Method and device for predicting health state of battery
By mapping multiple influencing factors of the battery to target variables and performing information fusion prediction, the prediction errors caused by the failure to consider user habits and environmental changes in existing technologies are solved, and more accurate battery SOH prediction is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-10
- Publication Date
- 2026-03-13
AI Technical Summary
Existing machine learning models do not take into account factors such as different users' charging and discharging habits and changes in ambient temperature in actual applications, which leads to a disconnect between the battery health status prediction results and the actual degradation path. Furthermore, the prediction results are inaccurate when SOH is predicted from a single dimension.
By acquiring the SOH value of the battery at the first moment and historical operating data over a preset time period, multiple variables affecting SOH are mapped to the target variable using a preset mapping relationship. The SOH value at the second moment is then predicted by combining historical operating data and mapping relationship, thus achieving multi-dimensional information fusion prediction.
It improves the accuracy of battery SOH prediction, solves the problem of the model prediction results being out of sync with the actual battery degradation path, and realizes personalized SOH prediction.
Smart Images
Figure CN121656887A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of energy storage technology, and in particular to a method and apparatus for predicting the health status of batteries. Background Technology
[0002] With the continuous growth of global power grid load, the market demand for energy storage systems is expanding rapidly. State of Health (SOH), as a key indicator reflecting the remaining lifespan of energy storage batteries, not only helps users understand the degradation trend of battery life but also facilitates the optimization of the operating environment of energy storage systems.
[0003] However, the State of Harm (SOH) of a battery cannot be directly obtained through sensors and requires indirect evaluation based on various parameters. Furthermore, the accuracy of SOH prediction is affected by a combination of factors. In existing technologies, machine learning models are often used to predict battery SOH. However, existing machine learning models do not consider factors such as different users' charging and discharging habits and changes in ambient temperature in actual applications, resulting in a disconnect between the model's prediction results and the actual degradation path of the battery. Moreover, existing machine learning models often predict SOH from a single dimension (such as the number of cycles), leading to inaccurate prediction results. Summary of the Invention
[0004] This invention provides a method for predicting battery health status, addressing the technical problems of existing machine learning models failing to consider factors such as different users' charging and discharging habits and environmental temperature changes in actual applications, leading to a disconnect between model predictions and the actual battery degradation path, and inaccurate predictions based on a single dimension of SOH. The method includes: Obtain the first SOH value of the battery at the first moment, as well as the historical operating data for a preset time period; the preset time period is the same historical time period as the time period from the first moment to the second moment. According to the preset first mapping relationship, the first SOH value is mapped to the first target variable fusion value; wherein, the first mapping relationship is the mapping relationship between the target variable fusion value and the SOH value; the target variable fusion value refers to the fusion value obtained by determining a target variable from multiple variables affecting the battery SOH and mapping the other variables among the multiple variables to the target variable; Based on the fusion value of the first target variable, historical operating data, and a preset second mapping relationship, predict the fusion value of the second target variable at the second time point; the second mapping relationship includes the mapping relationship between each other variable and the target variable. Based on the first mapping relationship, the fused value of the second target variable is mapped to the second SOH value of the battery at the second time.
[0005] Optionally, based on the fused value of the first target variable, historical operating data, and a preset second mapping relationship, the fused value of the second target variable at the second time point is predicted, including: Based on historical operational data, determine the change of at least one variable between the first and second time points; According to the second mapping relationship, the change of each variable between the first time point and the second time point is fused with the fusion value of the first target variable to obtain the fusion value of the second target variable at the second time point.
[0006] Optionally, based on the second mapping relationship, the change of each variable between the first and second time points is fused with the fused value of the first target variable to obtain the fused value of the second target variable at the second time point, including: The change of the target variable between the first and second time points is combined with the fused value of the first target variable to obtain the intermediate value of the first target variable; For each of the other variables, repeat the following steps to update the intermediate value of the first target variable: Based on the mapping relationship between other variables and the target variable, the intermediate value of the first target variable is mapped to the fused value of the first other variables; The changes of other variables between the first and second time points are combined with the first other variable fusion value to obtain the second other variable fusion value at the second time point. Based on the mapping relationship between other variables and the target variable, the fused value of the second other variables is mapped to the intermediate value of the first target variable; The final updated intermediate value of the first objective variable is used as the fused value of the second objective variable.
[0007] Optionally, based on the second mapping relationship, the change of each variable between the first and second time points is fused with the fused value of the first target variable to obtain the fused value of the second target variable at the second time point, including: Use the fused value of the first objective variable as the intermediate value of the second objective variable; For each of the other variables, repeat the following steps to update the intermediate value of the second target variable: Based on the mapping relationship between other variables and the target variable, the intermediate value of the second target variable is mapped to the fused value of the third set of other variables; The changes of other variables between the first and second time points are combined with the third other variable fusion value to obtain the fourth other variable fusion value at the second time point; Based on the mapping relationship between other variables and the target variable, the fused value of the fourth other variable is mapped to the intermediate value of the second target variable; The final updated intermediate value of the second objective variable is used as the fused value of the second objective variable.
[0008] Optionally, the method further includes determining the first mapping relationship and the second mapping relationship through the following steps: Obtain the correlation data between multiple variables and SOH, and construct a mathematical model of the target variable fusion value and SOH, and a mathematical model of each other variable and the target variable; Based on the correlation data between multiple variables and SOH, the mathematical model of the target variable fusion value and SOH is fitted to determine the first mapping relationship; and based on the correlation data between multiple variables and SOH, the mathematical model of each other variable and the target variable is fitted to determine the second mapping relationship.
[0009] Optionally, construct a mathematical model of the target variable and SOH, and a mathematical model of each other variable and the target variable, including: Using quadratic or exponential functions, construct a mathematical model of the target variable fusion value and SOH, and a mathematical model of each other variable and the target variable.
