Battery cluster state prediction method and related device

Abstract

The invention provides a state prediction method of a battery cluster and a related device, and the method comprises the steps: determining the battery state data of each single battery in a target battery cluster; then, inputting the battery state data of each single battery into a trained state prediction model, and determining predicted voltage entropy data of each single battery according to the output of the trained state prediction model; then, determining an abnormal single battery according to the predicted voltage entropy data of each single battery; and finally, constructing an equivalent circuit model for the abnormal single battery to monitor the abnormal single battery, and if it is determined that the abnormal single battery is abnormal, generating warning information. According to the invention, early warning can be carried out on the battery cluster composed of a large number of batteries, the abnormal battery monomers in the battery cluster are determined in advance for a long time, the abnormal battery monomers are monitored, and the early warning accuracy of the battery cluster is greatly improved while the power consumption of the system is saved.

Description

Technical Field

[0001] The present application relates to the technical field of energy storage systems, and in particular to a state prediction method and related devices for a battery cluster. Background Art

[0002] Battery Management System (BMS) is a system used in battery applications such as electric vehicles, energy storage systems, and mobile devices. Its main functions are to monitor, protect, control, and optimize the battery charging and discharging process to ensure the safety of the battery pack, extend battery life, improve system efficiency, and provide accurate status information for the system.

[0003] Battery state of health (SOH) is used to evaluate battery performance and life. SOH plays an important role in BMS and is an important indicator of battery aging. SOH estimation research is crucial for battery safety warning, maintenance management, and remaining useful life (RUL) prediction. Most of the current battery warning technologies use data-driven and model-driven methods, but can only warn of battery changes in a short period of time to respond in advance to the effects of thermal runaway, capacity decay, and increased internal resistance. Summary of the invention

[0004] In view of this, the present application provides a battery cluster status prediction method and related devices, which can provide early warning for a battery cluster composed of a large number of batteries, determine abnormal battery cells in the battery cluster in advance, and monitor the abnormal battery cells, thereby greatly improving the early warning accuracy of the battery cluster while saving system power consumption.

[0005] In a first aspect, an embodiment of the present application provides a method for predicting a state of a battery cluster, the method comprising:

[0006] Determine battery status data of each battery cell in the target battery cluster;

[0007] Inputting the battery state data of each battery cell into a trained state prediction model, and determining the predicted voltage entropy data of each battery cell according to the output of the trained state prediction model;

[0008] Determine an abnormal battery cell according to the predicted voltage entropy data of each battery cell;

[0009] An equivalent circuit model is constructed for the abnormal battery cell to monitor the abnormal battery cell, and a warning message is generated if it is determined that the abnormal battery cell is abnormal.

[0010] In a second aspect, an embodiment of the present application provides a battery cluster state prediction device, the device comprising:

[0011] A state determination unit, used to determine battery state data of each battery cell in the target battery cluster;

[0012] A first prediction unit, configured to input the battery state data of each battery cell into a trained state prediction model, and determine predicted voltage entropy data of each battery cell according to an output of the trained state prediction model;

[0013] an abnormality determination unit, configured to determine an abnormal battery cell according to the predicted voltage entropy data of each battery cell;

[0014] The second prediction unit is used to construct an equivalent circuit model for the abnormal battery cell to monitor the abnormal battery cell, and generate warning information if it is determined that the abnormal battery cell is abnormal.

[0015] In a third aspect, an embodiment of the present application provides an electronic device, comprising a processor, a memory, a communication interface, and one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the processor, and the program includes instructions for executing the steps of any method in the first aspect of the embodiment of the present application.

[0016] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute part or all of the steps described in any method of the first aspect of the embodiment of the present application.

[0017] In a fifth aspect, an embodiment of the present application provides a computer program product, wherein the computer program product includes a non-transitory computer-readable storage medium storing a computer program, and the computer program is operable to cause a computer to execute some or all of the steps described in any method of the first aspect of the embodiment of the present application. The computer program product may be a software installation package.

[0018] It can be seen that through the above-mentioned battery cluster state prediction method and related devices, first, the battery state data of each battery cell in the target battery cluster is determined; then, the battery state data of each battery cell is input into the trained state prediction model, and the predicted voltage entropy data of each battery cell is determined according to the output of the trained state prediction model; then, the abnormal battery cell is determined according to the predicted voltage entropy data of each battery cell; finally, an equivalent circuit model is constructed for the abnormal battery cell to monitor the abnormal battery cell, and a warning message is generated if it is determined that the abnormal battery cell is abnormal. Early warning can be given for a battery cluster composed of a large number of batteries, and abnormal battery cells in the battery cluster can be determined in advance for a long time, and abnormal battery cells can be monitored, which greatly improves the accuracy of early warning of the battery cluster while saving system power consumption. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying any creative work.

