Intelligent prediction method and device for battery health state

Abstract

The invention discloses an intelligent prediction method and device for the state of health of a battery, and the method comprises the steps: constructing an initial analysis model containing three key parameters, namely, the SOH (state of health), the operation mileage and the operation age limit of a battery pack, so that compared with a prediction model of a single factor, multi-dimensional factors influencing the performance degradation of the battery can be more comprehensively considered, and the prediction accuracy is improved. The prediction accuracy of the future use state of the battery pack can be improved, the trend of battery performance reduction can be identified in advance, and a scientific basis is provided for battery maintenance and management; besides, the built target analysis model can realize intelligent prediction of the health state of the battery pack, and the prediction capability enables a battery management system to dynamically adjust a maintenance strategy according to real-time data / prediction data, so that the service efficiency of the battery is optimized, and the service life of the battery is prolonged.

Description

Technical Field

[0001] The present invention relates to the technical field of battery data processing, and in particular to a method and device for intelligently predicting the health status of a battery. Background Art

[0002] Currently, lithium-ion batteries are widely used in automobiles, electronic mobile devices and other fields. In actual use, as the battery is recycled, irreversible physical or chemical changes will occur inside the battery, causing the battery to age. As an important topic in battery system management, the health status (SOH) of lithium-ion batteries is an important research object.

[0003] The existing battery SOH prediction methods have the following shortcomings: First, many methods only rely on a single battery pack SOH data or limited operating parameters for prediction, ignoring the complex relationship between battery performance degradation and multiple factors, resulting in inaccurate prediction results; Second, although some methods take into account multiple influencing factors, their use of multiple image factors is relatively fixed and single, and the data utilization rate is not high, which reduces the practicality and reliability of the prediction. Summary of the invention

[0004] The present invention provides a method and device for intelligently predicting the health status of a battery, which can broaden the data analysis dimension of the battery pack SOH and improve the prediction accuracy, reliability and practicality of the battery pack SOH.

[0005] In order to solve the above technical problems, the first aspect of the present invention discloses an intelligent prediction method for battery health status, the method comprising:

[0006] Acquire battery pack data to be analyzed, the battery pack data including battery pack SOH data, operating mileage corresponding to the battery pack SOH data, and operating years corresponding to the battery pack SOH data;

[0007] Constructing an initial analysis model corresponding to the battery pack data, the initial analysis model comprising a first analysis model and a second analysis model, the first analysis model being an analysis model corresponding to the battery pack SOH data and the operating mileage; the second analysis model being an analysis model corresponding to the battery pack SOH data and the operating years;

[0008] generating future battery pack data corresponding to the battery pack data according to the battery pack data and the initial analysis model, and generating a target data set according to the battery pack data and the future battery pack data;

[0009] According to the target data set, a target analysis model corresponding to the initial analysis model is generated, and the target analysis model is a comprehensive analysis model for the three parameters of new battery pack SOH data, new operating mileage and new operating years in the target data set; and the target analysis model is used to determine the SOH prediction data of any target battery pack.

[0010] As an optional implementation manner, in the first aspect of the present invention, the constructing an initial analysis model corresponding to the battery pack data includes:

[0011] Taking the battery pack SOH data and the operating mileage as first benchmark data, constructing a first linear model corresponding to the first benchmark data;

[0012] Taking the battery pack SOH data and the operating years as second benchmark data, constructing a second linear model corresponding to the second benchmark data;

[0013] Substituting the first benchmark data into the first linear model to obtain a first slope and a first intercept corresponding to the first linear model; and substituting the second benchmark data into the second linear model to obtain a second slope and a second intercept corresponding to the second linear model;

[0014] The first linear model is updated to a first analysis model according to the first slope and the first intercept; and the second linear model is updated to a second analysis model according to the second slope and the second intercept, and the first analysis model and the second analysis model are determined as initial analysis models.

[0015] As an optional implementation, in the first aspect of the present invention, generating future battery pack data corresponding to the battery pack data according to the battery pack data and the initial analysis model includes:

[0016] determining a numerical filter interval for performing data filtering;

[0017] Determining target screening data matching the numerical screening interval from the battery pack SOH data corresponding to the battery pack data;

[0018] Inputting the target screening data into the first analysis model and the second analysis model to obtain the future operating mileage corresponding to the first analysis model and the future operating years corresponding to the second analysis model;

[0019] Determine the future operating mileage and the corresponding target screening data, the future operating years and the corresponding target screening data as future battery pack data corresponding to the battery pack data;

[0020] And, generating a target data set according to the battery pack data and the future battery pack data includes:

[0021] The battery pack data and the future battery pack data are performed data splicing to obtain a target data set.