[0010] This invention also provides a battery health state prediction device to address the technical problems of existing machine learning models failing to consider factors such as different users' charging and discharging habits and environmental temperature changes in actual applications, leading to a disconnect between the model's prediction results and the actual battery degradation path, and inaccurate prediction results due to performing SOH prediction from a single dimension. The device includes: The data acquisition module is used to acquire the first SOH value of the battery at the first moment, as well as historical operating data for a preset time period; the preset time period is the same historical time period as the time period from the first moment to the second moment. The first mapping module is used to map the first SOH value to the first target variable fusion value according to the preset first mapping relationship; wherein, the first mapping relationship is the mapping relationship between the target variable fusion value and the SOH value; the target variable fusion value refers to the fusion value obtained by determining a target variable from multiple variables affecting the battery SOH and mapping the other variables among the multiple variables to the target variable; The second mapping module is used to predict the second target variable fusion value at the second time step based on the first target variable fusion value, historical running data, and a preset second mapping relationship; the second mapping relationship includes the mapping relationship between each other variable and the target variable; The third mapping module is used to map the fused value of the second target variable to the second SOH value of the battery at the second time point according to the first mapping relationship.
[0011] This invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the above-described method for predicting battery health status.
[0012] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for predicting battery health status.
[0013] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-described method for predicting battery health status.
[0014] The battery health state prediction method and apparatus provided in this invention obtains a first SOH value of the battery at a first moment and historical operating data for a preset time period; the preset time period is the same historical time period as the time period from the first moment to the second moment; according to a preset first mapping relationship, the first SOH value is mapped to a first target variable fusion value; wherein, the first mapping relationship is the mapping relationship between the target variable fusion value and the SOH value; the target variable fusion value refers to the fusion value obtained by determining a target variable from multiple variables affecting the battery SOH and mapping the other variables among the multiple variables to the target variable; according to the first target variable fusion value, historical operating data, and a preset second mapping relationship, a second target variable fusion value is predicted at a second moment; the second mapping relationship includes the mapping relationship between each other variable and the target variable; according to the first mapping relationship, the second target variable fusion value is mapped to the second SOH value of the battery at the second moment. Compared with existing battery SOH prediction methods, this invention, based on the user's historical operating data, maps multiple variables affecting battery SOH to a single target variable. The SOH value is then predicted based on the fused value of the mapped target variable. This achieves SOH prediction by combining the user's historical operating data with multi-variable fusion prediction. This addresses the technical problems in existing technologies, such as failure to consider different users' charging and discharging habits and environmental temperature changes in actual applications, leading to a disconnect between the model prediction results and the actual battery degradation path, and inaccurate SOH predictions based on a single dimension. Attached Figure Description
[0015] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0016] In the attached diagram: Figure 1 A flowchart illustrating a method for predicting battery health status according to an embodiment of the present invention; Figure 2 A flowchart illustrating another method for predicting battery health status provided in an embodiment of the present invention; Figure 3 A schematic diagram illustrating the construction process of a first mapping relationship and a second mapping relationship provided in an embodiment of the present invention; Figure 4 This is a schematic diagram of the battery health state predicted by the battery health state prediction method provided in an embodiment of the present invention. Figure 5 A schematic diagram of a battery health status prediction device provided in an embodiment of the present invention; Figure 6 This is a schematic diagram of a computer device provided in an embodiment of the present invention. Detailed Implementation
[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.
[0018] In the description of this specification, the terms "comprising," "including," "having," and "containing" are open-ended terms, meaning that they include but are not limited to. The terms "an embodiment," "a specific embodiment," "some embodiments," and "for example," etc., refer to specific features, structures, or characteristics described in connection with that embodiment or example that are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, or characteristics described can be combined in any suitable manner in one or more embodiments or examples. The order of steps involved in the various embodiments is used to illustrate the implementation of this application, and the order of steps is not limited and can be adjusted appropriately as needed.
[0019] In the description of this specification, the terms "first" and "second," etc., are used to distinguish different objects or to distinguish different treatments of the same object, rather than to describe a specific order of objects.
[0020] In the description of this specification, "and / or" is merely a way of describing the relationship between related objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone.
[0021] Research has revealed that current technologies often use machine learning models to predict the state of harmonics (SOH) of batteries. However, these models do not consider factors such as different users' charging and discharging habits and environmental temperature changes in actual applications, leading to a disconnect between the model's predictions and the actual battery degradation path. Furthermore, existing machine learning models often predict SOH from a single dimension (such as the number of cycles), while battery SOH degradation is the result of the coupling effect of multiple factors, including electrochemical and thermodynamic factors. A single dimension cannot ensure prediction accuracy.
[0022] Based on this, embodiments of the present invention provide a method and apparatus for predicting battery health status. The method predicts the battery's state of health (SOH) based on factors such as the user's charging and discharging habits and changes in ambient temperature during actual use. Furthermore, it combines multiple factors affecting the battery's SOH to perform multi-dimensional information fusion prediction. This addresses the technical problems in the prior art, such as the failure to consider factors like different users' charging and discharging habits and changes in ambient temperature during actual use, which leads to a disconnect between the model prediction results and the actual battery degradation path, and the inaccurate prediction results due to SOH prediction from a single dimension.
[0023] Figure 1 This is a flowchart illustrating a method for predicting battery health status according to an embodiment of the present invention. The subject executing this method can be an electronic device, or a functional module or entity within an electronic device capable of implementing the battery health status prediction method. Such electronic devices include, but are not limited to, mobile terminals, tablet computers, computers, cameras, and wearable devices.
