Battery parameter prediction method, device, equipment, medium and program product

Abstract

The present application provides a battery parameter prediction method, device, equipment, medium and program product. It relates to the technical field of battery parameter prediction. The method comprises: when the operating data of the battery to be tested is obtained, the operating state corresponding to the battery to be tested is determined according to the trained state determination model and the operating data; the operating data includes the current operating data and / or historical operating data of the battery to be tested; the state model is obtained by training the constructed initial state determination model according to the historical operating data of multiple batteries and their corresponding operating states; the target parameter prediction model corresponding to the operating state is obtained, and the parameters of the battery to be tested are determined according to the current operating data and the target parameter prediction model; the parameters include the voltage parameters of the battery to be tested and the battery surface temperature parameters. The method of the present application solves the technical problem of low accuracy of battery parameter prediction results in the prior art.

Description

Technical Field

[0001] The present application relates to the technical field of battery parameter prediction, and in particular to a battery parameter prediction method, device, equipment, medium and program product. Background Art

[0002] In today's energy field, lithium-ion batteries are widely used in many fields such as electric vehicles and portable electronic devices due to their high energy density, long cycle life and good charge and discharge performance. During the use of lithium-ion batteries, accurate prediction of various battery parameters, such as voltage parameters and battery surface temperature, is crucial to ensure safe and efficient operation of the battery.

[0003] At present, the technology of predicting battery voltage and surface temperature based on the current parameters of lithium-ion batteries and parameters such as ambient temperature has been studied and applied to a certain extent. However, there is a key problem in the existing technology, that is, the differences between different working conditions such as charging, discharging, and static are not fully considered. For example, under charging conditions, the battery accepts external input electrical energy, the current continues to flow in, complex electrochemical reactions occur inside the battery, and the structure and chemical composition of the electrode material are constantly changing, which not only causes the voltage to rise rapidly, but also causes the battery temperature to rise due to the heat release of the reaction and the heat generated by the internal resistance. In the discharge condition, the battery outputs electrical energy, the current flows out, the electrochemical reaction direction is opposite, the voltage gradually decreases, and the battery temperature is also affected by the reaction heat and internal resistance, but the change trend is different from that during charging. Under static conditions, although the battery seems to be in a static state, there is still a slow self-discharge phenomenon inside, which will cause the battery voltage to slowly decrease. At the same time, the battery temperature will also fluctuate with the change of ambient temperature, and the chemical balance inside the battery is also slowly adjusted.

[0004] Since existing prediction technologies fail to effectively distinguish these complex working conditions, there are limitations in model construction and parameter setting, and it is impossible to accurately capture the changing patterns of battery parameters under different working conditions. This results in a large deviation between the prediction results and the actual situation, and the prediction accuracy is low, which in turn affects the assessment of the health status of lithium-ion batteries and the reliable operation of related equipment. Summary of the invention

[0005] The present application provides a battery parameter prediction method, device, equipment, medium and program product to solve the technical problem of low accuracy of battery parameter prediction results in the prior art.

[0006] In a first aspect, the present application provides a battery parameter prediction method, comprising:

[0007] When the operating data of the battery to be tested is obtained, the operating state corresponding to the battery to be tested is determined according to the trained state determination model and the operating data; the operating data includes the current operating data and / or historical operating data of the battery to be tested; the state model is obtained by training the constructed initial state determination model according to the historical operating data of multiple batteries and their corresponding operating states; the operating state includes the operating condition of the battery to be tested, and the operating condition corresponding to the operating condition;

[0008] A target parameter prediction model corresponding to the operating state is obtained, and parameters of the battery to be tested are determined according to the current operating data and the target parameter prediction model; the parameters include voltage parameters and battery surface temperature parameters of the battery to be tested.

[0009] In an optional implementation manner, the current operation data includes current current data and current ambient temperature data corresponding to the operation of the battery to be tested;

[0010] Determining the operating state corresponding to the battery to be tested according to the trained state determination model and the operating data includes:

[0011] Calculate a first feature set according to the current current data and the current ambient temperature data to obtain a first feature set of the battery to be tested;

[0012] The first feature set is input into the state determination model to obtain the operating state corresponding to the battery to be tested.

[0013] In an optional implementation manner, the historical operation data includes historical current data, historical voltage data, and historical ambient temperature data corresponding to the operation of the battery to be tested;

[0014] Determining the operating state corresponding to the battery to be tested according to the trained state determination model and the operating data includes:

[0015] Calculating a second feature set according to the historical current data, the historical voltage data, and the historical ambient temperature data to obtain a second feature set of the battery to be tested;

[0016] The second feature set is input into the state determination model to obtain the operating state corresponding to the battery to be tested; the operating state includes the operating condition of the battery to be tested and the operating state corresponding to the operating condition.

[0017] In an optional implementation manner, obtaining a target parameter prediction model corresponding to the operating state, and determining the parameters of the battery to be tested according to the current operating data and the target parameter prediction model, includes:

[0018] Determining a target parameter prediction model corresponding to the battery to be tested from a plurality of candidate parameter prediction models according to the operating state;

[0019] The current operating parameters of the battery to be tested are input into the target parameter prediction model to obtain the parameters of the battery to be tested.