[0022] As an optional implementation manner, in the first aspect of the present invention, generating a target analysis model corresponding to the initial analysis model according to the target data set includes:

[0023] Taking the new battery pack SOH data, new operating mileage and new operating years in the target data set as the third benchmark data, a third linear model corresponding to the third benchmark data is constructed, and the third linear model is a multivariate linear regression model; the third linear model includes a first regression coefficient, a second regression coefficient and a third intercept;

[0024] generating a coefficient vector according to the first regression coefficient, the second regression coefficient and the third intercept;

[0025] Generate a target matrix according to the new operating mileage and the new operating years;

[0026] According to the target matrix and the new battery pack SOH data, a preset data calculation is performed in combination with a least squares method to obtain a solution value corresponding to the coefficient vector;

[0027] Update the first regression coefficient, the second regression coefficient and the third intercept according to the solution value corresponding to the coefficient vector;

[0028] The third linear model is updated to a target analysis model according to the first regression coefficient, the second regression coefficient and the third intercept.

[0029] As an optional embodiment, in the first aspect of the present invention, the method further comprises:

[0030] When a use demand for the target analysis model is detected, a parameter to be predicted is determined according to the use demand, and the parameter to be predicted includes an operating mileage to be predicted and an operating life to be predicted;

[0031] Generate a target data combination according to the to-be-predicted operating mileage and the to-be-predicted operating years, wherein the target data combination includes a plurality of data pairs, each of which is composed of a mileage value of the to-be-predicted operating mileage and a year value of the to-be-predicted operating years;

[0032] Combining the target data and outputting the target analysis model to obtain target SOH prediction data;

[0033] The target SOH prediction data is SOH change information associated with the battery pack and the parameter to be predicted.

[0034] As an optional implementation, in the first aspect of the present invention, the model formula corresponding to the first linear model is specifically:

[0035] Soh mile =k1*mile1+b1

[0036] Among them, Soh mile is the first linear model; mile1 is the operating mileage corresponding to the battery pack SOH data; k1 is the first slope; b1 is the first intercept;

[0037] The model formula corresponding to the second linear model is specifically:

[0038] Soh days =k2*days1+b2

[0039] Among them, Soh days is the second linear model; days1 is the operating life corresponding to the battery pack SOH data, and the operating life is in days; k2 is the second slope; b2 is the second intercept.

[0040] As an optional implementation, in the first aspect of the present invention, the model formula corresponding to the third linear model is specifically:

[0041] Soh=k3*mile2+k4*days2+b

[0042] Wherein, Soh is the third linear model; mile2 is the new operating mileage; days2 is the new operating years; k3 is the first regression coefficient; k4 is the second regression coefficient; b is the third intercept;

[0043] The calculation formula used for performing preset data calculation based on the target matrix and the new battery pack SOH data in combination with the least squares method to obtain the solution value corresponding to the coefficient vector is specifically:

[0044] X T Xβ=X T y

[0045] Wherein, X is the target matrix; y is the new battery pack SOH data; β is the coefficient vector;

[0046] Then we can get the variant:

[0047] β=(X T X) -1 X Ty

[0048] The β calculated by this variant is used to reversely determine the first regression coefficient, the second regression coefficient and the third intercept.

[0049] A second aspect of the present invention discloses an intelligent prediction device for a battery health state, the device comprising:

[0050] A data acquisition module, used to acquire battery pack data to be analyzed, wherein the battery pack data includes battery pack SOH data, operating mileage corresponding to the battery pack SOH data, and operating years corresponding to the battery pack SOH data;

[0051] A model building module, used to build an initial analysis model corresponding to the battery pack data, the initial analysis model comprising a first analysis model and a second analysis model, the first analysis model being an analysis model corresponding to the battery pack SOH data and the operating mileage; the second analysis model being an analysis model corresponding to the battery pack SOH data and the operating years;

[0052] a data generation module, configured to generate future battery pack data corresponding to the battery pack data according to the battery pack data and the initial analysis model, and to generate a target data set according to the battery pack data and the future battery pack data;

[0053] A model generation module is used to generate a target analysis model corresponding to the initial analysis model based on the target data set, wherein the target analysis model is a comprehensive analysis model for three parameters in the target data set: new battery pack SOH data, new operating mileage, and new operating years; and the target analysis model is used to determine the SOH prediction data of any target battery pack.