[0020] Figure 1 A system architecture diagram of a battery cluster state prediction method provided in an embodiment of the present application;

[0021] Figure 2 A schematic diagram of a flow chart of a method for predicting a state of a battery cluster provided in an embodiment of the present application;

[0022] Figure 3 A schematic diagram of the structure of a trained state prediction model provided in an embodiment of the present application;

[0023] Figure 4 A schematic diagram of a flow chart of another method for predicting the state of a battery cluster provided in an embodiment of the present application;

[0024] Figure 5 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application;

[0025] Figure 6 A block diagram of the functional units of a battery cluster state prediction device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described 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.

[0027] The terms "first", "second", etc. in the specification and claims of this application and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices.

[0028] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article indicates that the associated objects before and after are in an "or" relationship. The "plurality" appearing in the embodiments of the present application refers to two or more.

[0029] In the embodiments of the present application, "at least one item" or similar expressions refer to any combination of these items, including any combination of single items or plural items, and refer to one or more, and multiple refers to two or more. For example, at least one item of a, b, or c can represent the following seven situations: a, b, c, a and b, a and c, b and c, a, b, and c. Among them, each of a, b, and c can be an element or a set containing one or more elements.

[0030] The "connection" that appears in the embodiments of the present application refers to various connection methods such as direct connection or indirect connection to achieve communication between devices, and the embodiments of the present application do not impose any limitations on this.

[0031] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0032] The following is an explanation of the relevant contents, concepts, meanings, technical issues, technical solutions, beneficial effects, etc. involved in the embodiments of the present application.

[0033] Battery Cell: A single battery, the smallest unit of a battery. A battery cell is the smallest energy storage unit of an energy storage system. It consists of a positive electrode, a negative electrode, a separator, and an electrolyte, and is responsible for the conversion of chemical energy into electrical energy. Battery cells can be of different types, such as lithium-ion, nickel-cadmium, lead-acid, etc.

[0034] Battery Module / Pack: A standard package of a series of individual batteries. A battery module usually consists of multiple cells that can be connected in series (to increase voltage) or in parallel (to increase current). Battery modules are the basis for building larger battery units.

[0035] Battery Cluster: A storage unit consisting of a series of battery modules. A battery cluster consists of multiple battery modules that are managed as a unit to provide the required voltage and current levels. A battery cluster can contain tens of thousands of cells, depending on the design and scale of the energy storage system.

[0036] In the battery energy storage system (BESS), the above components work together to ensure the effective storage and conversion of energy to meet the needs of different application scenarios. The battery cluster is the core of energy storage.

[0037] In the embodiments of the present application, a battery cell may refer to a battery module or a battery cell, which is not specifically limited herein.

[0038] At present, as the degree of aging increases, the battery capacity continues to decay. The capacity definition is characterized by the ratio of the current available capacity of the battery to the initial rated capacity of the battery, that is, the battery health status is predicted by comparing the battery capacity; another manifestation of battery aging is the increase in battery internal resistance, which can also be used to predict the battery health status. However, most of the current methods predict the battery change status in a short period of time for early warning, and how to predict the battery status after a long time has become a difficult problem.

[0039] To solve the above problems, the embodiments of the present application provide a state prediction method and related devices for a battery cluster, which can provide early warning for a battery cluster composed of a large number of batteries, determine abnormal battery cells in the battery cluster in advance, and monitor the abnormal battery cells, thereby greatly improving the early warning accuracy of the battery cluster while saving system power consumption.

[0040] First, see Figure 1 , Figure 1 This is a system architecture diagram of a battery cluster state prediction method provided in an embodiment of the present application. Figure 1 As shown, a battery cluster 110 and a processing system 120 are included, wherein the processing system 120 is electrically connected to the battery cluster 110 .

[0041] The battery cluster 110 may include a plurality of battery cells.

[0042] Among them, the processing system 120 can be a battery management system (BMS), including an analog front end unit (AFE), a main control unit, a slave control unit, a communication interface, etc., wherein the analog front end unit is the part of the BMS used to collect battery voltage, temperature and current readings. It is usually responsible for measurements that are physically closer to the battery and controls the circuit breaker to disconnect the battery from the system in the event of a fault. Among them, the main control unit is responsible for the control of the battery cell level in the battery module, including battery cell voltage and temperature signal monitoring, balancing control, charge state estimation, battery health state estimation and thermal management control. Among them, the slave control unit is responsible for module-level control, including control of system relays, alarm signal output, and input and output with external devices. It summarizes the status information of all batteries in the series battery pack and can be displayed on a touch screen or communicated with a computer. Among them, the BMS exchanges data with external devices through a communication interface such as a CAN bus to achieve remote monitoring and control.

[0043] In the embodiment of the present application, the processing system 120 is equipped with a trained state prediction model, which can first determine the battery state data of each battery cell in the target battery cluster, and then determine the abnormal battery cells that will become abnormal after a long period of time through the trained state prediction model, so as to perform targeted monitoring and timely early warning, without the need to monitor all battery cells, which greatly improves the early warning accuracy of the battery cluster while saving system power consumption.