[0020] In an optional implementation, the method further includes:

[0021] Acquire training data, the training data including first operation data not including voltage parameters of the battery when any battery is operated at a historical moment, second operation data including voltage parameters of the battery, and an operation state corresponding to the battery;

[0022] According to the training data and the preset optimization algorithm, the constructed initial state determination model is optimized and trained to obtain a trained state determination model; the trained state determination model is used to determine the operating state of the battery to be tested according to the operating data of the battery to be tested.

[0023] In an optional implementation, the state determination model is a tree model; and the parameter prediction model is a time series model.

[0024] In a second aspect, the present application provides a battery parameter prediction device, comprising:

[0025] An operating state determination module is used to determine the operating state corresponding to the battery to be tested according to the trained state determination model and the operating data when the operating data of the battery to be tested is obtained; the operating data includes the current operating data and / or historical operating data of the battery to be tested; the state model is obtained by training the constructed initial state determination model according to the historical operating data of multiple batteries and their corresponding operating states;

[0026] A parameter prediction module is used to obtain a target parameter prediction model corresponding to the operating state, and determine the parameters of the battery to be tested according to the current operating data and the target parameter prediction model; the parameters include the voltage parameters and battery surface temperature parameters of the battery to be tested.

[0027] In a third aspect, the present application provides an electronic device, comprising: a processor, and a memory communicatively connected to the processor;

[0028] The memory stores computer-executable instructions;

[0029] The processor executes the computer-executable instructions stored in the memory to implement the method according to the first aspect.

[0030] In a fourth aspect, the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the method described in the first aspect.

[0031] In a fifth aspect, the present application provides a computer program product, including a computer program, which implements the method described in the first aspect when executed by a processor.

[0032] The battery parameter prediction technology provided in this application can accurately determine the operating state of the battery to be tested through the trained state determination model and the operating data of the battery to be tested; specifically, it includes whether the battery to be tested is in operating conditions such as charging, discharging or static, and whether it is in transient or steady state under the above conditions. The judgment of the above operating state can lay a solid foundation for subsequent parameter prediction, because the dispersion of battery parameters and the randomness of actual working conditions are coupled with each other, resulting in completely different characteristics and parameter change laws of the battery under different operating conditions. After clarifying the operating state, the corresponding target parameter prediction model is obtained, which changes the previous "one size fits all" prediction mode. Because the electrochemical reactions and energy conversion mechanisms inside the battery are different under different operating conditions, the exclusive model can be more in line with the actual situation and capture the parameter change laws in a targeted manner; on this basis, the voltage parameters and battery surface temperature parameters of the battery to be tested are determined based on the current operating data and the adapted target parameter prediction model, which fully considers the influence of different operating conditions and greatly improves the accuracy of the prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0034] Figure 1 An application scenario diagram of the battery parameter prediction method provided in this application;

[0035] Figure 2 A schematic diagram of a flow chart of a battery parameter prediction method provided in an embodiment of the present application;

[0036] Figure 3 A schematic diagram of the structure of a battery parameter prediction device provided in an embodiment of the present application;

[0037] Figure 4 It is a block diagram of an electronic device shown in an embodiment of the present application.

[0038] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0039] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0040] In the technical solution of this application, the collection, storage, use, processing, transmission, provision and disclosure of information such as financial data or user data involved shall comply with the provisions of relevant laws and regulations and shall not violate public order and good morals.

[0041] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0042] In today's energy field, lithium-ion batteries are widely used in many fields such as electric vehicles and portable electronic devices due to their high energy density, long cycle life and good charge and discharge performance. During the use of lithium-ion batteries, accurate prediction of various battery parameters, such as voltage parameters and battery surface temperature, is crucial to ensure safe and efficient operation of the battery.

[0043] At present, the technology of predicting battery voltage and surface temperature based on the current parameters of lithium-ion batteries and parameters such as ambient temperature has been studied and applied to a certain extent. However, there is a key problem in the existing technology, that is, the differences between different working conditions such as charging, discharging, and static are not fully considered. For example, under charging conditions, the battery accepts external input electrical energy, the current continues to flow in, complex electrochemical reactions occur inside the battery, and the structure and chemical composition of the electrode material are constantly changing, which not only causes the voltage to rise rapidly, but also causes the battery temperature to rise due to the heat release of the reaction and the heat generated by the internal resistance. In the discharge condition, the battery outputs electrical energy, the current flows out, the electrochemical reaction direction is opposite, the voltage gradually decreases, and the battery temperature is also affected by the reaction heat and internal resistance, but the change trend is different from that during charging. Under static conditions, although the battery seems to be in a static state, there is still a slow self-discharge phenomenon inside, which will cause the battery voltage to slowly decrease. At the same time, the battery temperature will also fluctuate with the change of ambient temperature, and the chemical balance inside the battery is also slowly adjusted.

[0044] Since existing prediction technologies fail to effectively distinguish these complex working conditions, there are limitations in model construction and parameter setting, and it is impossible to accurately capture the changing patterns of battery parameters under different working conditions. This results in a large deviation between the prediction results and the actual situation, and the prediction accuracy is low, which in turn affects the assessment of the health status of lithium-ion batteries and the reliable operation of related equipment.