[0054] As an optional implementation, in the second aspect of the present invention, the model building module builds an initial analysis model corresponding to the battery pack data in a manner that specifically includes:

[0055] Taking the battery pack SOH data and the operating mileage as first benchmark data, constructing a first linear model corresponding to the first benchmark data;

[0056] Taking the battery pack SOH data and the operating years as second benchmark data, constructing a second linear model corresponding to the second benchmark data;

[0057] Substituting the first benchmark data into the first linear model to obtain a first slope and a first intercept corresponding to the first linear model; and substituting the second benchmark data into the second linear model to obtain a second slope and a second intercept corresponding to the second linear model;

[0058] The first linear model is updated to a first analysis model according to the first slope and the first intercept; and the second linear model is updated to a second analysis model according to the second slope and the second intercept, and the first analysis model and the second analysis model are determined as initial analysis models.

[0059] As an optional implementation, in the second aspect of the present invention, the data generation module generates future battery pack data corresponding to the battery pack data according to the battery pack data and the initial analysis model, specifically including:

[0060] determining a numerical filter interval for performing data filtering;

[0061] Determining target screening data matching the numerical screening interval from the battery pack SOH data corresponding to the battery pack data;

[0062] Inputting the target screening data into the first analysis model and the second analysis model to obtain the future operating mileage corresponding to the first analysis model and the future operating years corresponding to the second analysis model;

[0063] Determine the future operating mileage and the corresponding target screening data, the future operating years and the corresponding target screening data as future battery pack data corresponding to the battery pack data;

[0064] And, the data generation module generates a target data set according to the battery pack data and the future battery pack data in a manner that specifically includes:

[0065] The battery pack data and the future battery pack data are performed data splicing to obtain a target data set.

[0066] As an optional implementation, in the second aspect of the present invention, the model generation module generates a target analysis model corresponding to the initial analysis model according to the target data set, specifically comprising:

[0067] Taking the new battery pack SOH data, new operating mileage and new operating years in the target data set as the third benchmark data, a third linear model corresponding to the third benchmark data is constructed, and the third linear model is a multivariate linear regression model; the third linear model includes a first regression coefficient, a second regression coefficient and a third intercept;

[0068] generating a coefficient vector according to the first regression coefficient, the second regression coefficient and the third intercept;

[0069] Generate a target matrix according to the new operating mileage and the new operating years;

[0070] According to the target matrix and the new battery pack SOH data, a preset data calculation is performed in combination with a least squares method to obtain a solution value corresponding to the coefficient vector;

[0071] Update the first regression coefficient, the second regression coefficient and the third intercept according to the solution value corresponding to the coefficient vector;

[0072] The third linear model is updated to a target analysis model according to the first regression coefficient, the second regression coefficient and the third intercept.

[0073] As an optional implementation, in the second aspect of the present invention, the device further includes:

[0074] A determination module, configured to determine, when a use demand for the target analysis model is detected, parameters to be predicted according to the use demand, the parameters to be predicted including operating mileage to be predicted and operating years to be predicted;

[0075] The data generation module is further used to generate a target data combination according to the to-be-predicted operating mileage and the to-be-predicted operating years, wherein the target data combination includes a plurality of data pairs, each of which is composed of a mileage value of the to-be-predicted operating mileage and a year value of the to-be-predicted operating years;

[0076] A prediction module, used for combining the target data and outputting the target analysis model to obtain target SOH prediction data;

[0077] The target SOH prediction data is SOH change information associated with the battery pack and the parameter to be predicted.

[0078] As an optional implementation, in the second aspect of the present invention, the model formula corresponding to the first linear model is specifically:

[0079] Soh mile =k1*mile1+b1

[0080] Among them, Soh mile is the first linear model; mile1 is the operating mileage corresponding to the battery pack SOH data; k1 is the first slope; b1 is the first intercept;

[0081] The model formula corresponding to the second linear model is specifically:

[0082] Soh days =k2*days1+b2

[0083] Among them, Soh daysis the second linear model; days1 is the operating life corresponding to the battery pack SOH data, and the operating life is in days; k2 is the second slope; b2 is the second intercept.