[0044] After understanding the software and hardware architecture in the embodiment of the present application, Figure 2 A method for predicting the state of a battery cluster in an embodiment of the present application is described. Figure 2 A schematic flow chart of a battery cluster state prediction method provided in an embodiment of the present application specifically includes the following steps:

[0045] Step 201 : determining the battery status data of each battery cell in the target battery cluster.

[0046] The target battery cluster may include multiple battery cells, and the battery status data may reflect multiple parameter indicators of the battery cells.

[0047] In a possible embodiment, the voltage data of each battery cell within a preset voltage range and the current data within the preset voltage range can be obtained, and then the voltage mean, voltage standard deviation, voltage entropy, voltage skewness, voltage kurtosis and voltage slope of each battery cell can be determined based on the voltage data of each battery cell; and the current mean, current standard deviation, current entropy, current skewness, current kurtosis and current slope of each battery cell can be determined based on the current data of each battery cell.

[0048] Specifically, the preset voltage range can be the range from 0.2V before the charge cut-off voltage to the charge cut-off voltage. The voltage Shannon entropy is obtained by calculating the voltage change from 0.2V before the charge cut-off voltage to the charge cut-off voltage, and the voltage mean, voltage standard deviation, voltage skewness, voltage kurtosis and voltage slope of the same time period are calculated at the same time. Similarly, the current data in the range from 0.2V before the charge cut-off voltage to the charge cut-off voltage are selected to calculate the current mean, current standard deviation, current Shannon entropy, current skewness, current kurtosis and current slope. The changes in voltage and current during the charging time are considered to contain rich battery information, so the data of this time period is specially selected as the model input, which also reduces the impact of data noise at other times on the model. At the same time, the battery is more likely to fail when it reaches the charge cut-off voltage. Selecting the interval of 0.2V before the charge cut-off voltage of the battery cell as the battery status data of each battery cell can make the subsequent prediction results more accurate. It should be noted that the method for calculating voltage entropy includes selecting a time window and dividing it into intervals, that is, selecting a time window containing all battery cell voltage data, and dividing it into intervals in the time dimension. The time window here can be the period from 0.2V before the charging cut-off voltage to the charging cut-off voltage, and then calculating the entropy value of each battery cell. The entropy value calculation formula can be used to calculate the entropy value of each battery cell, and then the time window is slid to obtain the voltage entropy value of each battery cell in a time period. Similarly, the method for calculating current entropy includes selecting a time window and dividing it into intervals, that is, selecting a time window containing all battery cell current data, and dividing it into intervals in the time dimension. The time window here can be the period from 0.2V before the charging cut-off voltage to the charging cut-off voltage, and then calculating the entropy value of each battery cell. The entropy value calculation formula can be used to calculate the entropy value of each battery cell, and then the time window is slid to obtain the current entropy value of each battery cell in a time period. I will not go into details here.

[0049] Step 202: input the battery state data of each battery cell into a trained state prediction model, and determine the predicted voltage entropy data of each battery cell according to the output of the trained state prediction model.

[0050] Among them, the trained state prediction model includes a convolutional layer module, a long short-term memory network layer module and a fully connected layer module. The battery state data of each battery cell can be input into the convolutional layer module, and the first feature data can be determined according to the output of the convolutional layer module; the first feature data can be input into the long short-term memory network layer module, and the second feature data can be determined according to the output of the long short-term memory network layer module; the second feature data can be input into the fully connected layer module, and the predicted voltage entropy data of each battery cell can be determined according to the output of the fully connected layer module.

[0051] In a possible embodiment, the convolution layer module may include a convolution module, which includes convolution, activation and pooling functions. It can be understood that the convolution module can extract local features of the battery status data and reduce the dimension through pooling to retain the main features to obtain the first feature data.

[0052] In a possible embodiment, the convolutional layer module may include at least two convolutional modules, each of which includes convolution, activation and pooling functions. It can be understood that the first convolutional module can extract local features of the battery status data and reduce the dimension through pooling to retain the main features. The second convolutional module can deepen the feature extraction and reduce the dimension again through pooling to obtain the first feature data. It can be understood that the convolutional layer module can also include three or even more convolutional modules. The above description is only an exemplary description and does not represent a limitation on the embodiments of the present application, so it will not be repeated here.

[0053] It can be understood that the convolutional layer module is responsible for extracting local features in the input data, such as patterns or trends, which is very effective for short-term dependencies or periodic patterns in battery cycle data. Through the convolution operation, the convolutional layer module can effectively capture short-term dependencies and reduce the amount of data input to the long short-term memory network layer module, making the model more efficient, especially when processing multi-dimensional time series data.

[0054] In a possible embodiment, the long short-term memory network layer module can be used to capture long-term dependencies. This is because the memory units of the long short-term memory network layer module can store sequence information, so it can capture important time dependencies in multiple time steps and enhance the ability of long-term prediction.

[0055] In a possible embodiment, the predicted voltage entropy data of each battery cell output by the fully connected layer module may be the voltage entropy data of each battery cell after ten battery cycles. The number of battery cycles refers to the number of cycles when the battery completes a complete charging cycle, that is, a complete charging and discharging process, and the number of cycles is increased by one.