[0045] The battery parameter prediction method provided in this application is intended to solve the above technical problems of the prior art. Specifically, before the battery parameter prediction is performed, the operating state of the battery to be tested is accurately determined by the trained state determination model and the operating data of the battery to be tested; specifically, the operating conditions of the battery to be tested, such as charging, discharging or static, and the operating state of the transient or steady state under the above conditions. The judgment of the above operating state can lay a solid foundation for the subsequent parameter prediction, because the dispersion of battery parameters and the randomness of actual working conditions are coupled with each other, resulting in completely different characteristics and parameter change laws of the battery under different operating conditions. After clarifying the operating state, the corresponding target parameter prediction model is obtained, which changes the previous "one size fits all" prediction mode. Because the electrochemical reactions and energy conversion mechanisms inside the battery are different under different operating conditions, the exclusive model can be more in line with the actual situation and capture the parameter change law in a targeted manner; on this basis, the voltage parameters and battery surface temperature parameters of the battery to be tested are determined based on the current operating data and the adapted target parameter prediction model, which fully considers the influence of different operating states and greatly improves the accuracy of the prediction.

[0046] Figure 1An application scenario diagram of the battery parameter prediction method provided in this application. The battery parameter prediction technical solution provided in this application is of great value in multiple fields and application scenarios, especially in situations where efficient management and optimization of battery performance are required. The following are some of the main application scenarios: 1. In the field of electric vehicles, specifically by predicting the battery voltage and temperature to obtain the battery health status, optimize the charging and discharging strategy, extend the battery life, and improve the vehicle's endurance; 2. In the field of renewable energy storage, specifically in solar and wind energy systems, predict battery performance to optimize energy storage and release, and improve system efficiency and reliability; 3. In the field of consumer electronics products, specifically, through battery parameter prediction, optimize charging management, and extend device usage time and battery life.

[0047] In summary, the above application scenarios demonstrate the extensive value of battery parameter prediction technology in improving equipment performance, optimizing energy management, improving safety and extending battery life. Specifically, through accurate battery status prediction, various systems can operate more efficiently and take timely measures when necessary to avoid potential failures and losses.

[0048] For ease of understanding, the following Figure 1 Describe any application scenario to which the embodiments of the present application are applicable. Figure 1 Taking the battery in the field of energy storage system as an example, the battery parameter prediction solution provided in this application involves an operation data acquisition module 11 and a parameter prediction module 12.

[0049] Specifically, the operation data acquisition module 11 acquires historical operation data and / or current operation data generated by the battery under test during actual operation or test operation, and transmits the historical operation data and / or current operation data to the parameter prediction module 12 for parameter prediction.

[0050] Specifically, in the parameter prediction module 12, the operating state corresponding to the battery to be tested is determined according to the historical operating data and / or the current operating data; then the target parameter prediction model corresponding to the operating state is obtained, and the parameters of the battery to be tested are determined according to the current operating data and the target parameter prediction model. In the above process, the operating data is matched with a suitable target parameter prediction model, and the voltage parameters and battery surface temperature parameters of the battery to be tested are determined by the current operating data and the adapted target parameter prediction model. In this way, compared with the prediction method that does not distinguish the operating state, it can more truly reflect the parameter changes of the battery in actual operation, and provide strong guarantees for the safe use of the battery, performance evaluation and stable operation of the equipment.

[0051] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0052] Figure 2 A flowchart of a battery parameter prediction method provided in an embodiment of the present application. The method can be executed by a battery parameter prediction device, which can be a server or an electronic device. The following is an example of an electronic device. The method in this embodiment can be implemented by software, hardware, or a combination of software and hardware, such as Figure 2 As shown, the method includes the following steps.

[0053] S201 . When the operating data of a battery to be tested is obtained, the operating state corresponding to the battery to be tested is determined according to a trained state determination model and the operating data.

[0054] In the present application, the operating data includes the current operating data and / or historical operating data of the battery to be tested; the state model is obtained by training a constructed initial state determination model based on the historical operating data of multiple batteries and their corresponding operating states; the operating state includes the operating condition of the battery to be tested, and the operating condition state corresponding to the operating condition.

[0055] In this application, a large amount of operating data can be obtained from a preset database of the energy storage system, for example, operating data of a large amount of batteries to be tested included in some commercial projects.

[0056] The operating data in the present application includes historical operating data of battery parameters at corresponding times, and current operating data of battery parameter positions at corresponding times.

[0057] For example, the battery to be tested is continuously operated from a preset historical moment to the current moment, and the entire operation process is stored in a preset database. When performing parameter prediction, the above operation data is acquired at one time and can be used as the current operation data of the battery to be tested. The operating state of the battery to be tested is predicted based on the current operation data, and then the corresponding battery parameters are predicted based on the operating state and the current operation data.