[0084] As an optional implementation, in the second aspect of the present invention, the model formula corresponding to the third linear model is specifically:

[0085] Soh=k3*mile2+k4*days2+b

[0086] Wherein, Soh is the third linear model; mile2 is the new operating mileage; days2 is the new operating years; k3 is the first regression coefficient; k4 is the second regression coefficient; b is the third intercept;

[0087] The calculation formula used for performing preset data calculation based on the target matrix and the new battery pack SOH data in combination with the least squares method to obtain the solution value corresponding to the coefficient vector is specifically:

[0088] X T Xβ=X T y

[0089] Wherein, X is the target matrix; y is the new battery pack SOH data; β is the coefficient vector;

[0090] Then we can get the variant:

[0091] β=(X T X) -1 X T y

[0092] The β calculated by this variant is used to reversely determine the first regression coefficient, the second regression coefficient and the third intercept.

[0093] The third aspect of the present invention discloses another intelligent prediction device for battery health status, the device comprising:

[0094] A memory storing executable program code;

[0095] a processor coupled to the memory;

[0096] The processor calls the executable program code stored in the memory to execute the intelligent prediction method for the battery health status disclosed in the first aspect of the present invention.

[0097] The fourth aspect of the present invention discloses a computer storage medium, wherein the computer storage medium stores computer instructions, and when the computer instructions are called, they are used to execute the intelligent prediction method of battery health status disclosed in the first aspect of the present invention.

[0098] Compared with the prior art, the embodiments of the present invention have the following beneficial effects:

[0099] In an embodiment of the present invention, a method for intelligently predicting the health status of a battery is provided, and the method comprises: It can be seen that, by implementing the present invention, by constructing an initial analysis model including three key parameters of the battery pack SOH (health status), operating mileage and operating years, compared with a single-factor prediction model, it is possible to more comprehensively consider the multi-dimensional factors that affect the battery performance degradation, which is conducive to improving the prediction accuracy of the future use status of the battery pack, and is helpful to identify the trend of battery performance degradation in advance, and provide a scientific basis for battery maintenance and management; in addition, the constructed target analysis model can realize the intelligent prediction of the health status of the battery pack, and the prediction capability enables the battery management system to dynamically adjust the maintenance strategy according to the real-time data / predicted data, thereby optimizing the battery efficiency and extending the battery life. BRIEF DESCRIPTION OF THE DRAWINGS

[0100] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0101] Figure 1 It is a flowchart of a method for intelligently predicting the health status of a battery disclosed in an embodiment of the present invention;

[0102] Figure 2 It is a flowchart of another intelligent prediction method of battery health status disclosed in an embodiment of the present invention;

[0103] Figure 3 It is a structural schematic diagram of a battery health status intelligent prediction device disclosed in an embodiment of the present invention;

[0104] Figure 4 It is a structural schematic diagram of another intelligent prediction device for battery health status disclosed in an embodiment of the present invention;

[0105] Figure 5 It is a structural schematic diagram of another intelligent prediction device for battery health status disclosed in an embodiment of the present invention;

[0106] Figure 6 It is a schematic diagram of the correlation information between the operating mileage and operating years and the battery pack SOH disclosed in an embodiment of the present invention. DETAILED DESCRIPTION

[0107] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0108] The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, device, product or end including a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units inherent to these processes, methods, products or ends.

[0109] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

[0110] The present invention discloses a method and device for intelligent prediction of battery health status. By constructing an initial analysis model including three key parameters of battery pack SOH (health status), operating mileage and operating years, compared with the prediction model of a single factor, it can more comprehensively consider the multi-dimensional factors affecting battery performance degradation, which is conducive to improving the prediction accuracy of the future use status of the battery pack, and helps to identify the trend of battery performance degradation in advance, providing a scientific basis for battery maintenance and management; in addition, the constructed target analysis model can realize the intelligent prediction of the health status of the battery pack. This prediction capability enables the battery management system to dynamically adjust the maintenance strategy according to real-time data / predicted data, thereby optimizing the battery use efficiency and extending the battery life. The following are detailed descriptions.

[0111] Embodiment 1

[0112] See also Figure 1 , Figure 1 1 is a flow chart of a method for intelligently predicting the health status of a battery disclosed in an embodiment of the present invention. Figure 1 The described intelligent prediction method of battery health status can be applied to an intelligent prediction device of battery health status, and the embodiment of the present invention does not limit it. Figure 1 As shown, the intelligent prediction method of the battery health status may include the following operations:

[0113] 101. Obtain battery pack data to be analyzed, where the battery pack data includes battery pack SOH data, operating mileage corresponding to the battery pack SOH data, and operating years corresponding to the battery pack SOH data.