[0024] like Figure 1 As shown, methods for predicting battery health status may include: Step 101: Obtain the first SOH value of the battery at the first moment, and the historical operating data for a preset time period; the preset time period is the same historical time period as the time period from the first moment to the second moment. Step 102: Map the first SOH value to the first target variable fusion value according to the preset first mapping relationship; wherein, the first mapping relationship is the mapping relationship between the target variable fusion value and the SOH value; the target variable fusion value refers to the fusion value obtained by determining a target variable from multiple variables affecting the battery SOH and mapping the other variables among the multiple variables to the target variable; Step 103: Based on the fusion value of the first target variable, historical running data, and the preset second mapping relationship, predict the fusion value of the second target variable at the second time point; the second mapping relationship includes the mapping relationship between each other variable and the target variable; Step 104: Based on the first mapping relationship, map the fused value of the second target variable to the second SOH value of the battery at the second time.
[0025] In this embodiment of the invention, the first SOH value of the battery at a first moment and historical operating data for a preset time period are obtained; the preset time period is the same historical time period as the time period from the first moment to the second moment; according to a preset first mapping relationship, the first SOH value is mapped to a first target variable fusion value; wherein, the first mapping relationship is the mapping relationship between the target variable fusion value and the SOH value; the target variable fusion value refers to the fusion value obtained by determining a target variable from multiple variables affecting the battery SOH and mapping the other variables among the multiple variables to the target variable; according to the first target variable fusion value, historical operating data and a preset second mapping relationship, the second target variable fusion value at the second moment is predicted; the second mapping relationship includes the mapping relationship between each other variable and the target variable; according to the first mapping relationship, the second target variable fusion value is mapped to the second SOH value of the battery at the second moment. Compared with existing battery SOH prediction methods, this invention, based on the user's historical operating data, maps multiple variables affecting battery SOH to a single target variable. The SOH value is then predicted based on the fused value of the mapped target variable. This achieves SOH prediction by combining the user's historical operating data with multi-variable fusion prediction. This addresses the technical problems in existing technologies, such as failure to consider different users' charging and discharging habits and environmental temperature changes in actual applications, leading to a disconnect between the model prediction results and the actual battery degradation path, and inaccurate SOH predictions based on a single dimension.
[0026] The method for predicting the health status of the battery described above will be explained below.
[0027] In step 101 above, firstly, the first SOH value of the battery at the first moment and the historical operating data for a preset time period are obtained.
[0028] In specific implementation, the first moment can refer to the current moment, and the second moment can refer to a future moment. In this embodiment of the invention, the SOH value at a future moment is predicted. The second moment can be determined based on the prediction duration. For example, if the SOH of the battery is predicted 10 days later (prediction duration), then the second moment is the current moment plus the moment after 10 days.
[0029] The preset time period can be the same historical time period as the time period from the first moment to the second moment. For example, if the time period from the first moment to the second moment is from January 1st to 10th, then the preset time period can be from January 1st to 10th of last year.
[0030] Historical operating data can be the historical operating data of a user's energy storage system, which can be a residential energy storage system. The historical operating data of a residential energy storage system can reflect information such as changes in the ambient temperature of the user's battery and the user's charging and discharging habits. Specifically, historical operating data can include the battery's average temperature, charging capacity, discharging capacity, charge / discharge rate, and resting time within a preset time period.
[0031] In this embodiment of the invention, the first State of Health (SOH) value can be obtained through the Battery Management System (BMS) of the energy storage system. Historical operating data can be stored in the cloud or the local memory of the energy storage system. Therefore, the historical operating data of the user's energy storage system within a preset time period can be obtained from the cloud or the local memory of the energy storage system.
[0032] In step 102 above, the first SOH value can be mapped to the first target variable fusion value according to the preset first mapping relationship.
[0033] In specific implementation, the first mapping relationship is the mapping relationship between the target variable fusion value and the SOH value. The specific construction of the first mapping relationship will be explained in the following embodiment, and will not be elaborated here.
[0034] The target variable fusion value can refer to the fusion value obtained by determining a target variable from multiple variables that affect the battery's state of equilibrium (SOH) and mapping the other variables to the target variable.
[0035] In this embodiment of the invention, multiple variables affecting the battery's state of equilibrium (SOH) can be considered based on the battery's cycle aging and calendar aging factors. For example, multiple variables may include: high temperature cycle count, low temperature cycle count, high temperature storage time, and low temperature storage time. The distinction between high temperature and low temperature can be determined based on a preset temperature threshold. If the temperature is greater than the temperature threshold, it is considered high temperature; if the temperature is less than or equal to the temperature threshold, it is considered low temperature.
[0036] Taking the target variable as the number of high-temperature cycles as an example, the number of low-temperature cycles, high-temperature storage time, and low-temperature storage time of the battery can be mapped to the number of high-temperature cycles to obtain a fused value of the number of high-temperature cycles. Therefore, by constructing a mapping relationship between the fused value of the number of high-temperature cycles and the SOH value, the known first SOH value can be converted into a fused value of the first number of high-temperature cycles (i.e., the fused value of the first target variable).
[0037] In step 103 above, the second target variable fusion value at the second time point can be predicted based on the first target variable fusion value, historical running data, and the preset second mapping relationship.
[0038] In specific implementation, the second mapping relationship can include the mapping relationship between each other variable and the target variable. For example, if the target variable is the number of high-temperature cycles, the second mapping relationship can include the mapping relationship between the number of low-temperature cycles and the number of high-temperature cycles, the mapping relationship between high-temperature storage time and the number of high-temperature cycles, and the mapping relationship between low-temperature storage time and the number of high-temperature cycles. How the second mapping relationship is specifically constructed will be explained in the following embodiments, and will not be elaborated further here.