[0058] For another example, in some cases, during the operation of the battery to be tested, the operation data of the shorter duration is stored in a preset database at short intervals, and parameter prediction is performed for the shorter duration. It should be noted that, because at the current moment, the battery parameters corresponding to the operation data at the previous historical moment have been predicted, that is, the voltage data has been obtained, so in order to more accurately predict the operating conditions of the battery to be tested, the operation data with a shorter duration from the current moment and the battery parameters can be combined as the historical operation data when predicting the operating state of the battery to be tested.

[0059] It should be understood that the acquired operation data is the collected raw data, and in order to facilitate the accurate identification of the subsequent operation status, the present application also performs data preprocessing on the acquired raw data, including but not limited to data deletion, vacancy filling, and interpolation smoothing.

[0060] On this basis, the processed operating data of each battery to be tested are respectively input into the pre-trained state determination model to obtain the operating state corresponding to each battery to be tested as output by the model, so as to facilitate subsequent targeted parameter prediction and improve the accuracy of the prediction results.

[0061] In the embodiment of the present application, the operating condition corresponding to the battery to be tested may be determined by using the historical operating data of the battery to be tested or the current operating data.

[0062] It can be explained that the current operation data in this application can be interpreted as the operation data corresponding to the battery under test obtained at the current moment within a longer period of time; specifically including the current current data and current ambient temperature data of the battery under test. The historical operation data can be interpreted as the historical current data and historical ambient temperature data corresponding to the battery under test at historical moments, and on this basis, the historical operation data also includes the historical voltage data corresponding to the above historical current data.

[0063] Optionally, when the acquired operating data of the battery to be tested is the current operating data, correspondingly, determining the operating state of the battery to be tested specifically includes: calculating a first feature set based on the current current data and the current ambient temperature data to obtain a first feature set of the battery to be tested; inputting the first feature set into the state determination model to obtain the operating state corresponding to the battery to be tested.

[0064] In the present application, the first feature set is at least one data set calculated based on the battery current data and ambient temperature data according to a pre-set setting; specifically including but not limited to: average current, cumulative charging / discharging / resting time, cumulative charging / discharging amount, current extreme difference, current change rate, current acceleration (i.e., the rate of change of the rate of change) and other data.

[0065] Specifically, based on obtaining the current current data and the current ambient temperature data of the battery to be tested, the current current data and the current ambient temperature data are calculated and processed according to a calculation method corresponding to a preset first feature data set to obtain a first feature set corresponding to the battery to be tested.

[0066] Furthermore, the first feature set is input into a pre-trained state determination model to obtain the operating state corresponding to the battery to be tested output by the model.

[0067] In some possible examples, the present application also performs feature dimensionality reduction processing on the above-mentioned first feature set according to predetermined feature correlations, so as to avoid that high-correlation features may affect the accuracy of subsequent parameter prediction results.

[0068] Exemplarily, the first feature set of the battery to be tested obtained according to the above calculation method can be input into a pre-trained dimensionality reduction model to obtain the reduced first feature set output by the dimensionality reduction model, and the reduced first feature set can be input into the state determination model to obtain the operating state of the battery to be tested.

[0069] It should be understood that when the battery to be tested is in different states under the same operating condition, it will also have different impacts on parameter prediction. Therefore, the operating state predicted in the present application includes the operating condition of the battery to be tested, that is, (charging, static or discharging), and also includes the operating state corresponding to each operating condition, that is, (steady state or transient state).

[0070] In another optional embodiment, when the acquired operating data of the battery to be tested is historical operating data, correspondingly, determining the operating state of the battery to be tested specifically includes: calculating a second feature set based on the historical current data, the historical voltage data and the historical ambient temperature data to obtain a second feature set of the battery to be tested; inputting the second feature set into the state determination model to obtain the operating state corresponding to the battery to be tested; the operating state includes the operating condition of the battery to be tested, and the operating state corresponding to the operating condition.

[0071] In the present application, the second feature set, based on the first feature set, also includes a data set calculated based on the voltage data of the battery; specifically, the second feature set includes but is not limited to: average current, cumulative charging / discharging / resting time, cumulative charging / discharging amount, current range, current change rate, current acceleration (i.e., the rate of change of the rate of change), peak / average voltage, correlation coefficient between current and voltage, voltage skewness / kurtosis, current skewness / kurtosis, maximum / minimum / average value of the ratio of current to voltage, and maximum / minimum / average value of the square of the ratio of current to voltage.

[0072] Specifically, since the historical operation data of the battery to be tested includes historical current data, historical voltage data and historical temperature data, based on this, based on the calculation method of each feature in the second feature set, the above historical operation data is used for calculation and processing to obtain the second feature set of the battery to be tested.

[0073] Furthermore, the second feature set is input into a pre-trained state determination model to obtain the operating state corresponding to the battery to be tested output by the model.

[0074] In some possible examples, the present application also performs feature dimensionality reduction processing on the above-mentioned second feature set according to predetermined feature correlations, so as to also avoid that high-correlation features will affect the accuracy of subsequent parameter prediction results.

[0075] Exemplarily, the first feature set of the battery to be tested obtained according to the above calculation method can be input into a pre-trained dimensionality reduction model to obtain a reduced-dimensional second feature set output by the dimensionality reduction model, and the reduced-dimensional second feature set can be input into the state determination model to obtain the operating state of the battery to be tested.