[0114] In the embodiment of the present invention, the battery pack SOH (State of Health) is the battery health state, which is a key indicator for measuring the battery performance state.

[0115] In an embodiment of the present invention, the battery pack data may refer to the data of the battery corresponding to the electric vehicle / electronic mobile device, and the battery may be a lithium battery. Furthermore, the battery pack data actually obtained in the present invention may refer to the battery data of the electric vehicle. Specifically, the battery pack SOH data may select SOH data with a numerical range of 85%-105%; the operating mileage may select an electric vehicle mileage of 0km-250,000km; and the operating years may select a year of 0-800 days.

[0116] 102. Construct an initial analysis model corresponding to the battery pack data, where the initial analysis model includes a first analysis model and a second analysis model.

[0117] In the embodiment of the present invention, the first analysis model is an analysis model corresponding to the battery pack SOH data and the operating mileage; the second analysis model is an analysis model corresponding to the battery pack SOH data and the operating years;

[0118] 103. Generate future battery pack data corresponding to the battery pack data according to the battery pack data and the initial analysis model, and generate a target data set according to the battery pack data and the future battery pack data.

[0119] In the embodiment of the present invention, the future battery pack data may also be referred to as predicted battery pack data, which is used to expand the data volume of the battery pack.

[0120] 104. Generate a target analysis model corresponding to the initial analysis model based on the target data set. The target analysis model is a comprehensive analysis model for three parameters in the target data set: new battery pack SOH data, new operating mileage, and new operating years.

[0121] In the embodiment of the present invention, the target analysis model is used to determine the SOH prediction data of any target battery pack.

[0122] In the embodiment of the present invention, the battery pack SOH prediction data of the target analysis model can help to immediately discover potential safety hazards of the battery and prevent safety accidents such as battery thermal runaway and short circuit, thereby improving the safety and reliability of the battery pack.

[0123] It can be seen that implementation Figure 1 The described intelligent prediction method of battery health status, by constructing an initial analysis model including three key parameters of battery pack SOH (health status), operating mileage and operating years, can more comprehensively consider the multi-dimensional factors affecting battery performance degradation compared to the prediction model of a single factor, which is conducive to improving the prediction accuracy of the future use status of the battery pack, and helps to identify the trend of battery performance degradation in advance, providing a scientific basis for battery maintenance and management; in addition, the constructed target analysis model can realize the intelligent prediction of the health status of the battery pack. This prediction capability enables the battery management system to dynamically adjust the maintenance strategy according to real-time data / predicted data, thereby optimizing the battery usage efficiency and extending the battery life.

[0124] In an optional embodiment, the method of constructing the initial analysis model corresponding to the battery pack data in step 102 specifically includes:

[0125] Taking the battery pack SOH data and the operating mileage as the first benchmark data, constructing a first linear model corresponding to the first benchmark data;

[0126] Taking the battery pack SOH data and the operating years as the second benchmark data, a second linear model corresponding to the second benchmark data is constructed;

[0127] Substituting the first benchmark data into the first linear model to obtain a first slope and a first intercept corresponding to the first linear model; and substituting the second benchmark data into the second linear model to obtain a second slope and a second intercept corresponding to the second linear model;

[0128] The first linear model is updated to a first analysis model according to the first slope and the first intercept; and the second linear model is updated to a second analysis model according to the second slope and the second intercept, and the first analysis model and the second analysis model are determined as initial analysis models.

[0129] In this optional embodiment, the model formula corresponding to the first linear model is specifically:

[0130] Soh mile =k1*mile1+b1

[0131] Among them, Soh mile is the first linear model; mile1 is the operating mileage corresponding to the battery pack SOH data; k1 is the first slope; b1 is the first intercept;

[0132] The model formula corresponding to the second linear model is as follows:

[0133] Soh days =k2*days1+b2

[0134] Among them, Soh days is the second linear model; days1 is the operating life corresponding to the battery pack SOH data, and the operating life is in days; k2 is the second slope; b2 is the second intercept.

[0135] In this optional embodiment, by substituting the first benchmark data and the second benchmark data, the first slope and the first intercept of the first linear model, and the second slope and the second intercept of the second linear model are calculated respectively, thereby achieving parameterization of the model. This parameterized model form is not only easy to understand and explain, but also easy to adjust and optimize according to new data, thereby continuously improving the accuracy of prediction.