[0039] Based on historical operating data, the changes or usage of each variable of the battery from the first moment to the second moment can be determined. The second mapping relationship can be used to map the changes or usage of each variable to the target variable. In this way, based on the fusion value of the first target variable corresponding to the first moment, the changes or usage of each variable from the first moment to the second moment and the second mapping relationship, the fusion value of the second target variable at the second moment can be predicted.
[0040] In one embodiment, such as Figure 2 As shown, step 103 above may specifically include: Step 201: Based on historical operating data, determine the change of at least one variable between the first time point and the second time point; Step 202: According to the second mapping relationship, the change of each variable between the first time point and the second time point is fused with the fusion value of the first target variable to obtain the fusion value of the second target variable at the second time point.
[0041] In specific implementation, in step 201, the change of at least one variable between the first time and the second time can be determined based on historical operating data; the change of at least one variable between the first time and the second time can be the change of any other variable between the first time and the second time, and / or the change of the target variable between the first time and the second time.
[0042] For example, based on the charging capacity, discharging capacity, and pre-obtained rated battery capacity from historical operating data, the change in the number of cycles within a preset time period can be calculated. This change in the number of cycles within the preset time period determines the change in the number of high-temperature (or low-temperature) cycles from the first moment to the second moment. Similarly, based on the charge / discharge rate and the duration of the preset time period from historical operating data, the change in storage time within the preset time period can be calculated. Alternatively, if the historical operating data includes resting time, the resting time can be used as the change in storage time within the preset time period, determining the change in high-temperature (or low-temperature) storage time from the first moment to the second moment. Where the average battery temperature in the historical operating data is greater than a preset threshold, it represents the change in the number of high-temperature cycles and the change in high-temperature storage time; if the average battery temperature is less than or equal to the preset threshold, it represents the change in the number of low-temperature cycles and the change in low-temperature storage time.
[0043] In step 202 above, the change of each variable between the first time point and the second time point can be fused with the fusion value of the first target variable according to the mapping relationship between each other variable and the target variable, so as to obtain the fusion value of the second target variable at the second time point.
[0044] In this way, by using the above method, the changes of multiple variables from the first moment to the second moment are determined based on the user's historical operating data. This allows the fused value of the second target variable to incorporate information such as the user's charging and discharging habits and environment, thereby enabling personalized prediction based on the user's usage habits and environment, improving prediction accuracy, and enhancing the user experience.
[0045] In one embodiment, if the change of at least one variable between the first time point and the second time point includes the change of the target variable between the first time point and the second time point, the mapping can be achieved in step 202 above through the following steps: The change of the target variable between the first and second time points is combined with the fused value of the first target variable to obtain the intermediate value of the first target variable; For each of the other variables, repeat the following steps to update the intermediate value of the first target variable: Based on the mapping relationship between other variables and the target variable, the intermediate value of the first target variable is mapped to the fused value of the first other variables; The changes of other variables between the first and second time points are combined with the first other variable fusion value to obtain the second other variable fusion value at the second time point. Based on the mapping relationship between other variables and the target variable, the fused value of the second other variables is mapped to the intermediate value of the first target variable; The final updated intermediate value of the first objective variable is used as the fused value of the second objective variable.
[0046] In practice, if the change of the target variable between the first and second time points is included, the change of the target variable between the first and second time points can be merged with the first target variable's fused value first. For example, the change of the target variable between the first and second time points can be added to the first target variable's fused value to obtain the first target variable's intermediate value (here, the first target variable's intermediate value is the second target variable's fused value at the second time point after only merging the change of the target variable). Then, the changes of other variables between the first and second time points need to be merged with the first target variable's intermediate value in turn.
[0047] Specifically, for other variables: First, based on the mapping relationship between the other variable and the target variable, map the intermediate value of the first target variable to the fused value of the first other variable. Then, fuse the fused value of the first other variable with the change of the other variable between the first and second time points. For example, add the change of the other variable between the first and second time points to the fused value of the first other variable to obtain the fused value of the second other variable at the second time point. Then, convert the fused value of the second other variable at the second time point into the intermediate value of the first target variable. In this way, the fusion of the change of one other variable between the first and second time points is completed. The intermediate value of the first target variable here is the fused value of the second target variable after fusing the change of the target variable and the change of one other variable between the first and second time points. Then, for the latest intermediate value of the first target variable, continue to fuse the change of the next other variable between the first and second time points until the change of each other variable between the first and second time points is fused to the target variable. Then, the final intermediate value of the first target variable can be used as the fused value of the second target variable.
[0048] In one embodiment, if the change of at least one variable between the first and second time points does not include the change of the target variable between the first and second time points, the mapping can be achieved in step 202 above through the following steps: Use the fused value of the first objective variable as the intermediate value of the second objective variable; For each of the other variables, repeat the following steps to update the intermediate value of the second target variable: Based on the mapping relationship between other variables and the target variable, the intermediate value of the second target variable is mapped to the fused value of the third set of other variables; The changes of other variables between the first and second time points are combined with the third other variable fusion value to obtain the fourth other variable fusion value at the second time point; Based on the mapping relationship between other variables and the target variable, the fused value of the fourth other variable is mapped to the intermediate value of the second target variable; The final updated intermediate value of the second objective variable is used as the fused value of the second objective variable.
[0049] In practice, if the change in the target variable between the first and second time points is not included, the fused value of the first target variable can be used as the intermediate value of the second target variable for updating.
[0050] Specifically, for other variables: First, based on the mapping relationship between the other variable and the target variable, map the intermediate value of the second target variable to the fused value of the third other variable. Then, fuse the fused value of the third other variable with the change of the other variable between the first and second time points. For example, add the change of the other variable between the first and second time points to the fused value of the third other variable to obtain the fused value of the fourth other variable at the second time point. Then, convert the fused value of the fourth other variable at the second time point into the intermediate value of the second target variable. In this way, the fusion of the change of one other variable between the first and second time points is completed. The intermediate value of the second target variable here is the fused value of the second target variable after incorporating the change of one other variable between the first and second time points. Then, for the latest intermediate value of the second target variable, continue to fuse the change of the next other variable between the first and second time points until the change of each other variable between the first and second time points is fused to the target variable. Then, the final intermediate value of the second target variable can be used as the fused value of the second target variable.