[0076] Based on the above implementation, the state prediction model pre-built and trained in the present application may be a tree model. For example, the tree model may be a GBDT model, an XGBOOST model, a LightGBM model, and the like.

[0077] In the present application, when constructing an initial state prediction model based on a tree model, the following method can be used for training to obtain a trained state prediction model. The training process of the tree model includes: obtaining training data, wherein the training data includes first operating data including the voltage parameters of the battery when any battery is operated at a historical moment, second operating data not including the voltage parameters of the battery, and the operating state corresponding to the battery; optimizing and training the constructed initial state determination model according to the training data and the selected optimization algorithm to obtain a trained state determination model; the trained state determination model is used to determine the operating state of the battery to be tested according to the operating data of the battery to be tested.

[0078] It should be understood that the basic data used to predict the operating status of the battery in this application is the operating data of the battery. Since the voltage data of the battery cannot be obtained in real time during operation, the voltage data corresponding to the battery when it was running at a historical moment can be obtained. Therefore, in order to meet the requirement that the subsequent battery can also predict the operating status during real-time operation, the training data in this application may include two types of data, one is the first operating data that does not include the voltage parameters of the battery, and the other is the second operating data that includes the voltage parameters of the battery. Of course, the labels of the operating status corresponding to these two types of data are the same.

[0079] It should be noted that the label of the running status includes the running condition label and the condition status label corresponding to the running condition.

[0080] Specifically, the operating conditions include charging, discharging or static. Exemplarily, the operating condition threshold range is set to -5A~5A according to the charging and discharging status; for example, the battery with current data greater than 5A is marked as a charging condition, the battery with current data between -5A and 5A is marked as a static condition, and the battery with current data less than -5A is marked as a discharging condition. Furthermore, the operating state includes transient or steady state. Exemplarily, the battery state is marked according to the charging and discharging state and the duration after the state switching; for example, if the current state is maintained for more than 30 minutes, it is marked as a steady state, otherwise it is marked as a transient state.

[0081] Furthermore, feature set calculation is performed on the operating data in the above training data to obtain the corresponding feature set, and the preset optimization algorithm is used to process the above feature set in the constructed tree model to obtain the predicted operating state; and based on the predicted operating state and the labeled operating state, the model parameters of the tree model are optimized, and after multiple rounds of training, a trained state determination model is obtained.

[0082] Exemplarily, the optimization algorithm may be a grid search, a gradient search, a genetic algorithm, or the like.

[0083] S202: Obtain a target parameter prediction model corresponding to the operating state, and determine the parameters of the battery to be tested according to the current operating data and the target parameter prediction model.

[0084] The present application includes multiple candidate parameter prediction models, any of which is a parameter prediction model obtained by training a pre-built timing model with multiple sets of operating data under the same operating state as input and preset battery characteristic parameters as output.

[0085] In the present application, the pre-built timing model may include but is not limited to: Long Short-Term Memory (LSTM) model, Gated Recurrent Unit (GRU), Transformer model, etc.

[0086] Before training the time series model, the acquired operation data is pre-checked for data quality and cleaned to improve the training performance of the model. Specifically, data cleaning includes marking abnormal current, abnormal voltage, and abnormal temperature data in the operation data, and processing the marked data according to a preset strategy to complete the cleaning. Exemplary preset strategies include: data deletion, vacancy filling, interpolation smoothing, etc.

[0087] On this basis, taking the current data and ambient temperature data of any battery within a period of time (for example, 2 hours) as an example, feature extraction processing is performed on them to obtain current series and ambient temperature series.

[0088] Furthermore, the sequence data corresponding to the multiple batteries are divided into training sets, validation sets and test sets to complete the model training. In addition, in order to ensure the robustness of the model, the entire model training process requires multiple data set divisions and multiple rounds of model training.

[0089] It should be noted that in any round of training, the current series data and ambient temperature series of any sample battery are used as input, and a preset optimization algorithm is used to perform parameter prediction on the above data in the timing model to obtain the predicted voltage and predicted battery surface temperature of the sample battery. Then, according to the predicted voltage and predicted surface temperature of the sample battery, as well as the label voltage and label surface temperature of the sample battery, the model parameters of the timing model are adjusted to obtain a trained candidate parameter prediction model.

[0090] Exemplarily, the preset optimization algorithm may include but is not limited to: grid search, gradient search, genetic algorithm, etc.

[0091] It should be noted that in the existing solutions, during the modeling and training of the battery, parameters such as the battery health status and the battery internal resistance are usually used as labels to train the completed model. However, during the above training process, it is difficult to obtain accurate labels containing the above data, resulting in poor model training quality and subsequent poor prediction results. In the present application, the voltage and surface temperature that can be directly measured are used as labels, and the above data can be directly and accurately measured, resulting in a better training effect of the constructed battery parameter prediction model, and the battery parameters obtained based on this prediction are more accurate, and the battery health results obtained based on the battery parameter reasoning evaluation are also more accurate.

