Battery SOH estimation method and device, storage medium and terminal

Abstract

The invention discloses a battery SOH estimation method and device, a storage medium and a terminal. The method comprises the following steps: acquiring SN codes of all batteries in a battery cluster and charging data of the battery cluster; the charging data comprises charging duration, charging voltage and charging temperature, and the charging duration covers any cross-current charging stage; allocating an address for a battery according to the SN code, and determining a host of the battery cluster based on the address; the host is used for collecting charging data and sending the charging data to the upper computer; and calculating a multi-dimensional feature array based on the charging data, and inputting the multi-dimensional feature array into a trained SOH estimation model to obtain the SOH of the battery cluster. Full charge or full discharge is not needed, time consumption is greatly shortened, battery damage is reduced, and in-service requirements can be met.

Description

Technical Field

[0001] This invention relates to the field of battery technology, and specifically to a battery SOH estimation method, apparatus, storage medium, and terminal. Background Technology

[0002] In the field of high-voltage battery energy storage, the State of Health (SOH) is a key parameter for measuring the ratio of the battery's actual usable capacity to its rated capacity. Estimating SOH is crucial for the safe operation, lifespan management, and maintenance efficiency of energy storage systems. Existing SOH estimation methods require obtaining the battery's current total capacity through a full charge or full discharge process, and then combining this with the rated capacity to calculate SOH. However, this process is time-consuming, and during full charge, lithium ion insertion at the negative electrode approaches its theoretical upper limit, easily leading to the precipitation of lithium dendrites and their reaction with the electrolyte to form solid products. Simultaneously, the positive electrode's lattice structure becomes unstable or even collapses. During full discharge, both positive and negative electrode materials also experience wear and tear, all of which exacerbate battery aging and pose safety hazards. Furthermore, neither full charge nor full discharge provides immediate usability, failing to meet the in-service requirements of energy storage systems.

[0003] Therefore, how to overcome the shortcomings of existing technologies using effective methods has become an urgent technical challenge. Summary of the Invention

[0004] The purpose of this invention is to address the above-mentioned problems by providing a battery SOH estimation method, apparatus, storage medium, and terminal.

[0005] The technical solution of the present invention is as follows: a battery SOH estimation method, comprising the following steps: acquiring the serial number (SN) codes of all batteries in a battery cluster and the charging data of the battery cluster; the charging data includes charging duration, charging voltage, and charging temperature, wherein the charging duration covers any one of the constant current charging stages; assigning addresses to the batteries according to the SN codes, and determining the host of the battery cluster based on the addresses; the host is used to collect charging data and send the charging data to a host computer; calculating a multidimensional feature array based on the charging data, and inputting the multidimensional feature array into a trained SOH estimation model to obtain the SOH of the battery cluster.

[0006] As an improvement to this embodiment of the invention, the "any one constant current stage" specifically refers to any stage in which the battery is in a constant current charging state and the battery's state of charge (SOC) is within a first preset range.

[0007] As an improvement of this embodiment of the invention, the first preset range is 20%~80%.

[0008] As a kind of improvement of the embodiment of the application, the "address is assigned to battery according to the SN code, and the host of the battery cluster is determined based on the address", specifically includes: the SN code of all batteries in battery cluster is entered into host computer, year, month, batch number, serial number in SN code are compared and sorted in turn to obtain sorting result, the corresponding address number is assigned to each battery according to the sorting result, and the battery with the smallest address number is the host of the battery cluster.

[0009] As a kind of improvement of the embodiment of the application, the "multi-dimensional feature array is calculated based on the charging data", specifically includes: the charging data is filtered; the average temperature rise rate is calculated based on the filtered charging temperature and charging duration; the average voltage rise rate is calculated based on the filtered charging voltage and charging duration; the charging duration, average temperature rise rate and average voltage rise rate are combined as three-dimensional features to obtain the multi-dimensional feature array.