[0051] In step 104 above, the fused value of the second target variable can be transformed into the second SOH value at the second time step according to the first mapping relationship.
[0052] Thus, multiple variables can affect the battery's State of Health (SOH). The battery health prediction method provided in this invention maps multiple variables to a single variable, and then calculates the SOH based on the fused variable. This avoids calculating an SOH for each variable, which would lead to redundancy and conflicts in the output. Furthermore, it enables multi-dimensional information fusion to predict the battery's SOH, improving the prediction accuracy. In addition, the historical operating data of the user's energy storage system reflects the user's charging and discharging habits, ambient temperature changes, etc., in actual applications. Combining the historical operating data of the user's energy storage system to predict the battery's SOH allows for personalized predictions for the user, further improving the prediction accuracy of the battery's SOH.
[0053] The construction process of the first and second mapping relationships is explained below.
[0054] Figure 3This is a schematic diagram illustrating the construction process of the first and second mapping relationships provided in an embodiment of the present invention. Figure 3 As shown, it may include the following steps: Step 301: Obtain the correlation data between multiple variables and SOH, and construct a mathematical model of the target variable fusion value and SOH, and a mathematical model of each other variable and the target variable; Step 302: Based on the correlation data between multiple variables and SOH, fit the mathematical model of the target variable fusion value and SOH to determine the first mapping relationship; and based on the correlation data between multiple variables and SOH, fit the mathematical model of each other variable and the target variable to determine the second mapping relationship.
[0055] In specific implementation, in step 301 above, the correlation data between each variable and SOH can be obtained. For example, battery aging tests of high temperature / low temperature cycle number and high temperature / low temperature storage time can be carried out to obtain the relationship curves between high temperature cycle number and SOH, low temperature cycle number and SOH, high temperature storage time and SOH, and low temperature storage time and SOH.
[0056] In step 301 above, a mathematical model of the target variable fusion value and SOH, and a mathematical model of each other variable and the target variable are also constructed. The mathematical model of the target variable fusion value and SOH, and the mathematical model of each other variable and the target variable can be constructed using a quadratic function or an exponential function.
[0057] For example, a quadratic function in one variable is shown below: The exponential function is shown below: In step 302 above, the mathematical model of the target variable fusion value and SOH can be fitted based on the correlation data between multiple variables and SOH to determine the first mapping relationship; and the mathematical model of each other variable and the target variable can be fitted based on the correlation data between multiple variables and SOH to determine the second mapping relationship.
[0058] In practical implementation, it is assumed that the fused value of the target variable, constructed based on a quadratic function, is related to the mathematical model of SOH. x When expressed as the fused value of the target variable, Expressed as SOH value; x When expressed as SOH value, This is represented as the fused value of the target variable; P1, P2, and P3 are the coefficients obtained from the fitting. In the mathematical model of other variables and the target variable constructed based on the quadratic function, x When represented as other variables, Represented as the target variable; x When expressed as the target variable, Other variables are represented by P1, P2, and P3, which are the coefficients obtained from the fitting.
[0059] Assuming that in the mathematical model of the fusion value of the target variable constructed based on the exponential function and the SOH, x When expressed as the fused value of the target variable, Expressed as SOH value; x When expressed as SOH value, This is represented as the fused value of the target variable; a, b, c, and d are the coefficients obtained from the fit. In the mathematical model of other variables and the target variable constructed based on the exponential function, x When represented as other variables, Represented as the target variable; x When expressed as the target variable, Other variables are represented as 'a', 'b', 'c', and 'd', which are the coefficients obtained from the fitting.
[0060] The specific fitting process can be found in existing fitting algorithms, and will not be elaborated further here.
[0061] It should be noted that the variables affecting the battery SOH in the embodiments of the present invention include, but are not limited to, the number of high-temperature cycles, the number of low-temperature cycles, the high-temperature storage time, and the low-temperature storage time, and may also include the battery state of charge (SOC) value when at rest, the depth of charge and discharge, etc.
[0062] It should also be noted that the mathematical model for the fusion value of the target variable and SOH, and the mathematical model for each other variable and the target variable in the embodiments of the present invention, are not limited to quadratic functions or exponential functions, but can also be other functions, such as power functions, logarithmic functions, etc. In practical applications, the appropriate function can be selected according to the requirements.
[0063] To better understand the battery health status prediction method provided in the embodiments of the present invention, a specific example is given below.
[0064] In this example, multiple variables may include: number of high-temperature cycles, number of low-temperature cycles, high-temperature storage time, and low-temperature storage time; with the number of high-temperature cycles as the target variable, a mapping relationship (first mapping relationship) between the fused value of the number of high-temperature cycles and the SOH value is pre-constructed; and mapping relationships (second mapping relationships) between the number of low-temperature cycles and the number of high-temperature cycles, the high-temperature storage time and the number of high-temperature cycles, and the low-temperature storage time and the number of high-temperature cycles are pre-constructed.
[0065] The SOH value at time k (the first SOH value) is known, i.e. ; and to acquire historical operating data for a preset time period, namely the battery average temperature T and charging capacity within the preset time period. Discharge capacity 1. Charge / discharge rate (m). Next, predict the battery health status as follows.
[0066] (1) Calculate the number of high-temperature cycles at time k according to Formula 1 (i.e., the first mapping relationship). (First target variable fusion value): Formula 1 Where A1, A2, and A3 are fitting coefficients.