[0092] Optionally, based on multiple sets of training data under different operating states and the above-mentioned training method, candidate parameter prediction models corresponding to different operating states are trained respectively. Further, a first mapping relationship is generated according to each operating state and the corresponding candidate parameter prediction model, so that the parameter prediction model matching the different operating states can be determined according to the first mapping relationship, so as to facilitate more accurate parameter prediction.

[0093] On this basis, in the present application, when the operating state of the battery to be tested is determined, the process of predicting the parameters of the battery to be tested may include: determining a target parameter prediction model corresponding to the battery to be tested from multiple candidate parameter prediction models according to the operating state; inputting the current operating parameters of the battery to be tested into the target parameter prediction model to obtain the parameters of the battery to be tested.

[0094] Specifically, according to the operating state of the battery to be tested, a candidate parameter prediction model corresponding to the operating state of the battery to be tested is searched in the first mapping relationship, and the model is determined as the target parameter prediction model.

[0095] Furthermore, the current operating data of the battery to be tested is obtained, and the current operating data is input into the target parameter prediction model to obtain the voltage data and battery surface temperature data of the battery to be tested output by the model.

[0096] On this basis, the battery health status and performance evaluation can be evaluated based on the voltage data and battery surface temperature data. For example, by monitoring the battery voltage, the battery health status can be determined. For example, too low a voltage may indicate over-discharge of the battery, while too high a voltage may indicate over-charging; and an abnormal increase in the battery surface temperature may indicate an internal short circuit or other fault in the battery. For another example, by combining the voltage and temperature data, the remaining capacity and health status of the battery can be estimated, and the battery charge and discharge efficiency at different temperatures can be analyzed to optimize the use conditions.

[0097] In the above technical solution, the operating state of the battery to be tested is accurately determined by the trained state determination model and the operating data of the battery to be tested; specifically, the operating conditions of the battery to be tested, such as charging, discharging or static, and the operating conditions of the transient or steady state under the above conditions. The above operating state judgment can lay a solid foundation for the subsequent parameter prediction, because the dispersion of battery parameters and the randomness of actual working conditions are coupled with each other, resulting in completely different characteristics and parameter change laws of the battery under different operating conditions. After the operating state is clarified, the corresponding target parameter prediction model is obtained, which changes the previous "one size fits all" prediction mode. Because the electrochemical reactions and energy conversion mechanisms inside the battery are different under different operating conditions, the exclusive model can be more in line with the actual situation and capture the parameter change laws in a targeted manner; on this basis, the voltage parameters and battery surface temperature parameters of the battery to be tested are determined based on the current operating data and the adapted target parameter prediction model, which fully considers the influence of different operating conditions and greatly improves the accuracy of the prediction.

[0098] Figure 3 This is a schematic diagram of the structure of a battery parameter prediction device provided in an embodiment of the present application. Figure 3The battery parameter prediction device 30 includes: an operation state determination module 301 and a parameter prediction module 302; wherein,

[0099] The operating state determination module 301 is used to determine the operating state corresponding to the battery to be tested according to the trained state determination model and the operating data when the operating data of the battery to be tested is obtained; the operating data includes the current operating data and / or historical operating data of the battery to be tested; the state model is obtained by training the constructed initial state determination model according to the historical operating data of multiple batteries and their corresponding operating states;

[0100] The parameter prediction module 302 is used to obtain the target parameter prediction model corresponding to the operating state, and determine the parameters of the battery to be tested according to the current operating data and the target parameter prediction model; the parameters include the voltage parameters and battery surface temperature parameters of the battery to be tested.

[0101] In an optional implementation manner, the current operation data includes current current data and current ambient temperature data corresponding to the operation of the battery to be tested; the operation state determination module 301 includes:

[0102] A first feature set obtaining unit, configured to calculate a first feature set according to the current current data and the current ambient temperature data, to obtain a first feature set of the battery to be tested;

[0103] The first operating state obtaining unit is used to input the first feature set into the state determination model to obtain the operating state corresponding to the battery to be tested.

[0104] In an optional implementation manner, the historical operation data includes historical current data, historical voltage data, and historical ambient temperature data corresponding to the operation of the battery to be tested; the operation state determination module 301 includes:

[0105] A second feature set obtaining unit, configured to calculate a second feature set according to the historical current data, the historical voltage data and the historical ambient temperature data, to obtain a second feature set of the battery to be tested;

[0106] The second operating state obtaining unit is used to input the second feature set into the state determination model to obtain the operating state corresponding to the battery to be tested; the operating state includes the operating condition of the battery to be tested and the operating state corresponding to the operating condition.

[0107] In an optional implementation, the parameter prediction module includes:

[0108] a target parameter prediction model determining unit, configured to determine a target parameter prediction model corresponding to the battery to be tested from a plurality of candidate parameter prediction models according to the operating state;

[0109] The parameter prediction unit is used to input the current operating parameters of the battery to be tested into the target parameter prediction model to obtain the parameters of the battery to be tested.