[0010] As a kind of improvement of the embodiment of the application, the "trained SOH estimation model" is trained by the following steps: obtaining charging data in feature collection interval, filtering the charging data; the feature collection interval is the cross-flow stage when battery state of charge SOC is 40%-50%; the multi-dimensional feature array and SOH true value corresponding to the charging data are calculated; the above steps are repeatedly executed until the battery is retired; the SOH estimation model based on back propagation neural network is created, the multi-dimensional feature array is taken as the input of the SOH estimation model, the corresponding SOH true value is taken as the output of the SOH estimation model, the training set, validation set and test set are divided, the SOH estimation model is trained, verified and tested, when the goodness of fit of the SOH estimation model on the training set, validation set and test set is greater than the preset threshold, the training is stopped to obtain the trained SOH estimation model.

[0011] As a kind of improvement of the embodiment of the application, the sliding average filtering algorithm is used to filter the charging data.

[0012] To achieve one of the above-mentioned purposes, one embodiment of the present application provides a battery SOH estimation method device, comprising the following modules: a data acquisition module for acquiring SN codes of all batteries in a battery cluster and charging data of the battery cluster; the charging data includes charging duration, charging voltage and charging temperature, wherein the charging duration covers any one cross-current charging phase; an address allocation module for allocating an address to the battery according to the SN code and determining a host of the battery cluster based on the address; the host is used to collect charging data and send the charging data to an upper computer; an SOH estimation module for calculating a multi-dimensional feature array based on the charging data, inputting the multi-dimensional feature array into a trained SOH estimation model to obtain the SOH of the battery cluster.

[0013] To achieve one of the above-mentioned purposes, one embodiment of the present application provides a storage medium storing program instructions, which, when executed, implement the battery SOH estimation method according to any one of the above.

[0014] To achieve one of the above-mentioned purposes, one embodiment of the present application provides an electronic terminal comprising a processor and a memory, the memory storing program instructions, and the processor running the program instructions to implement the battery SOH estimation method according to any one of the above.

[0015] The battery SOH estimation method, device, storage medium and terminal provided by the embodiments of the present application have the following advantages: the battery SOH estimation method of the present application only needs to cover any one cross-current charging phase, without full charging or full discharging, greatly shortening the time consumption; and does not involve the extreme charging and discharging interval of SOC>90% or SOC<10%, reducing the damage to the battery; the whole process does not need to interrupt the normal operation of the energy storage system, meeting the in-service demand of the energy storage system. BRIEF DESCRIPTION OF DRAWINGS

[0016] Figure 1 is a flowchart of the battery SOH estimation method of the present application; Figure 2 is a schematic diagram of the SN code of the present application; Figure 3 is a schematic diagram of the iteration process of the SOH estimation model of the present application; Figure 4 is a schematic diagram of the fitting effect of the SOH estimation model of the present application; Figure 5 is a schematic diagram of the time saving rate of the battery SOH estimation method of the present application; Figure 6 is a schematic diagram of the battery SOH estimation method device of the present application; Figure 7 is a schematic diagram of the structure of the electronic terminal of the present application. DETAILED DESCRIPTION

[0017] The present application will be described in detail below with reference to specific embodiments illustrated in the attached drawings. These embodiments are not intended to limit the present application, and the structural, method, or functional changes made by those of ordinary skill in the art based on these embodiments are included in the protection scope of the present application.

[0018] If the present application involves orientation (for example, up, down, left, right, front, back, outside, inside, etc.) when expressing, the orientation involved needs to be defined.

[0019] The scope of the embodiments herein includes the entire scope of the claims, and all available equivalents of the claims. Herein, the terms "first", "second", and the like are only used to distinguish one element from another element, and do not require or imply any actual relationship or order between the elements. In fact, the first element can also be referred to as the second element, and vice versa. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the structure, device or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such structure, device or equipment. Without more limitations, the element defined by the statement "including a" does not exclude the presence of another identical element in the structure, device or equipment including the element. Various embodiments herein are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between various embodiments can be referred to each other.

[0020] The terms "longitudinal", "transverse", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", and the like in this document indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of description and simplification of the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application. In the description herein, unless otherwise specified and limited, the terms "mount", "connect", "connect" should be understood broadly, for example, it can be a mechanical connection or an electrical connection, it can be a communication between two elements inside, it can be directly connected, or indirectly connected through an intermediate medium. For those of ordinary skill in the art, the specific meaning of the above terms can be understood according to the specific circumstances.