[0067] If the average battery temperature T within the preset time period is less than or equal to the preset temperature threshold, it is considered low temperature. This means that the period from time k to time k+1 (the second time) is considered low temperature. Steps (1) to (9) are then executed to obtain the fusion value of the number of high-temperature cycles at time k+1. (The fusion value of the second target variable).
[0068] (2) According to Formula 2 (i.e., the mapping relationship between the number of low-temperature cycles and the number of high-temperature cycles in the second mapping relationship), (i.e., the intermediate value 1 of the second objective variable) is transformed into the number of low-temperature cycles at time k. (i.e., the third other variable fusion value 1): Formula 2 Among them, B1, B2, and B3 are fitting coefficients.
[0069] (3) Based on the charging capacity within the preset time period Discharge capacity The pre-obtained rated capacity of the battery Using Formula 3, calculate the change in the number of low-temperature cycles from time k to time k+1. : Formula 3 (4) According to Formula 4, and Add them together to get the number of low-temperature cycles at time k+1. (i.e., the fourth other variable fusion value 1): Formula 4 (5) According to Formula 5, Transformed into fusion High temperature cycle number fusion value (Second target variable intermediate value 2): Formula 5 In Formula 5, B1, B2, and B3 are the same as B1, B2, and B3 in Formula 2. Both Formula 5 and Formula 2 are mapping relationships between the number of low-temperature cycles and the number of high-temperature cycles in the second mapping relationship, only the representation is different.
[0070] (6) According to Formula 6 (i.e., the mapping relationship between low-temperature storage time and high-temperature cycle number in the second mapping relationship), Converted to cryogenic storage time at time k (i.e., the third other variable fusion value 2): Formula 6 Where C1, C2, and C3 are fitting coefficients.
[0071] (7) Based on the change in charge / discharge rate m and the number of low-temperature cycles within the preset time period. Given the known duration n of the preset time period and Formula 7, calculate the change in cryogenic storage time from time k to time k+1. : Formula 7 (8) According to Formula 8, and Adding them together, we get the cryogenic storage time at time k+1. (i.e., the fourth other variable fusion value 2): Formula 8 (9) According to Formula 9, Transformed into fusion and High temperature cycle number fusion value (That is, the intermediate value 2 of the second objective variable. Since there is no other variable, the intermediate value 2 of the second objective variable here is the final fused value of the number of high-temperature cycles at time k+1, that is, the fused value of the second objective variable): Formula 9 In Formula 9, C1, C2, and C3 are the same as C1, C2, and C3 in Formula 6. Both Formula 9 and Formula 6 are mapping relationships between low-temperature storage time and high-temperature cycle number in the second mapping relationship, only the representation is different.
[0072] If the average battery temperature T within the preset time period is greater than the preset temperature threshold, it is considered high temperature, meaning that the period from time k to time k+1 is considered high temperature. Steps (10) to (15) are then executed to obtain the fused value of the number of high temperature cycles at time k+1. (The fusion value of the second target variable).
[0073] (10) Based on the charging capacity within the preset time period Discharge capacity The pre-obtained rated capacity of the battery Using Formula 10, calculate the change in the number of high-temperature cycles from time k to time k+1. :, Formula 10 (11) According to formula 11, and Adding them together yields only the fused ones. Number of high-temperature cycles at time k+1 (i.e., the intermediate value 1 of the first target variable): Formula 11 (12) According to Formula 12 (i.e., the mapping relationship between high-temperature storage time and high-temperature cycle number in the second mapping relationship), Converted to high-temperature storage time at time k (i.e., the first other variable fusion value 1): Formula 12 Where D1, D2, and D3 are fitting coefficients.
[0074] (13) Based on the change in charge / discharge rate m and the number of high-temperature cycles within the preset time period. Given the known duration n of the preset time period and Formula 13, calculate the change in high-temperature storage time from time k to time k+1. : Formula 13 (14) According to formula 14, and Adding them together, we get the high-temperature storage time at time k+1. (i.e., the second other variable fusion value 1): Formula 14 (15) According to Formula 15, Transformed into fusion and High temperature cycle number fusion value (That is, the intermediate value 2 of the first objective variable. Since there is no other variable, the intermediate value 2 of the first objective variable here is the final fused value of the number of high-temperature cycles at time k+1, which is the fused value of the second objective variable): Formula 15 In Formula 15, D1, D2, and D3 are the same as those in Formula 12. Both Formula 15 and Formula 12 are mapping relationships between high-temperature storage time and high-temperature cycle count in the second mapping relationship, only the representation is different.
[0075] Based on the two scenarios of high and low temperatures described above, if the temperature is low between time k and time k+1 (the second time), then the change in at least one variable between the first and second time times includes the change in low-temperature storage time and the change in the number of low-temperature cycles, but not the change in the target variable (the number of high-temperature cycles). Therefore, the value of the number of high-temperature cycles at time k can be merged first. The result is converted into the number of low-temperature cycles, then the change in the number of low-temperature cycles is added, and finally converted back into the fused value of the number of high-temperature cycles. The latest fused value of the number of high-temperature cycles is then converted into the low-temperature storage time, and the change in the low-temperature storage time is added, and finally converted back into the fused value of the number of high-temperature cycles, thus obtaining the fused value of the number of high-temperature cycles at time k+1. .
[0076] If the temperature is high between time k and time k+1 (the second time), then the change in at least one variable between the first and second time includes: the change in high-temperature storage time and the change in the number of high-temperature cycles, including the change in the target variable (number of high-temperature cycles). Therefore, the number of high-temperature cycles at time k can be merged first. After adding the change in the number of high-temperature cycles, it is converted into high-temperature storage time; after adding the change in high-temperature storage time to the high-temperature storage time, it is converted into a fused value of the number of high-temperature cycles, which is the fused value of the number of high-temperature cycles at time k+1. .