[0110] In an optional embodiment, the device further comprises:

[0111] a training data acquisition module, configured to acquire training data, the training data comprising first operation data not including voltage parameters of the battery when any battery is operated at a historical moment, second operation data including voltage parameters of the battery, and an operation state corresponding to the battery;

[0112] The operating status acquisition module is used to optimize and train the constructed initial state determination model according to the training data and a preset optimization algorithm to obtain a trained state determination model; the trained state determination model is used to determine the operating state of the battery to be tested according to the operating data of the battery to be tested.

[0113] In an optional implementation, the state determination model is a tree model; and the parameter prediction model is a time series model.

[0114] Figure 4 is a block diagram of an electronic device shown in an embodiment of the present application, and the device may be a computer, a digital broadcast terminal, etc. Figure 4 , the device 800 may include one or more of the following components: a processing component 802 , a memory 804 , a power component 806 , a multimedia component 808 , an audio component 810 , an input / output interface 812 , a sensor component 814 , and a communication component 816 .

[0115] The processing component 802 generally controls the overall operation of the device 800, such as operations associated with display, phone calls, data communications, camera operations, and recording operations. The processing component 802 may include one or more processors 820 to execute instructions to complete all or part of the steps of the above-mentioned method. In addition, the processing component 802 may include one or more modules to facilitate the interaction between the processing component 802 and other components. For example, the processing component 802 may include a multimedia module to facilitate the interaction between the multimedia component 808 and the processing component 802.

[0116] The memory 804 is configured to store various types of data to support operations on the device 800. Examples of such data include instructions for any application or method operating on the device 800, contact data, phone book data, messages, pictures, videos, etc. The memory 804 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0117] The power supply component 806 provides power to the various components of the device 800. The power supply component 806 can include a power management system, one or more power supplies, and other components associated with generating, managing, and distributing power for the device 800.

[0118] The multimedia component 808 includes a screen that provides an output interface between the device 800 and the user. In some embodiments, the screen may include a liquid crystal display (LCD) and a touch panel (TP). If the screen includes a touch panel, the screen may be implemented as a touch screen to receive input signals from the user. The touch panel includes one or more touch sensors to sense touch, slide, and gestures on the touch panel. The touch sensor may not only sense the boundaries of the touch or slide action, but also detect the duration and pressure associated with the touch or slide operation. In some embodiments, the multimedia component 808 includes a front camera and / or a rear camera. When the device 800 is in an operating mode, such as a shooting mode or a video mode, the front camera and / or the rear camera may receive external multimedia data. Each front camera and rear camera may be a fixed optical lens system or have a focal length and optical zoom capability.

[0119] The audio component 810 is configured to output and / or input audio signals. For example, the audio component 810 includes a microphone (MIC), and when the device 800 is in an operating mode, such as a call mode, a recording mode, and a speech recognition mode, the microphone is configured to receive an external audio signal. The received audio signal can be further stored in the memory 804 or sent via the communication component 816. In some embodiments, the audio component 810 also includes a speaker for outputting audio signals.

[0120] The input / output interface 812 provides an interface between the processing component 802 and the peripheral interface modules, which may be keyboards, click wheels, buttons, etc. These buttons may include but are not limited to: a home button, a volume button, a start button, and a lock button.

[0121] The sensor assembly 814 includes one or more sensors for providing various aspects of status assessment for the device 800. For example, the sensor assembly 814 can detect the open / closed state of the device 800, the relative positioning of components, such as the display and keypad of the device 800, and the sensor assembly 814 can also detect the position change of the device 800 or a component of the device 800, the presence or absence of user contact with the device 800, the orientation or acceleration / deceleration of the device 800, and the temperature change of the device 800. The sensor assembly 814 may include a proximity sensor configured to detect the presence of nearby objects without any physical contact. The sensor assembly 814 may also include an optical sensor, such as a solid image (Complementary Metal Oxide Semiconductor, CMOS) sensor or a semiconductor image (Charge-coupled Device, CCD) sensor, for use in imaging applications. In some embodiments, the sensor assembly 814 may also include an acceleration sensor, a gyroscope sensor, a magnetic sensor, a pressure sensor, or a temperature sensor.

[0122] The communication component 816 is configured to facilitate wired or wireless communication between the device 800 and other devices. The device 800 can access a wireless network based on a communication standard, such as WiFi, 4G or 5G, or a combination thereof. In an exemplary embodiment, the communication component 816 receives a broadcast signal or broadcast-related information from an external broadcast management system via a broadcast channel. In an exemplary embodiment, the communication component 816 also includes a near field communication (NFC) module to facilitate short-range communication. For example, the NFC module can be implemented based on radio frequency identification (RFID) technology, infrared data association (IrDA) technology, ultra-wideband (UWB) technology, Bluetooth (BT) technology and other technologies.

[0123] In an exemplary embodiment, the device 800 may be implemented by one or more application specific integrated circuits (ASIC), digital signal processors (DSP), digital signal processing devices (DSPD), programmable logic devices (PLD), field programmable gate arrays (FPGA), controllers, microcontrollers, microprocessors or other electronic components to perform the above method.

[0124] In an exemplary embodiment, a non-transitory computer-readable storage medium including instructions is also provided, such as a memory 804 including instructions, and the instructions can be executed by a processor 820 of the device 800 to perform the above method. For example, the non-transitory computer-readable storage medium can be a ROM, a random access memory (RAM), a CD-ROM, a magnetic tape, a floppy disk, an optical data storage device, etc.