[0021] The present application provides a battery SOH estimation method, as shown in Figure 1 The method comprises the following steps: Step 101: obtaining SN codes of all batteries in a battery cluster and charging data of the battery cluster; the charging data includes charging duration, charging voltage and charging temperature, wherein the charging duration covers any one of the constant current charging stages; In practice, the SN codes of all batteries in the battery cluster can be collected one by one and uploaded to the upper computer by manual input or scanning by a code scanning gun. During the charging process, the charging current of the battery can be collected by a Hall sensor, the charging voltage and the charging temperature of each battery can be collected by an AFE (Analog Front-End) rectifier unit, and the charging duration can be obtained by recording the start time and the end time of the constant current charging stage, and the complete charging data including the charging duration, the charging voltage and the charging temperature can be integrated.

[0022] Here, the "any one of the constant current stages" specifically refers to any stage in which the battery is in a constant current charging state, and the state of charge SOC of the battery is in a first preset interval.

[0023] In practice, the battery cluster can be charged by using the MCC (Multistage Constant Current) charging method according to the charging map preset by the battery manufacturer. The charging map determines the corresponding constant current charging current value for the battery cluster under different working conditions according to the SOC interval and the temperature range of the battery. By setting the constant current charging current in stages and under different conditions, the charging efficiency can be improved as much as possible while ensuring the safety of the battery charging, and the charging needs of the battery under different states can be adapted. The charging duration is controlled to cover any one of the constant current charging stages, and the state of charge SOC of the battery is controlled to be in the first preset interval. Preferably, the first preset interval is 20% to 80%. It can be understood that this can avoid the high state of charge of SOC>90% and the low state of charge of SOC<10%, which can effectively reduce the side reactions such as loss of positive and negative electrode materials and precipitation of lithium dendrites, and reduce the influence of the SOH estimation process on the battery life.

[0024] Step 102: assigning an address to the battery according to the SN code, and determining a host of the battery cluster based on the address; the host is used to collect the charging data and send the charging data to the upper computer; Here, as Figure 2As shown, the SN code, from left to right, includes the product code, instruction identification code, year, month, batch number, and serial number. The year, month, batch number, and serial number in the SN code are compared and sorted sequentially to obtain a sorting result. Based on this sorting result, a corresponding address number is assigned to each battery. The battery with the smallest address number becomes the master battery of the battery cluster, and the other batteries become slave batteries. This approach standardizes the address allocation logic based on the time and production identifier of the SN code, avoiding the randomness of manually setting addresses.

[0025] Step 103: Calculate a multidimensional feature array based on the charging data, and input the multidimensional feature array into the trained SOH estimation model to obtain the SOH of the battery cluster.

[0026] Here, the "trained SOH estimation model" is obtained through the following steps: Acquire charging data within the feature acquisition range, and then filter the charging data. The feature acquisition range is the cross-current stage when the battery's state of charge (SOC) is between 40% and 50%. Calculate the multidimensional feature array and SOH true value corresponding to the charging data; Repeat the above steps until the battery is decommissioned; In practice, the charging data is first filtered using a moving average filtering algorithm. Specifically, based on the data acquisition frequency and the expected duration of the constant current charging phase, the sliding window size is set to 5-20 data points. The window size selection must balance filtering effectiveness and data timeliness, avoiding both incomplete filtering and residual random noise due to a small window, and data lag and loss of local variation characteristics due to a large window. Next, using continuously acquired charging voltage or temperature data as the processing object, the arithmetic mean of all data within the window is calculated, and the resulting average is used as the filtered data for the current window. Then, the sliding window is traversed sequentially according to the data acquisition time, repeating the above averaging steps until all charging voltage and temperature data for the entire constant current charging phase are filtered. It is understood that filtering the charging data effectively removes abnormal data points caused by sensor fluctuations, electromagnetic interference, and other factors during the charging process, making the filtered charging data closer to the actual operating state of the battery.