[0077] The fusion value of the number of high-temperature cycles at time k+1 is obtained. Then, the SOH value at time k+1 can be calculated according to Formula 16: Formula 16 In Formula 16, A1, A2, and A3 are the same as A1, A2, and A3 in Formula 1. Both Formula 16 and Formula 1 are first mapping relationships, only differing in their representation.
[0078] In this way, by using the above method, the change of at least one variable between the first time point and the second time point can be fused into the fused value of the first target variable to obtain the fused value of the second target variable at the second time point. Then, based on the fused value of the second target variable at the second time point, the SOH value at the second time point can be obtained.
[0079] like Figure 4 The image shown is a schematic diagram of the battery health state predicted based on the aforementioned battery health state prediction method. Figure 4 The prediction cycle is 15 days, meaning the interval between the first and second points is 15 days. The initial health status is 94%, and the prediction is made for the health status 10 years later. This results in a low-temperature cycle for the first 90 months and a high-temperature cycle for the following 30 months. Figure 4 The three curves in the diagram represent changes in health status when the number of daily cycles is 1, 1.5, and 2.
[0080] This invention also provides a battery health state prediction device, as described in the following embodiments. Since the principle by which this device solves the problem is similar to the battery health state prediction method described above, the implementation of this device can refer to the implementation of the battery health state prediction method, and repeated details will not be elaborated further.
[0081] like Figure 5 The diagram shown is a schematic of a battery health status prediction device provided in an embodiment of the present invention. The device may include the following modules: The data acquisition module 501 is used to acquire the first SOH value of the battery at the first moment, as well as the historical operating data for a preset time period; the preset time period is the same historical time period as the time period from the first moment to the second moment. The first mapping module 502 is used to map the first SOH value to the first target variable fusion value according to the preset first mapping relationship; wherein, the first mapping relationship is the mapping relationship between the target variable fusion value and the SOH value; the target variable fusion value refers to the fusion value obtained by determining a target variable from multiple variables affecting the battery SOH and mapping the other variables among the multiple variables to the target variable; The second mapping module 503 is used to predict the second target variable fusion value at a second time step based on the first target variable fusion value, historical running data, and a preset second mapping relationship; the second mapping relationship includes the mapping relationship between each other variable and the target variable; The third mapping module 504 is used to map the fused value of the second target variable to the second SOH value of the battery at the second time point according to the first mapping relationship.
[0082] In one embodiment, the second mapping module 503 can be specifically used for: Based on historical operational data, determine the change of at least one variable between the first and second time points; According to the second mapping relationship, the change of each variable between the first time point and the second time point is fused with the fusion value of the first target variable to obtain the fusion value of the second target variable at the second time point.
[0083] The second mapping module 503 can also be used for: The change of the target variable between the first and second time points is combined with the fused value of the first target variable to obtain the intermediate value of the first target variable; For each of the other variables, repeat the following steps to update the intermediate value of the first target variable: Based on the mapping relationship between other variables and the target variable, the intermediate value of the first target variable is mapped to the fused value of the first other variables; The changes of other variables between the first and second time points are combined with the first other variable fusion value to obtain the second other variable fusion value at the second time point. Based on the mapping relationship between other variables and the target variable, the fused value of the second other variables is mapped to the intermediate value of the first target variable; The final updated intermediate value of the first objective variable is used as the fused value of the second objective variable.
[0084] In one embodiment, the second mapping module 503 can also be used for: Use the fused value of the first objective variable as the intermediate value of the second objective variable; For each of the other variables, repeat the following steps to update the intermediate value of the second target variable: Based on the mapping relationship between other variables and the target variable, the intermediate value of the second target variable is mapped to the fused value of the third set of other variables; The changes of other variables between the first and second time points are combined with the third other variable fusion value to obtain the fourth other variable fusion value at the second time point; Based on the mapping relationship between other variables and the target variable, the fused value of the fourth other variable is mapped to the intermediate value of the second target variable; The final updated intermediate value of the second objective variable is used as the fused value of the second objective variable.
[0085] In one embodiment, a relationship building module is further included, used to determine the first mapping relationship and the second mapping relationship through the following steps: Obtain the correlation data between multiple variables and SOH, and construct a mathematical model of the target variable fusion value and SOH, and a mathematical model of each other variable and the target variable; Based on the correlation data between multiple variables and SOH, the mathematical model of the target variable fusion value and SOH is fitted to determine the first mapping relationship; and based on the correlation data between multiple variables and SOH, the mathematical model of each other variable and the target variable is fitted to determine the second mapping relationship.
[0086] In one embodiment, the relationship building module can also be used for: Using quadratic or exponential functions, construct a mathematical model of the target variable fusion value and SOH, and a mathematical model of each other variable and the target variable.
[0087] This invention also provides a computer device. Figure 6 This is a schematic diagram of a computer device in an embodiment of the present invention. The computer device 600 includes a memory 610, a processor 620, and a computer program 630 stored in the memory 610 and executable on the processor 620. When the processor 620 executes the computer program 630, it implements the above-mentioned method for predicting battery health status.
[0088] This invention also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for predicting battery health status.
[0089] This invention also provides a computer program product, which includes a computer program that, when executed by a processor, implements the above-mentioned method for predicting battery health status.
