the server issues the model firmware required for updating the battery model, and S633 is executed subsequently.

[0115] S633, the local end corrects the battery model upgrade firmware. If the correction is correct, S634 is executed subsequently. If the correction is incorrect, S631 is executed again.

[0116] S634, the local end receives the model firmware and performs self-upgrade IAP, and S635 is executed subsequently.

[0117] S635, the local end judges whether the battery model upgrade firmware upgrade is successful. If the firmware upgrade is successful, S636 is executed subsequently. If the firmware upgrade fails, S631 is executed again.

[0118] S636, the local end feeds back the upgrade firmware success result to the server, requires the server to issue a model parameter configuration and start running new model command, and S637 is executed subsequently.

[0119] S637, the server issues a model parameter configuration and start running new model command, and S638 is executed subsequently.

[0120] S638, the local BMS starts running according to the new model.

[0121] In the embodiment of the application, the local battery energy storage manager obtains battery data generated by the energy storage battery cluster during operation, and reports the battery data to the server at a preset time interval; receives the model update firmware or model update parameter issued by the server, and checks the model update firmware or model update parameter; based on the model update firmware or model update parameter that passes the check, the battery model is upgraded or the model parameter is updated, a new battery model is obtained, and the running state of the energy storage battery cluster is managed by using the new battery model, that is, the local energy storage battery manager can realize high-precision updating and use of the battery model through simple data uploading and data receiving processing, reducing the calculation demand of the local energy storage battery manager, improving the balancing consistency of the battery cell, improving the accurate evaluation of the state of charge SOC of the energy storage system, predicting the health degree SOH of the energy storage battery, and further improving the applicability and compatibility of the correction method of the application to different scenes on the basis of improving the fault safety warning, fault tracking, fault analysis, improving the performance of the energy storage battery, and prolonging the life cycle of the energy storage system.

[0122] Embodiment

[0123] Figure 7 The structure block diagram of the correction device of the energy storage battery management system provided by an embodiment of the application can be realized by software and / or hardware, and is usually deployed in a server, and the server is synchronously deployed with a twin model of a battery model in a local energy storage battery manager and a plurality of general battery models, and can include:

[0124] The data prediction module 710 is configured to generate prediction data based on historical battery data through the twin model, wherein the historical battery data is battery data generated by the energy storage battery cluster during operation and reported by the local energy storage battery manager.

[0125] The model training module 720 is configured to train the general battery model according to the prediction data by using a machine learning algorithm when a model correction event is detected, to obtain a target battery model.

[0126] The model issuing module 730 is configured to issue model update firmware or model update parameters of the target battery model to the local energy storage battery manager, so as to instruct the local energy storage battery manager to perform model firmware upgrade or model parameter update based on the model update firmware or model update parameters.

[0127] The correction device of the energy storage battery management system provided by the embodiments of the present application can execute the correction method of the energy storage battery management system provided by any of the embodiments of the present application, and has the function modules and beneficial effects corresponding to the execution method.

[0128] Optionally, the data prediction module 710 is specifically configured to predict, by the twin model, running data of the energy storage battery cluster in a second time period as prediction data based on historical battery data in a first time period.

[0129] Optionally, the device further comprises:

[0130] An event triggering module is configured to, after the prediction data is generated by the twin model based on the historical battery data, determine a first running curve according to the prediction data; acquire battery data reported by the local energy storage battery manager in real time, and determine a second running curve according to the battery data in the second time period; determine a battery working condition deviation according to the second running curve and the first running curve; and trigger a model correction event when the battery working condition deviation meets a set condition.

[0131] Optionally, the correction device of the energy storage battery management system further comprises a data storage module.

[0132] The data storage module is configured to classify the battery data based on types of the battery data, determine storage times of each type of the battery data, and store the corresponding battery data as historical battery data according to the storage times, wherein the battery data includes single cell temperature data, single cell voltage data, charging and discharging event data, charging capacity energy data, discharging capacity energy data, OCV-SOC data, internal resistance data, SOP data, cycle life data, and self-discharge rate data.

[0133] Optionally, the model training module 720 is specifically configured to, for each general battery model, train a plurality of alternative battery models by using a machine learning algorithm according to the prediction data; for each alternative battery model, predict alternative running data of the energy storage battery cluster in a second time period based on historical battery data in a first time period, and determine a third running curve according to the alternative running data; determine a weight of each alternative battery model according to a deviation of each third running curve from the first running curve; and generate a target battery model according to the alternative battery model whose weight meets a preset condition.

[0134] Optionally, the model issuing module 730 is specifically configured to issue the model update firmware of the target battery model to the local energy storage battery manager when the battery working condition deviation is greater than the set threshold; and issue the model update parameter of the target battery model to the local energy storage battery manager when the battery working condition deviation is less than or equal to the set threshold.