[0056] For ease of understanding, the following Figure 3A trained state prediction model in an embodiment of the present application is described. Figure 3 A schematic diagram of the structure of a trained state prediction model provided in an embodiment of the present application. It can be seen that the convolution layer module includes n convolution modules, CNN1-CNN2 respectively. n It means that after the battery status data of each battery cell is input into the convolutional layer module, the convolutional layer module will output the first feature data (not shown in the figure) to the long short-term memory network layer module, and the long short-term memory network will determine the second feature data according to the first feature data. It should be noted that the long short-term memory network layer module will process the first feature data and obtain internal feature data. The internal feature data is represented by b1-b n and B1-B n Indicates that the second feature data is h1-h n It indicates that the fully connected layer module outputs the predicted voltage entropy data through the second feature data, which will not be elaborated here.

[0057] It can be seen that by inputting the battery status data of each battery cell into the trained status prediction model and determining the predicted voltage entropy data of each battery cell according to the output of the trained status prediction model, the preset voltage entropy of each battery cell after a long time can be obtained, providing reliable data support for status prediction.

[0058] Step 203 : determining abnormal battery cells according to the predicted voltage entropy data of each battery cell.

[0059] Among them, the average voltage entropy data and the standard deviation voltage entropy data can be determined according to the predicted voltage entropy data of each battery cell, and a voltage entropy confidence interval can be constructed according to the average voltage entropy data and the standard deviation voltage entropy data, and the battery cell corresponding to the predicted voltage entropy data whose value does not belong to the voltage entropy confidence interval is determined as the abnormal battery cell.

[0060] It can be understood that the confidence interval is set to determine whether the battery is in a normal state. This design can improve the generalization ability of the model, because the model does not need to accurately determine the health status of the battery by judging the absolute value of the voltage entropy, and the working conditions of each energy storage power station are quite different, so the absolute value of its voltage entropy will also vary greatly. For example, the voltage entropy confidence interval calculation method is as follows: a 95% confidence interval (μ-2σ, μ+2σ) is set, where μ is the mean voltage entropy and σ is the standard deviation of the voltage entropy. This means that if a data is randomly selected from the sample, there is a 95% probability that it is in this (μ-2σ, μ+2σ) range.

[0061] It can be seen that by determining abnormal battery cells based on the predicted voltage entropy data of each battery cell, abnormal battery cells after a long period of time can be preliminarily determined, which is convenient for targeted monitoring. There is no need to monitor each battery cell, which greatly reduces system power consumption.

[0062] Step 204 , constructing an equivalent circuit model for the abnormal battery cell to monitor the abnormal battery cell, and generating a warning message if it is determined that the abnormal battery cell is abnormal.

[0063] Among them, an equivalent circuit model of the abnormal battery cell can be constructed; then, parameter identification is performed on the equivalent circuit model to determine the battery voltage data of the abnormal battery cell; finally, if the battery voltage data of the abnormal battery cell does not fall within the battery voltage threshold range, the warning information is generated.

[0064] Specifically, the parameter identification of the equivalent circuit model to determine the battery voltage data of the abnormal battery cell includes: updating the state space equation of the equivalent circuit model according to the voltage change data and current change data of the abnormal battery cell; determining the measurement error and Kalman gain by recursive least square method with forgetting factor, and updating the covariance matrix and model parameters of the battery to determine the battery voltage data of the abnormal battery cell. The existing parameter identification algorithm can be referred to, and will not be described in detail here.

[0065] It should be noted that if it is determined that the abnormal battery cell is abnormal after several battery cycles, the equivalent circuit model of the abnormal battery cell can be constructed and monitored when the several battery cycles are about to be reached. If no abnormality actually occurs after several battery cycles, the monitoring can be stopped. In a possible embodiment, if it is determined at the current moment that the abnormal battery cell is abnormal after several battery cycles, and at the next moment it is determined that the abnormal battery cell is abnormal after several battery cycles, it can be understood that the abnormal period of the abnormal battery cell will last from the current moment to the next moment. Then, the equivalent circuit model of the abnormal battery cell can be constructed and monitored when the above-mentioned several battery cycles are about to be reached, and it will last from the current moment to the next moment, and so on, which will not be elaborated here.

[0066] It can be seen that an equivalent circuit model is constructed for the abnormal battery cell to monitor the abnormal battery cell. If the abnormal battery cell is determined to be abnormal, a warning message is generated. Only the abnormal battery cell can be finely monitored to improve the accuracy of state prediction.

[0067] It can be seen that through the above-mentioned battery cluster state prediction method, first, the battery state data of each battery cell in the target battery cluster is determined; then, the battery state data of each battery cell is input into the trained state prediction model, and the predicted voltage entropy data of each battery cell is determined according to the output of the trained state prediction model; then, the abnormal battery cell is determined according to the predicted voltage entropy data of each battery cell; finally, an equivalent circuit model is constructed for the abnormal battery cell to monitor the abnormal battery cell, and a warning message is generated if it is determined that the abnormal battery cell is abnormal. Early warning can be given for a battery cluster composed of a large number of batteries, and abnormal battery cells in the battery cluster can be determined in advance for a long time, and abnormal battery cells can be monitored, which greatly improves the accuracy of early warning of the battery cluster while saving system power consumption.