[0027] Based on the filtered charging temperature and charging time, the average temperature rise rate is calculated; based on the filtered charging voltage and charging time, the average voltage rise rate is calculated; the charging time, average temperature rise rate, and average voltage rise rate are combined as three-dimensional features to obtain the multi-dimensional feature array.

[0028] Here, battery aging increases ohmic and polarization resistance, leading to increased heat generation within the same time frame during any constant current charging phase, thus increasing the average temperature rise rate within that range. Furthermore, aging causes physical factors such as active particle breakage and positive electrode material lattice collapse, reducing battery capacity and resulting in a smaller charging capacity during any constant current charging phase. Externally, this manifests as an aged battery requiring less charging time compared to a new battery within the same time frame, thereby increasing the voltage rise rate. Understandably, using these three characteristics as training data for the SOH estimation model allows the model to comprehensively reflect the changing patterns of battery health, improving the model's estimation accuracy.

[0029] A backpropagation neural network-based SOH estimation model is created. The multidimensional feature array is used as the input of the SOH estimation model, and the corresponding SOH ground truth value is used as the output of the SOH estimation model. The model is divided into training set, validation set, and test set. The SOH estimation model is trained, validated, and tested. When the goodness of fit of the SOH estimation model on the training set, validation set, and test set is greater than a preset threshold, the training is stopped and the trained SOH estimation model is obtained.

[0030] In practice, the aforementioned ,in This represents the current actual capacity. Here, n represents the number of stages in the MCC (Multi-Channel Classification). The Levenberg-Marquardt algorithm can be used as an optimization algorithm, with an appropriate number of hidden layer nodes set. Then, the multi-dimensional feature array is used as the model input, and the corresponding SOH (Sort of Ohms) ground truth value is used as the model output. The training, validation, and test sets can be divided in a 7:1:2 ratio for training, validation, and testing. Figure 3 As shown, with the increase in the number of iterations, the mean squared errors of training, validation, and testing decrease rapidly and tend to stabilize, indicating that the model is continuously optimizing. Training stops when the model's goodness of fit on the training, validation, and test sets all reach a preset threshold, resulting in a well-trained SOH estimation model. Figure 4 and Figure 5 As can be seen, after iterative optimization, the mean square error of the SOH estimation model described in this invention converges to a stable low level, while the time saving rate remains around 90% as SOH changes, greatly improving the estimation accuracy and efficiency of SOH.

[0031] This invention also provides a battery SOH estimation method apparatus, such as... Figure 6 As shown, it includes the following modules: The data acquisition module 201 is used to acquire the serial numbers (SNs) of all batteries in the battery cluster and the charging data of the battery cluster; the charging data includes charging duration, charging voltage and charging temperature, wherein the charging duration covers any one of the constant current charging stages; Address allocation module 202 is used to allocate an address to the battery according to the SN code, and determine the host of the battery cluster based on the address; the host is used to collect charging data and send the charging data to the host computer; SOH estimation module 203 is used to calculate a multi-dimensional feature array based on the charging data, and input the multi-dimensional feature array into the trained SOH estimation model to obtain the SOH of the battery cluster.

[0032] The present invention also provides a storage medium storing program instructions that, when executed, implement the battery SOH estimation method as described in any of the preceding claims.

[0033] The present invention also provides an electronic terminal, such as Figure 7 As shown, it includes a processor and a memory, the memory storing program instructions, and the processor executing the program instructions to implement the battery SOH estimation method as described in any of the preceding claims.

[0034] This invention can be an apparatus, method, and / or computer program product. A computer program product may include a readable storage medium having computer-readable program instructions loaded thereon for causing a processor to implement various aspects of the invention.

[0035] Storage media can be tangible devices that hold and store instructions for use by instruction execution devices. Storage media can include, but are not limited to, electrical storage devices, magnetic storage devices, optical storage devices, electromagnetic storage devices, semiconductor storage devices, or any suitable combination thereof. More specific examples (a non-exhaustive list) of readable storage media include: portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), static random access memory (SRAM), portable compact disc read-only memory (CD-ROM), digital multifunction disc (DVD), memory sticks, floppy disks, mechanical encoding devices, such as punch cards or recessed protrusions storing instructions thereon, and any suitable combination thereof.