[0135] The modified device for the energy storage battery management system provided by the embodiment of the application after further description can also execute the modified method for the energy storage battery management system provided by any embodiment of the application, and has the function modules and beneficial effects corresponding to the execution method.

[0136] Embodiment six

[0137] Figure 8 A structural block diagram of a modified device for an energy storage battery management system provided by another embodiment of the application, which can be implemented by software and / or hardware, and is usually configured in a local energy storage battery manager, a battery model in the local energy storage battery manager is deployed in a server to run a twin model of the battery model in the server, and the device can include:

[0138] The data reporting module 810 is configured to obtain battery data generated by the energy storage battery cluster during operation, report the battery data to the server at a preset time interval, instruct the server to trigger a model correction event based on the battery data, generate predicted data based on historical battery data through the twin model, train a general battery model to obtain a target battery model according to the predicted data by using a machine learning algorithm, and issue model update firmware or model update parameters of the target battery model to the local energy storage battery manager.

[0139] The model verification module 820 is configured to receive the model update firmware or model update parameters issued by the server, and verify the model update firmware or model update parameters.

[0140] The model update module 830 is configured to perform model firmware upgrade or model parameter update on the battery model based on the verified model update firmware or model update parameters, obtain a new battery model, and manage the operating state of the energy storage battery cluster by using the new battery model.

[0141] The modified device for the energy storage battery management system provided by the embodiment of the application can execute the modified method for the energy storage battery management system provided by any embodiment of the application, and has the function modules and beneficial effects corresponding to the execution method.

[0142] Embodiment seven

[0143] The seventh embodiment of the present application also provides a storage medium containing computer executable instructions, which, when executed by a computer processor, are used to perform a correction method of a storage battery management system.

[0144] The method can be performed by a server, which synchronously deploys a twin model of a battery model in a local storage battery manager and a plurality of general battery models, and specifically includes: generating prediction data based on historical battery data through the twin model, wherein the historical battery data is battery data generated in a running process of a storage battery cluster reported by the local storage battery manager; when a model correction event is detected, training the general battery model according to the prediction data by using a machine learning algorithm to obtain a target battery model; and issuing a model update firmware or a model update parameter of the target battery model to the local storage battery manager to instruct the local storage battery manager to perform model firmware upgrading or model parameter updating based on the model update firmware or the model update parameter.

[0145] Alternatively, the method can be performed by a local storage battery manager, and a battery model in the local storage battery manager is synchronously deployed in a server to run a twin model of the battery model in the server, and specifically includes: obtaining battery data generated in a running process of a storage battery cluster, and reporting the battery data to the server at a preset time interval to instruct the server to trigger a model correction event based on the battery data, generate prediction data based on historical battery data through the twin model, train a general battery model according to the prediction data by using a machine learning algorithm to obtain a target battery model, and issue a model update firmware or a model update parameter of the target battery model to the local storage battery manager; receiving the model update firmware or the model update parameter issued by the server, and verifying the model update firmware or the model update parameter; based on the verified model update firmware or model update parameter, performing model firmware upgrading or model parameter updating on the battery model to obtain a new battery model, and using the new battery model to manage a running state of the storage battery cluster.

[0146] Of course, the storage medium containing computer executable instructions provided by the embodiment of the present application is not limited to the method operations described above, and can also perform related operations in the correction method of the storage battery management system provided by any embodiment of the present application.

[0147] Through the above description of the embodiments, those skilled in the art can clearly understand that the present application can be realized by means of software and necessary universal hardware, and of course can also be realized by hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application can be embodied in the form of a software product in essence or in the form of a part of the prior art that makes a contribution, and the computer software product can be stored in a computer readable storage medium, such as a floppy disk, a read-only memory (ROM), a random access memory (RAM), a FLASH, a hard disk, or an optical disk, and includes a number of instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in various embodiments of the present application.

[0148] It is worth noting that in the above embodiments of the correction device of the energy storage battery management system, each unit and module included is only divided according to functional logic, but is not limited to the above division, as long as the corresponding functions can be realized; in addition, the specific names of each functional unit are only for easy mutual differentiation, and are not used to limit the protection scope of the present application.

[0149] Note that the above are only preferred embodiments of the present application and the technical principles applied. Those skilled in the art will understand that the present application is not limited to the specific embodiments described herein, and those skilled in the art can make various obvious changes, readjustments and substitutions without departing from the scope of the present application. Therefore, although the present application has been described in more detail through the above embodiments, the present application is not limited to the above embodiments, and can include more other equivalent embodiments without departing from the concept of the present application, and the scope of the present application is determined by the scope of the appended claims.