[0125] A non-transitory computer-readable storage medium, when the instructions in the storage medium are executed by a processor of a server, enables the server to execute the above-mentioned battery parameter prediction method.

[0126] The embodiment of the present application also provides a chip for running instructions, which is used to execute the technical solution of the battery parameter prediction method in the above embodiment.

[0127] An embodiment of the present application further provides a computer-readable storage medium, in which computer-executable instructions are stored. When the computer-executable instructions are executed on a computer, the computer executes the technical solution of the battery parameter prediction method of the above embodiment.

[0128] An embodiment of the present application also provides a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium. When at least one processor executes the computer program, the technical solution of the battery parameter prediction method in the above embodiment can be implemented.

[0129] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0130] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

[0131] Those skilled in the art will readily appreciate other embodiments of the present application after considering the specification and practicing the invention disclosed herein. The present application is intended to cover any modification, use or adaptation of the present application, which follows the general principles of the present application and includes common knowledge or customary techniques in the art that are not disclosed in the present application. The specification and examples are intended to be exemplary only, and the true scope and spirit of the present application are indicated by the following claims.

[0132] It should be understood that the present application is not limited to the precise structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present application is limited only by the appended claims.

Claims

1. A battery parameter prediction method, characterized in that: The method comprises: When the operating data of the battery to be tested is obtained, the operating state corresponding to the battery to be tested is determined according to the trained state determination model and the operating data; the operating data includes the current operating data and / or historical operating data of the battery to be tested; the state determination model is obtained by training the constructed initial state determination model according to the historical operating data of multiple batteries and their corresponding operating states; the operating state includes the operating condition of the battery to be tested, and the operating condition state corresponding to the operating condition; A target parameter prediction model corresponding to the operating state is obtained, and parameters of the battery to be tested are determined according to the current operating data and the target parameter prediction model; the parameters include voltage parameters and battery surface temperature parameters of the battery to be tested.

2. The method according to claim 1, characterized in that The current operation data includes current current data and current ambient temperature data corresponding to the operation of the battery to be tested; Determining the operating state corresponding to the battery to be tested according to the trained state determination model and the operating data includes: Calculate a first feature set according to the current current data and the current ambient temperature data to obtain a first feature set of the battery to be tested; The first feature set is input into the state determination model to obtain the operating state corresponding to the battery to be tested.

3. The method according to claim 1, characterized in that The historical operation data includes historical current data, historical voltage data and historical ambient temperature data corresponding to the operation of the battery to be tested; Determining the operating state corresponding to the battery to be tested according to the trained state determination model and the operating data includes: Calculating a second feature set according to the historical current data, the historical voltage data, and the historical ambient temperature data to obtain a second feature set of the battery to be tested; The second feature set is input into the state determination model to obtain the operating state corresponding to the battery to be tested; the operating state includes the operating condition of the battery to be tested and the operating state corresponding to the operating condition.

4. The method according to claim 1, characterized in that: Acquiring a target parameter prediction model corresponding to the operating state, and determining the parameters of the battery to be tested according to the current operating data and the target parameter prediction model, including: Determining a target parameter prediction model corresponding to the battery to be tested from a plurality of candidate parameter prediction models according to the operating state; The current operating parameters of the battery to be tested are input into the target parameter prediction model to obtain the parameters of the battery to be tested.

5. The method according to any one of claims 1 to 3, characterized in that: The method further comprises: Acquire training data, the training data including first operation data not including voltage parameters of the battery when any battery is operated at a historical moment, second operation data including voltage parameters of the battery, and an operation state corresponding to the battery; According to the training data and the preset optimization algorithm, the constructed initial state determination model is optimized and trained to obtain a trained state determination model; the trained state determination model is used to determine the operating state of the battery to be tested according to the operating data of the battery to be tested.

6. The method according to any one of claims 1 to 3, characterized in that The state determination model is a tree model; the parameter prediction model is a time series model.

7. A battery parameter prediction device, characterized in that: The device comprises: An operating state determination module is used to determine the operating state corresponding to the battery to be tested according to the trained state determination model and the operating data when the operating data of the battery to be tested is obtained; the operating data includes the current operating data and / or historical operating data of the battery to be tested; the state determination model is obtained by training the constructed initial state determination model according to the historical operating data of multiple batteries and their corresponding operating states; A parameter prediction module is used to obtain a target parameter prediction model corresponding to the operating state, and determine the parameters of the battery to be tested according to the current operating data and the target parameter prediction model; the parameters include the voltage parameters and battery surface temperature parameters of the battery to be tested.

8. An electronic device, characterized in that: include: A processor and a memory communicatively connected to the processor; The memory stores computer-executable instructions; When executing the computer-executable instructions, the processor is used to implement the battery parameter prediction method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a processor, they are used to implement the battery parameter prediction method according to any one of claims 1 to 6.

10. A computer program product, characterized in that The method comprises a computer program, which implements the method according to any one of claims 1 to 6 when the computer program is executed by a processor.

Images