[0136] It can be seen that in this optional embodiment, by respectively constructing a first linear model corresponding to the battery pack SOH data and operating mileage, and a second linear model corresponding to the battery pack SOH data and operating years, this modeling method based on linear relationships can more accurately capture the linear relationship between battery performance degradation and operating mileage and operating years, which is conducive to improving the reliability of subsequent data prediction analysis based on the initial analysis model.

[0137] Embodiment 2

[0138] See also Figure 2 , Figure 2 1 is a flow chart of another intelligent prediction method for battery health status disclosed in an embodiment of the present invention. Figure 2 The described intelligent prediction method of battery health status can be applied to an intelligent prediction device of battery health status, and the embodiment of the present invention does not limit it. Figure 2 As shown, the intelligent prediction method of the battery health status may include the following operations:

[0139] 201. Obtain battery pack data to be analyzed, where the battery pack data includes battery pack SOH data, operating mileage corresponding to the battery pack SOH data, and operating years corresponding to the battery pack SOH data.

[0140] 202. Construct an initial analysis model corresponding to the battery pack data, where the initial analysis model includes a first analysis model and a second analysis model.

[0141] 203. Generate future battery pack data corresponding to the battery pack data according to the battery pack data and the initial analysis model, and generate a target data set according to the battery pack data and the future battery pack data.

[0142] 204. Generate a target analysis model corresponding to the initial analysis model based on the target data set, where the target analysis model is a comprehensive analysis model for three parameters: new battery pack SOH data, new operating mileage, and new operating years in the target data set.

[0143] In the embodiment of the present invention, for other descriptions of step 201 to step 204, please refer to other specific descriptions of step 101 to step 104 in embodiment 1, and the embodiment of the present invention will not be repeated here.

[0144] 205. When the use demand for the target analysis model is detected, the parameters to be predicted are determined according to the use demand. The parameters to be predicted include the operating mileage to be predicted and the operating years to be predicted.

[0145] 206. Generate a target data combination based on the operating mileage to be predicted and the operating years to be predicted, the target data combination including a plurality of data pairs, each data pair consisting of a mileage value of the operating mileage to be predicted and a year value of the operating years to be predicted.

[0146] 207. Combining the target data to output a target analysis model, and obtaining target SOH prediction data.

[0147] In the embodiment of the present invention, the target SOH prediction data is SOH change information associated with the battery pack and the parameter to be predicted.

[0148] In the embodiment of the present invention, in specific application, 600,000 mileages can be selected as the operating mileage to be predicted, that is, the actual operating mileage to be predicted includes mileage data of 0-600,000 mileages; 8 years are selected as the operating years to be predicted, and the actual operating years to be predicted include mileage data of 0-8 years (specifically, data recording methods with days as units such as 1 day, 10 days, and 100 days can be adopted); then, for the target data combination, the data pairs included therein, each data pair is in the form of (operating mileage to be predicted, operating years to be predicted), or (operating years to be predicted, operating mileage to be predicted), for example (10km, 10 days) or (10 days, 10km); the numerical value of the operating mileage to be predicted corresponding to each operating years to be predicted is not limited in the embodiment of the present invention.

[0149] It can be seen that implementation Figure 2The described intelligent prediction method of battery health status can fully consider the joint impact of operating mileage and operating years on battery pack SOH by generating a target data combination containing multiple data pairs, each data pair consisting of a mileage value of the operating mileage to be predicted and a year value of the operating years to be predicted; this multi-dimensional data input method enables the target analysis model to more accurately capture the complex laws of battery performance degradation, thereby helping to improve the prediction accuracy; in addition, the target SOH prediction data is the SOH change information associated with the battery pack and the parameters to be predicted, which helps users understand the health status of the battery pack under different operating mileages and operating years in the future. This long-term and continuous prediction capability is conducive to improving the convenience of formulating battery maintenance plans and helping to achieve long-term health management of battery packs.