[0068] Combine the following Figure 4 Another method for predicting the state of a battery cluster in an embodiment of the present application is described. Figure 4 Another method for predicting the state of a battery cluster provided in an embodiment of the present application specifically includes the following steps:

[0069] Step 401: Acquire training battery status data of each training battery cell in a training battery cluster.

[0070] Among them, the training battery status data of each training battery cell includes the voltage mean, voltage standard deviation, voltage entropy, voltage skewness, voltage kurtosis and voltage slope of each training battery cell; and the current mean, current standard deviation, current entropy, current skewness, current kurtosis and current slope of each training battery cell.

[0071] Step 402: input the training battery state data into a basic model, and determine the first voltage entropy data of each training battery cell after a preset period of time according to the output of the basic model.

[0072] Among them, the basic model includes convolutional layer modules, long short-term memory network layer modules and fully connected layer modules.

[0073] Step 403 , comparing the first voltage entropy data with the second voltage entropy data of each training battery cell after the preset period of time, and updating the basic model through a loss function until the basic model converges to obtain the trained state prediction model.

[0074] The preset time period may be a time period of ten battery cycles.

[0075] Step 404 , determining the battery status data of each battery cell in the target battery cluster.

[0076] Step 405, inputting the battery state data of each battery cell into a trained state prediction model, and determining predicted voltage entropy data of each battery cell according to an output of the trained state prediction model;

[0077] Step 406, determining abnormal battery cells according to the predicted voltage entropy data of each battery cell;

[0078] Step 407 , constructing an equivalent circuit model for the abnormal battery cell to monitor the abnormal battery cell, and generating a warning message if it is determined that the abnormal battery cell is abnormal.

[0079] It can be seen that through the above-mentioned battery cluster state prediction method, first, the battery state data of each battery cell in the target battery cluster is determined; then, the battery state data of each battery cell is input into the trained state prediction model, and the predicted voltage entropy data of each battery cell is determined according to the output of the trained state prediction model; then, the abnormal battery cell is determined according to the predicted voltage entropy data of each battery cell; finally, an equivalent circuit model is constructed for the abnormal battery cell to monitor the abnormal battery cell, and if it is determined that the abnormal battery cell is abnormal, a warning message is generated. Early warning can be given to a battery cluster composed of a large number of batteries, and abnormal battery cells in the battery cluster can be determined in advance for a long time, and abnormal battery cells can be monitored, which greatly improves the accuracy of early warning of the battery cluster while saving system power consumption. Due to the use characteristics of energy storage batteries, the charging method is basically the same, and the discharging method is various based on the use requirements. Therefore, only the voltage data and current data of the last charging time period are used to predict the battery health, which can bypass the influence caused by different SOH estimation methods of the BMS system.

[0080] The steps not described in detail above can be referred to Figure 2 The description of the steps of some of the methods will not be repeated here.

[0081] Combine the following Figure 5 An electronic device in an embodiment of the present application is described. Figure 5 Schematic diagram of the structure of an electronic device according to an embodiment of the present application, wherein the electronic device 500 includes a processor 501 , a memory 502 , and a communication bus 503 for connecting the processor 501 and the memory 502 .

[0082] In some possible implementations, the memory 502 includes but is not limited to random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM) or portable read-only memory (CD-ROM), and the memory 502 is used to store the program code executed by the electronic device 500 and the transmitted data.

[0083] In some possible implementations, the electronic device 500 further includes a communication interface for receiving and sending data.

[0084] In some possible implementations, the processor 501 may be one or more central processing units (CPUs). When the processor 501 is a central processing unit (CPU), the central processing unit (CPU) may be a single-core central processing unit (CPU) or a multi-core central processing unit (CPU).

[0085] In some possible implementations, the processor 501 may be a baseband chip, a chip, a central processing unit (CPU), a general-purpose processor, a DSP, an ASIC, an FPGA or other programmable logic device, a transistor logic device, a hardware component or any combination thereof.

[0086] In a specific implementation, the processor 501 in the electronic device 500 is used to execute the program instructions 521 stored in the memory 502 to perform the following operations:

[0087] Determine battery status data of each battery cell in the target battery cluster;

[0088] Inputting the battery state data of each battery cell into a trained state prediction model, and determining the predicted voltage entropy data of each battery cell according to the output of the trained state prediction model;

[0089] Determine an abnormal battery cell according to the predicted voltage entropy data of each battery cell;

[0090] An equivalent circuit model is constructed for the abnormal battery cell to monitor the abnormal battery cell, and a warning message is generated if it is determined that the abnormal battery cell is abnormal.