[0090] The battery health state prediction method and apparatus provided in this invention obtains a first SOH value of the battery at a first moment and historical operating data for a preset time period; the preset time period is the same historical time period as the time period from the first moment to the second moment; according to a preset first mapping relationship, the first SOH value is mapped to a first target variable fusion value; wherein, the first mapping relationship is the mapping relationship between the target variable fusion value and the SOH value; the target variable fusion value refers to the fusion value obtained by determining a target variable from multiple variables affecting the battery SOH and mapping the other variables among the multiple variables to the target variable; according to the first target variable fusion value, historical operating data, and a preset second mapping relationship, a second target variable fusion value is predicted at a second moment; the second mapping relationship includes the mapping relationship between each other variable and the target variable; according to the first mapping relationship, the second target variable fusion value is mapped to the second SOH value of the battery at the second moment. Compared with existing battery SOH prediction methods, this invention, based on the user's historical operating data, maps multiple variables affecting battery SOH to a single target variable. The SOH value is then predicted based on the fused value of the mapped target variable. This achieves SOH prediction by combining the user's historical operating data with multi-variable fusion prediction. This addresses the technical problems in existing technologies, such as failure to consider different users' charging and discharging habits and environmental temperature changes in actual applications, leading to a disconnect between the model prediction results and the actual battery degradation path, and inaccurate SOH predictions based on a single dimension.
[0091] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0092] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0093] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0094] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0095] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. A method for predicting battery health status, characterized in that, include: Obtain the first SOH value of the battery at the first moment, as well as the historical operating data for a preset time period; the preset time period is the same historical time period as the time period from the first moment to the second moment. According to the preset first mapping relationship, the first SOH value is mapped to the first target variable fusion value; wherein, the first mapping relationship is the mapping relationship between the target variable fusion value and the SOH value; the target variable fusion value refers to the fusion value obtained by determining a target variable from multiple variables affecting the battery SOH and mapping the other variables among the multiple variables to the target variable; Based on the fusion value of the first target variable, historical operating data, and a preset second mapping relationship, predict the fusion value of the second target variable at the second time point; the second mapping relationship includes the mapping relationship between each other variable and the target variable. Based on the first mapping relationship, the fused value of the second target variable is mapped to the second SOH value of the battery at the second time.
2. The method as described in claim 1, characterized in that, Based on the fused value of the first objective variable, historical operating data, and a preset second mapping relationship, predict the fused value of the second objective variable at the second time point, including: Based on historical operational data, determine the change of at least one variable between the first and second time points; According to the second mapping relationship, the change of each variable between the first time point and the second time point is fused with the fusion value of the first target variable to obtain the fusion value of the second target variable at the second time point.
3. The method as described in claim 2, characterized in that, Based on the second mapping relationship, the change of each variable between the first and second time points is fused with the fusion value of the first target variable to obtain the fusion value of the second target variable at the second time point, including: The change of the target variable between the first and second time points is combined with the fused value of the first target variable to obtain the intermediate value of the first target variable; For each of the other variables, repeat the following steps to update the intermediate value of the first target variable: Based on the mapping relationship between other variables and the target variable, the intermediate value of the first target variable is mapped to the fused value of the first other variables; The changes of other variables between the first and second time points are combined with the first other variable fusion value to obtain the second other variable fusion value at the second time point. Based on the mapping relationship between other variables and the target variable, the fused value of the second other variables is mapped to the intermediate value of the first target variable; The final updated intermediate value of the first objective variable is used as the fused value of the second objective variable.
4. The method as described in claim 2, characterized in that, Based on the second mapping relationship, the change of each variable between the first and second time points is fused with the fusion value of the first target variable to obtain the fusion value of the second target variable at the second time point, including: Use the fused value of the first objective variable as the intermediate value of the second objective variable; For each of the other variables, repeat the following steps to update the intermediate value of the second target variable: Based on the mapping relationship between other variables and the target variable, the intermediate value of the second target variable is mapped to the fused value of the third set of other variables; The changes of other variables between the first and second time points are combined with the third other variable fusion value to obtain the fourth other variable fusion value at the second time point; Based on the mapping relationship between other variables and the target variable, the fused value of the fourth other variable is mapped to the intermediate value of the second target variable; The final updated intermediate value of the second objective variable is used as the fused value of the second objective variable.
5. The method according to any one of claims 1-4, characterized in that, It also includes determining the first mapping relationship and the second mapping relationship through the following steps: Obtain the correlation data between multiple variables and SOH, and construct a mathematical model of the target variable fusion value and SOH, and a mathematical model of each other variable and the target variable; Based on the correlation data between multiple variables and SOH, the mathematical model of the target variable fusion value and SOH is fitted to determine the first mapping relationship; Furthermore, based on the correlation data between multiple variables and SOH, a mathematical model is fitted to the target variable for each other variable to determine the second mapping relationship.
6. The method as described in claim 5, characterized in that, Construct a mathematical model of the target variable and SOH, and a mathematical model of each other variable and the target variable, including: Using quadratic or exponential functions, construct a mathematical model of the target variable fusion value and SOH, and a mathematical model of each other variable and the target variable.
7. A device for predicting battery health status, characterized in that, include: The data acquisition module is used to acquire the first SOH value of the battery at the first moment, as well as historical operating data for a preset time period; the preset time period is the same historical time period as the time period from the first moment to the second moment. The first mapping module is used to map the first SOH value to the first target variable fusion value according to the preset first mapping relationship; wherein, the first mapping relationship is the mapping relationship between the target variable fusion value and the SOH value; the target variable fusion value refers to the fusion value obtained by determining a target variable from multiple variables affecting the battery SOH and mapping the other variables among the multiple variables to the target variable; The second mapping module is used to predict the second target variable fusion value at the second time step based on the first target variable fusion value, historical running data, and a preset second mapping relationship; the second mapping relationship includes the mapping relationship between each other variable and the target variable; The third mapping module is used to map the fused value of the second target variable to the second SOH value of the battery at the second time point according to the first mapping relationship.
8. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the battery health state 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 a computer program that, when executed by a processor, implements the method for predicting battery health status as described in any one of claims 1 to 6.
10. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the battery health state prediction method according to any one of claims 1 to 6.