Claims

1. A correction method for an energy storage battery management system, characterized in that, The method is executed by a server, which synchronously deploys a twin model of the battery model in the local energy storage battery manager and multiple general battery models; the method includes: Predictive data is generated based on historical battery data through the twin model, wherein the historical battery data is the battery data generated during the operation of the energy storage battery cluster reported by the local energy storage battery manager; the twin model is a virtual entity run by the server and constructed in the same way as the battery model in the local energy storage battery manager and initialized with the same data; the general battery model is a battery model under ideal conditions constructed based on different principles. The first operating curve is determined based on the predicted data; When a model correction event is detected due to the need for optimization of the battery model in the local energy storage battery manager, a machine learning algorithm is used to train the general battery model based on the predicted data to obtain the target battery model. The model update firmware or model update parameters of the target battery model are sent to the local energy storage battery manager to instruct the local energy storage battery manager to upgrade the model firmware or update the model parameters of the battery model in the local energy storage battery manager based on the model update firmware or model update parameters. The step of training the general battery model based on the predicted data using a machine learning algorithm to obtain the target battery model includes: For each general battery model, a machine learning algorithm is used to train based on the predicted data to obtain multiple candidate battery models; For each candidate battery model, based on historical battery data in the first time period, the candidate operating data of the energy storage battery cluster in the second time period is predicted, and a third operating curve is determined based on the candidate operating data. Based on the deviation between each of the third operating curves and the first operating curve, the weight of each of the candidate battery models is determined, and a target battery model is generated based on the candidate battery models whose weights meet preset conditions. The step of generating predictive data based on historical battery data using the twin model includes: The twin model predicts the operating data of the energy storage battery cluster in a second time period based on historical battery data in the first time period, and uses this as the prediction data.

2. The method according to claim 1, characterized in that, After generating predictive data based on historical battery data using the twin model, the method further includes: Obtain the battery data reported in real time by the local energy storage battery manager, and determine the second operating curve based on the battery data within the second time period; The battery operating condition deviation is determined based on the second operating curve and the first operating curve; When the battery operating condition deviation meets the set conditions, a model correction event is triggered.

3. The method according to claim 2, characterized in that, After obtaining the battery data reported in real time by the local energy storage battery manager, the method further includes: The battery data is classified based on its type, and the storage time for each type of battery data is determined. The corresponding battery data is stored according to the storage time as historical battery data. The battery data includes individual cell temperature data, individual cell voltage data, charge and discharge event data, charging capacity energy data, discharging capacity energy data, OCV-SOC data, internal resistance data, SOP data, cycle life data, and self-discharge rate data.

4. The method according to claim 1, characterized in that, The step of sending the model update firmware or model update parameters of the target battery model to the local energy storage battery manager includes: When the battery operating condition deviation exceeds a set threshold, the model update firmware of the target battery model is sent to the local energy storage battery manager. When the battery operating condition deviation is less than or equal to a set threshold, the model update parameters of the target battery model are sent to the local energy storage battery manager.

5. A correction method for an energy storage battery management system, characterized in that, The method is executed by a local energy storage battery manager, wherein the battery model in the local energy storage battery manager is synchronously deployed on a server to run a twin model of the battery model on the server; the method includes: The system acquires battery data generated by the energy storage battery cluster during operation, reports the battery data to the server at preset time intervals to instruct the server to trigger a model correction event based on the battery data, generates predicted data based on historical battery data through the twin module, determines a first operating curve based on the predicted data, trains a general battery model using a machine learning algorithm based on the predicted data to obtain a target battery model, and sends the model update firmware or model update parameters of the target battery model to the local energy storage battery manager. The twin model is a virtual entity run by the server and constructed in the same way as the battery model in the local energy storage battery manager and initialized with the same data. The general battery model is a battery model under ideal conditions constructed based on different principles. The model correction event is triggered when the battery model in the local energy storage battery manager is detected to need optimization. The step of generating predicted data based on historical battery data through the twin model includes: predicting the operating data of the energy storage battery cluster in a second time period based on historical battery data in a first time period using the twin model, as the predicted data. Receive the model update firmware or model update parameters sent by the server, and verify the model update firmware or model update parameters; Based on the verified model update firmware or model update parameters, the battery model in the local energy storage battery manager is upgraded with model firmware or updated with model parameters to obtain a new battery model. The new battery model is then used to manage the operating status of the energy storage battery cluster. The specific process for generating the target battery model is as follows: For each general battery model, a machine learning algorithm is used to train multiple candidate battery models based on the predicted data. For each candidate battery model, based on historical battery data within the first time period, candidate operating data for the energy storage battery cluster within the second time period is predicted, and a third operating curve is determined based on the candidate operating data. The weights of each candidate battery model are determined based on the deviations between each of the third operating curves and the first operating curve, and a target battery model is generated based on the candidate battery models whose weights meet preset conditions.