[0091] It should be noted that the specific implementation of each operation can adopt the corresponding description of the method embodiment shown above, and the electronic device 500 can be used to execute the above method embodiment of the present application, which will not be described in detail.

[0092] It can be seen that through the above-mentioned battery cluster state prediction method and related devices, first, the battery state data of each battery cell in the target battery cluster is determined; then, the battery state data of each battery cell is input into the trained state prediction model, and the predicted voltage entropy data of each battery cell is determined according to the output of the trained state prediction model; then, the abnormal battery cell is determined according to the predicted voltage entropy data of each battery cell; finally, an equivalent circuit model is constructed for the abnormal battery cell to monitor the abnormal battery cell, and a warning message is generated if it is determined that the abnormal battery cell is abnormal. Early warning can be given for a battery cluster composed of a large number of batteries, and abnormal battery cells in the battery cluster can be determined in advance for a long time, and abnormal battery cells can be monitored, which greatly improves the accuracy of early warning of the battery cluster while saving system power consumption.

[0093] The above mainly introduces the scheme of the embodiment of the present application from the perspective of the execution process on the method side. It is understandable that, in order to realize the above functions, the electronic device includes a hardware structure and / or software module corresponding to the execution of each function. It should be easily appreciated by those skilled in the art that, in combination with the units and algorithm steps of each example described in the embodiments provided herein, the present application can be implemented in the form of hardware or a combination of hardware and computer software. Whether a function is executed in the form of hardware or computer software driving hardware depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present application.

[0094] The embodiment of the present application can divide the electronic device into functional units according to the above method example. For example, each functional unit can be divided according to each function, or two or more functions can be integrated into one processing unit. The above integrated unit can be implemented in the form of hardware or in the form of software functional units. It should be noted that the division of units in the embodiment of the present application is schematic and is only a logical function division. There may be other division methods in actual implementation.

[0095] In the case of dividing each functional module into corresponding functional modules, Figure 6 This is a block diagram of the functional units of a battery cluster state prediction device provided in an embodiment of the present application. The battery cluster state prediction device 600 includes:

[0096] A state determination unit 610, configured to determine battery state data of each battery cell in a target battery cluster;

[0097] A first prediction unit 620, configured to input the battery state data of each battery cell into a trained state prediction model, and determine predicted voltage entropy data of each battery cell according to an output of the trained state prediction model;

[0098] An abnormality determination unit 630, configured to determine an abnormal battery cell according to the predicted voltage entropy data of each battery cell;

[0099] The second prediction unit 640 is used to construct an equivalent circuit model for the abnormal battery cell to monitor the abnormal battery cell, and generate warning information if it is determined that the abnormal battery cell is abnormal.

[0100] In a possible embodiment, in determining the battery status data of each battery cell in the target battery cluster, the status determination unit 610 is specifically used to:

[0101] Acquiring voltage data of each battery cell within a preset voltage range and current data within the preset voltage range;

[0102] Determine the voltage mean, voltage standard deviation, voltage entropy, voltage skewness, voltage kurtosis and voltage slope of each battery cell according to the voltage data of each battery cell; and

[0103] A current mean, a current standard deviation, a current entropy, a current skewness, a current kurtosis, and a current slope of each battery cell are determined according to the current data of each battery cell.

[0104] In a possible embodiment, the trained state prediction model includes a convolutional layer module, a long short-term memory network layer module, and a fully connected layer module; in inputting the battery state data of each battery cell into the trained state prediction model, and determining the predicted voltage entropy data of each battery cell according to the output of the trained state prediction model, the first prediction unit 620 is specifically used to:

[0105] Inputting the battery status data of each battery cell into the convolutional layer module, and determining the first feature data according to the output of the convolutional layer module;

[0106] Inputting the first feature data into the long short-term memory network layer module, and determining the second feature data according to the output of the long short-term memory network layer module;

[0107] The second characteristic data is input into the fully connected layer module, and the predicted voltage entropy data of each battery cell is determined according to the output of the fully connected layer module.

[0108] In a possible embodiment, in determining the abnormal battery cell according to the predicted voltage entropy data of each battery cell, the abnormality determination unit 630 is specifically used to:

[0109] Determine average voltage entropy data and standard deviation voltage entropy data according to the predicted voltage entropy data of each battery cell;

[0110] Constructing a voltage entropy confidence interval according to the average voltage entropy data and the standard deviation voltage entropy data;

[0111] The battery cell corresponding to the predicted voltage entropy data whose value does not belong to the voltage entropy confidence interval is determined as the abnormal battery cell.

[0112] In a possible embodiment, in terms of constructing an equivalent circuit model for the abnormal battery cell to monitor the abnormal battery cell and generating warning information if it is determined that the abnormal battery cell is abnormal, the second prediction unit 640 is specifically used to:

[0113] constructing an equivalent circuit model of the abnormal battery cell;

[0114] Performing parameter identification on the equivalent circuit model to determine battery voltage data of the abnormal battery cell;

[0115] If the battery voltage data of the abnormal battery cell does not belong to the battery voltage threshold range, the warning information is generated.