[0036] It should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This way of describing the specification is only for clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

[0037] The detailed descriptions listed above are merely specific descriptions of feasible embodiments of the present invention, and are not intended to limit the scope of protection of the present invention. All equivalent embodiments or modifications made without departing from the spirit of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for estimating the state of harmonics (SOH) of a battery, characterized in that, Includes the following steps: Obtain the serial number (SN) of all batteries in the battery cluster and the charging data of the battery cluster; the charging data includes charging duration, charging voltage and charging temperature, wherein the charging duration covers any one of the constant current charging stages; The battery is assigned an address based on the SN code, and the host of the battery cluster is determined based on the address; the host is used to collect charging data and send the charging data to the host computer. A multidimensional feature array is calculated based on the charging data, and the multidimensional feature array is input into the trained SOH estimation model to obtain the SOH of the battery cluster.

2. The battery SOH estimation method according to claim 1, characterized in that, The "any one constant current stage" specifically refers to any stage in which the battery is in a constant current charging state, and the battery's state of charge (SOC) is within a first preset range.

3. The battery SOH estimation method according to claim 2, characterized in that, The first preset range is 20% to 80%.

4. The battery SOH estimation method according to claim 1, characterized in that, The phrase "assigning addresses to batteries based on the SN codes and determining the host of the battery cluster based on the addresses" specifically includes: inputting the SN codes of all batteries in the battery cluster into the host computer, comparing and sorting them according to the year, month, batch number, and serial number in the SN codes to obtain a sorting result, assigning a corresponding address number to each battery according to the sorting result, and using the battery with the smallest address number as the host of the battery cluster.

5. The battery SOH estimation method according to claim 1, characterized in that, The "calculation of a multi-dimensional feature array based on the charging data" specifically includes: filtering the charging data; calculating the average rate of temperature rise based on the filtered charging temperature and charging time; calculating the average rate of voltage rise based on the filtered charging voltage and charging time; and combining the charging time, the average rate of temperature rise, and the average rate of voltage rise as three-dimensional features to obtain the multi-dimensional feature array.

6. The battery SOH estimation method according to claim 1, characterized in that, The "trained SOH estimation model" is obtained through the following steps: Acquire charging data within the feature acquisition range and filter the charging data; the feature acquisition range is the constant current stage when the battery's state of charge (SOC) is 40%-50%. Calculate the multidimensional feature array and SOH true value corresponding to the charging data; Repeat the above steps until the battery is decommissioned; A backpropagation neural network-based SOH estimation model is created. The multidimensional feature array is used as the input of the SOH estimation model, and the corresponding SOH ground truth value is used as the output of the SOH estimation model. The model is divided into training set, validation set, and test set. The SOH estimation model is trained, validated, and tested. When the goodness of fit of the SOH estimation model on the training set, validation set, and test set is greater than a preset threshold, the training is stopped and the trained SOH estimation model is obtained.

7. The battery SOH estimation method according to claim 5, characterized in that, The charging data is filtered using a moving average filtering algorithm.

8. A battery SOH estimation device, characterized in that, Includes the following modules: The data acquisition module is used to acquire the serial number (SN) codes of all batteries in the battery cluster and the charging data of the battery cluster; the charging data includes charging duration, charging voltage and charging temperature, wherein the charging duration covers any one of the constant current charging stages; The address allocation module is used to allocate an address to the battery according to the SN code, and determine the host of the battery cluster based on the address; the host is used to collect charging data and send the charging data to the host computer. The SOH estimation module is used to calculate a multi-dimensional feature array based on the charging data, and input the multi-dimensional feature array into the trained SOH estimation model to obtain the SOH of the battery cluster.

9. A storage medium storing program instructions, characterized in that, When the program instructions are executed, the battery SOH estimation method as described in any one of claims 1 to 7 is implemented.

10. An electronic terminal, characterized in that, It includes a processor and a memory, the memory storing program instructions, and the processor executing the program instructions to implement the battery SOH estimation method as described in any one of claims 1 to 7.

Images