6. A correction device for an energy storage battery management system, characterized in that, Deployed on a server, the server synchronously deploys a twin model of the battery model in the local energy storage battery manager and multiple general battery models; the correction device includes: The data prediction module is used to generate prediction data based on historical battery data through the twin model. The historical battery data is the battery data generated during the operation of the energy storage battery cluster reported by the local energy storage battery manager. The twin model is a virtual entity run by the server and constructed in the same way as the battery model in the local energy storage battery manager and initialized with the same data. The general battery model is a battery model under ideal conditions constructed based on different principles. The event triggering module is used to determine the first running curve based on the predicted data; The model training module is used to train the general battery model based on the predicted data using a machine learning algorithm when a model correction event is detected that the battery model in the local energy storage battery manager needs to be optimized, so as to obtain the target battery model. The model delivery module is used to deliver the model update firmware or model update parameters of the target battery model to the local energy storage battery manager, so as to instruct the local energy storage battery manager to upgrade the model firmware or update the model parameters of the battery model in the local energy storage battery manager based on the model update firmware or model update parameters. Specifically, the model training module is used to: train each general battery model using a machine learning algorithm based on the predicted data to obtain multiple candidate battery models; for each candidate battery model, predict the candidate operating data of the energy storage battery cluster in a second time period based on historical battery data in a first time period, and determine a third operating curve based on the candidate operating data; determine the weight of each candidate battery model based on the deviation between each of the third operating curves and the first operating curve, and generate a target battery model based on the candidate battery models whose weights meet preset conditions; Specifically, the data prediction module is used to predict the operating data of the energy storage battery cluster in a second time period based on historical battery data in the first time period using the twin model, as the prediction data.

7. A correction device for an energy storage battery management system, characterized in that, Configured in a local energy storage battery manager, the battery model in the local energy storage battery manager is synchronously deployed on a server to run a twin model of the battery model on the server; The correction device includes: The data reporting module is used to acquire battery data generated by the energy storage battery cluster during operation, and report the battery data to the server at preset time intervals to instruct the server to trigger a model correction event based on the battery data. The module generates predicted data based on historical battery data through the twin module, determines a first operating curve based on the predicted data, trains a general battery model using a machine learning algorithm based on the predicted data to obtain a target battery model, and sends the model update firmware or model update parameters of the target battery model to the local energy storage battery manager. The twin model is a virtual entity run by the server and constructed in the same way as the battery model in the local energy storage battery manager, and initialized with the same data. The general battery model is a battery model under ideal conditions constructed based on different principles. The model correction event is triggered when the battery model in the local energy storage battery manager is detected to need optimization. The step of generating predicted data based on historical battery data through the twin model includes: predicting the operating data of the energy storage battery cluster in a second time period based on historical battery data within a first time period using the twin model, as the predicted data. The model verification model is used to receive the model update firmware or model update parameters sent by the server and verify the model update firmware or model update parameters. The model update module is used to upgrade the model firmware or update the model parameters of the battery model in the local energy storage battery manager based on the verified model update firmware or model update parameters to obtain a new battery model, and use the new battery model to manage the operating status of the energy storage battery cluster. Specifically, the model training module is used to: train each general battery model using a machine learning algorithm based on the predicted data to obtain multiple candidate battery models; for each candidate battery model, predict the candidate operating data of the energy storage battery cluster in a second time period based on historical battery data within the first time period, and determine a third operating curve based on the candidate operating data; determine the weight of each candidate battery model based on the deviation between each of the third operating curves and the first operating curve, and generate a target battery model based on the candidate battery models whose weights meet preset conditions.

8. A correction system for an energy storage battery management system, characterized in that, include: At least two servers, multiple local energy storage battery managers, and multiple energy storage battery clusters; The at least two servers include a primary server and a number of backup servers. The primary server and the backup servers operate synchronously. The backup servers are used to back up the data of the primary server and, in the event of a primary server failure, to replace the primary server in interacting with the local energy storage battery manager. The main server is communicatively connected to the plurality of local energy storage battery managers and is used to execute the correction method of the energy storage battery management system as described in any one of claims 1-4; The local energy storage battery manager is communicatively connected to the plurality of energy storage battery clusters and is used to execute the correction method of the energy storage battery management system as described in claim 5. The energy storage battery cluster is used to record battery data generated during operation and send the battery data to the corresponding local energy storage battery manager.

9. A storage medium containing computer-executable instructions, characterized in that, The computer-executable instructions, when executed by a computer processor, are used to perform the correction method of the energy storage battery management system as described in any one of claims 1-5.

Images