[0116] In a possible embodiment, in the aspect of performing parameter identification on the equivalent circuit model to determine the battery voltage data of the abnormal battery cell, the second prediction unit 640 is specifically used for:

[0117] updating the state space equation of the equivalent circuit model according to the voltage change data and the current change data of the abnormal battery cell;

[0118] The measurement error and the Kalman gain are determined by a recursive least square method with a forgetting factor, and the covariance matrix and model parameters of the battery are updated to determine the battery voltage data of the abnormal battery cell.

[0119] In a possible embodiment, before determining the battery status data of each battery cell in the target battery cluster, the first prediction unit 620 is specifically configured to:

[0120] Acquire training battery status data of each training battery cell in the training battery cluster;

[0121] Inputting the training battery state data into a basic model, and determining the first voltage entropy data of each training battery cell after a preset period of time according to the output of the basic model;

[0122] The first voltage entropy data is compared with the second voltage entropy data of each training battery cell after the preset period of time, and the basic model is updated by a loss function until the basic model converges to obtain the trained state prediction model.

[0123] It should be noted that the specific implementation of each operation may adopt the corresponding description of the method embodiment shown above, and the battery cluster state prediction device 600 may be used to execute the above method embodiment of the present application, which will not be described in detail.

[0124] An embodiment of the present application also provides a chip, including a processor, a memory, and a computer program or instructions stored in the memory, wherein the processor executes the computer program or instructions to implement the steps described in the above method embodiment.

[0125] An embodiment of the present application also provides a chip module, including a transceiver component and a chip, the chip including a processor, a memory and a computer program or instructions stored in the memory, wherein the processor executes the computer program or instructions to implement the steps described in the above method embodiment.

[0126] An embodiment of the present application also provides a computer storage medium, wherein the computer storage medium stores a computer program for electronic data exchange, wherein the computer program enables a computer to execute part or all of the steps of any method recorded in the above method embodiment, and the above computer includes an electronic device.

[0127] The embodiment of the present application also provides a computer program product, the computer program product includes a non-transitory computer-readable storage medium storing a computer program, the computer program is operable to cause a computer to execute some or all of the steps of any method described in the method embodiment. The computer program product may be a software installation package, and the computer includes an electronic device.

[0128] It should be noted that, for the above-mentioned various embodiments, for the sake of simple description, they are all expressed as a series of action combinations. Those skilled in the art should be aware that the present application is not limited by the described order of actions, because some steps in the embodiments of the present application can be performed in other orders or simultaneously. In addition, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions, steps, modules or units involved are not necessarily required by the embodiments of the present application.

[0129] In the above embodiments, the embodiments of the present application have different focuses on the description of each embodiment. For parts that are not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0130] The steps of the method or algorithm described in the embodiments of the present application can be implemented in hardware or by executing software instructions by a processor. The software instructions can be composed of corresponding software modules, and the software modules can be stored in RAM, flash memory, ROM, EPROM, electrically erasable programmable read-only memory (electrically EPROM, EEPROM), registers, hard disks, mobile hard disks, read-only compact disks (CD-ROMs) or any other form of storage medium known in the art. An exemplary storage medium is coupled to the processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an ASIC. In addition, the ASIC can be located in a terminal device or a management device. Of course, the processor and the storage medium can also be present in a terminal device or a management device as discrete components.

[0131] Those skilled in the art should be aware that in one or more of the above examples, the functions described in the embodiments of the present application can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiments of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, computer, server, or data center to another website site, computer, server, or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrations. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital video disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)).

[0132] The modules / units included in the devices and products described in the above embodiments may be software modules / units or hardware modules / units, or may be partially software modules / units and partially hardware modules / units. For example, for the devices and products applied to or integrated in the chip, the modules / units included therein may all be implemented in the form of hardware such as circuits, or at least some of the modules / units may be implemented in the form of software programs, which run on the processor integrated inside the chip, and the remaining (if any) modules / units may be implemented in the form of hardware such as circuits; for the devices and products applied to or integrated in the chip module, the modules / units included therein may all be implemented in the form of hardware such as circuits, and different modules / units may be located in the same component (such as a chip, circuit module, etc.) or in different components of the chip module, or at least some of the modules / units may be implemented in the form of software programs. The software programs run on the processor integrated inside the chip, and the remaining (if any) modules / units may be implemented in the form of hardware such as circuits. It is implemented in the form of a software program, which runs on a processor integrated inside the chip module, and the remaining (if any) modules / units can be implemented in hardware such as circuits; for various devices and products applied to or integrated in the terminal equipment, the various modules / units contained therein can be implemented in hardware such as circuits, and different modules / units can be located in the same component (for example, chip, circuit module, etc.) or in different components in the terminal equipment, or, at least some modules / units can be implemented in the form of a software program, which runs on a processor integrated inside the terminal equipment, and the remaining (if any) modules / units can be implemented in hardware such as circuits.

[0133] The specific implementation methods described above further illustrate the purpose, technical solutions and beneficial effects of the embodiments of the present application. It should be understood that the above description is only the specific implementation method of the embodiments of the present application and is not intended to limit the protection scope of the embodiments of the present application. Any modifications, equivalent substitutions, improvements, etc. made on the basis of the technical solutions of the embodiments of the present application should be included in the protection scope of the embodiments of the present application.

Claims

1. A method for predicting the state of a battery cluster, characterized in that: The method comprises: Determine battery status data of each battery cell in the target battery cluster; Inputting the battery state data of each battery cell into a trained state prediction model, and determining the predicted voltage entropy data of each battery cell according to the output of the trained state prediction model; Determine an abnormal battery cell according to the predicted voltage entropy data of each battery cell; An equivalent circuit model is constructed for the abnormal battery cell to monitor the abnormal battery cell, and a warning message is generated if it is determined that the abnormal battery cell is abnormal.

2. The method according to claim 1, characterized in that The step of determining the battery status data of each battery cell in the target battery cluster includes: Acquiring voltage data of each battery cell within a preset voltage range and current data within the preset voltage range; Determine the voltage mean, voltage standard deviation, voltage entropy, voltage skewness, voltage kurtosis and voltage slope of each battery cell according to the voltage data of each battery cell; and A current mean, a current standard deviation, a current entropy, a current skewness, a current kurtosis, and a current slope of each battery cell are determined according to the current data of each battery cell.

3. The method according to claim 1, characterized in that The trained state prediction model includes a convolutional layer module, a long short-term memory network layer module and a fully connected layer module; the battery state data of each battery cell is input into the trained state prediction model, and the predicted voltage entropy data of each battery cell is determined according to the output of the trained state prediction model, including: Inputting the battery status data of each battery cell into the convolutional layer module, and determining the first feature data according to the output of the convolutional layer module; Inputting the first feature data into the long short-term memory network layer module, and determining the second feature data according to the output of the long short-term memory network layer module; The second characteristic data is input into the fully connected layer module, and the predicted voltage entropy data of each battery cell is determined according to the output of the fully connected layer module.

4. The method according to claim 1, characterized in that: The determining of abnormal battery cells according to the predicted voltage entropy data of each battery cell comprises: Determine average voltage entropy data and standard deviation voltage entropy data according to the predicted voltage entropy data of each battery cell; Constructing a voltage entropy confidence interval according to the average voltage entropy data and the standard deviation voltage entropy data; The battery cell corresponding to the predicted voltage entropy data whose value does not belong to the voltage entropy confidence interval is determined as the abnormal battery cell.

5. The method according to claim 1, characterized in that: The constructing an equivalent circuit model for the abnormal battery cell to monitor the abnormal battery cell, and generating warning information if it is determined that the abnormal battery cell is abnormal, includes: constructing an equivalent circuit model of the abnormal battery cell; Performing parameter identification on the equivalent circuit model to determine battery voltage data of the abnormal battery cell; If the battery voltage data of the abnormal battery cell does not belong to the battery voltage threshold range, the warning information is generated.

6. The method according to claim 5, characterized in that The performing parameter identification on the equivalent circuit model to determine the battery voltage data of the abnormal battery cell includes: updating the state space equation of the equivalent circuit model according to the voltage change data and the current change data of the abnormal battery cell; The measurement error and the Kalman gain are determined by a recursive least square method with a forgetting factor, and the covariance matrix and model parameters of the battery are updated to determine the battery voltage data of the abnormal battery cell.

7. The method according to claim 1, characterized in that Before determining the battery status data of each battery cell in the target battery cluster, the method further includes: Obtaining training battery status data of each training battery cell in the training battery cluster; Inputting the training battery state data into a basic model, and determining the first voltage entropy data of each training battery cell after a preset period of time according to an output of the basic model; The first voltage entropy data is compared with the second voltage entropy data of each training battery cell after the preset period of time, and the basic model is updated by a loss function until the basic model converges to obtain the trained state prediction model.

8. A battery cluster state prediction device, characterized in that: The device comprises: A state determination unit, used to determine the battery state data of each battery cell in the target battery cluster; A first prediction unit, configured to input the battery state data of each battery cell into a trained state prediction model, and determine predicted voltage entropy data of each battery cell according to an output of the trained state prediction model; an abnormality determination unit, configured to determine an abnormal battery cell according to the predicted voltage entropy data of each battery cell; The second prediction unit is used to construct an equivalent circuit model for the abnormal battery cell to monitor the abnormal battery cell, and generate warning information if it is determined that the abnormal battery cell is abnormal.

9. An electronic device, characterized in that: include: A processor, a memory, and one or more programs; the one or more programs are stored in the memory and are configured to be executed by the processor, the programs comprising instructions for executing the steps in the method according to any one of claims 1 to 7.

10. A computer storage medium, characterized in that: The computer storage medium stores a computer program, wherein the computer program includes program instructions, and when the program instructions are executed by a processor, the processor is caused to perform the method according to any one of claims 1 to 7.

Images