[0150] In an optional embodiment, the method of generating future battery pack data corresponding to the battery pack data according to the battery pack data and the initial analysis model in step 203 specifically includes:

[0151] determining a numerical filter interval for performing data filtering;

[0152] Determine target screening data matching the numerical screening interval from the battery pack SOH data corresponding to the battery pack data;

[0153] Input the target screening data into the first analysis model and the second analysis model to obtain the future operating mileage corresponding to the first analysis model and the future operating years corresponding to the second analysis model;

[0154] Determine the future operating mileage and its corresponding target screening data, the future operating years and its corresponding target screening data as the future battery pack data corresponding to the battery pack data;

[0155] And, the method of generating the target data set according to the battery pack data and future battery pack data specifically includes:

[0156] The battery pack data is spliced ​​with the future battery pack data execution data to obtain the target data set.

[0157] In this optional embodiment, the numerical screening interval can be selected as 75%-85%; correspondingly, the target screening data is the data corresponding to 75%-85% of the battery pack SOH values ​​in the target data set.

[0158] In this optional embodiment, the battery pack data and future battery pack data are spliced ​​to obtain a complete target data set containing historical and future information. This data splicing method not only retains the integrity of the original data, but also achieves seamless connection between historical data and predicted data, providing a rich data source for subsequent data analysis and model training.

[0159] It can be seen that in this optional embodiment, firstly, by determining the numerical screening interval, the target screening data that better matches the target analysis model is screened out from the battery pack data, ensuring that the data input into the first analysis model and the second analysis model have higher relevance and accuracy, thereby improving the accuracy of future battery pack data generation; at the same time, through targeted data screening, the interference of irrelevant data on the prediction results is effectively avoided, thereby enhancing the targeted nature of the prediction; the target screening data is input into the first analysis model and the second analysis model respectively, so as to obtain future battery pack data related to operating mileage and operating years. This multi-dimensional prediction method not only takes into account the decline of battery pack performance with operating mileage, but also takes into account the natural aging over time (operating years), which is conducive to improving the comprehensiveness and accuracy of the prediction of the future state of the battery pack.

[0160] In another optional embodiment, the method of generating the target analysis model corresponding to the initial analysis model according to the target data set in step 204 specifically includes:

[0161] Taking the new battery pack SOH data, new operating mileage and new operating years in the target data set as the third benchmark data, a third linear model corresponding to the third benchmark data is constructed, and the third linear model is a multivariate linear regression model; the third linear model includes a first regression coefficient, a second regression coefficient and a third intercept;

[0162] Generate a coefficient vector according to the first regression coefficient, the second regression coefficient and the third intercept;

[0163] Generate a target matrix based on new operating mileage and new operating years;

[0164] According to the target matrix and the new battery pack SOH data, the preset data calculation is performed in combination with the least squares method to obtain the solution value corresponding to the coefficient vector;

[0165] Update the first regression coefficient, the second regression coefficient and the third intercept according to the solution value corresponding to the coefficient vector;

[0166] The third linear model is updated to a target analysis model according to the first regression coefficient, the second regression coefficient and the third intercept.

[0167] In this optional embodiment, the model formula corresponding to the third linear model is specifically:

[0168] Soh=k3*mile2+k4*days2+b

[0169] Among them, Soh is the third linear model; mile2 is the new operating mileage; days2 is the new operating years; k3 is the first regression coefficient; k4 is the second regression coefficient; b is the third intercept;

[0170] According to the target matrix and the new battery pack SOH data, the preset data calculation is performed in combination with the least squares method to obtain the solution value corresponding to the coefficient vector. The calculation formula used is as follows:

[0171] X T Xβ=X T y

[0172] Among them, X is the target matrix; y is the new battery pack SOH data; β is the coefficient vector;

[0173] Then we can get the variant:

[0174] β=(X T X) -1 X T y

[0175] The β calculated by this variant is used to reversely determine the first regression coefficient, the second regression coefficient and the third intercept.

[0176] In this optional embodiment, by constructing a target matrix and a coefficient vector, the solution process of the multivariate linear regression model is converted into a matrix operation, which is conducive to improving data processing efficiency. This matrix processing method not only simplifies the calculation process, but also reduces the calculation complexity, making model updates and predictions faster and more efficient.

[0177] For this optional embodiment, please refer to Figure 6 , Figure 6 Schematic diagram of the association between the operating mileage and operating years and the battery pack SOH disclosed in the embodiment of the present invention. Figure 6 As shown, the associated information diagram specifically shows the relationship between the operating mileage & operating years and the change of battery pack SOH, where: Figure 6 The left vertical axis in the diagram is the operating years_bin, Figure 6 The horizontal axis in the diagram is the operating mileage mile_bin, Figure 6 The vertical axis on the right side of the diagram is the battery pack SOH, corresponding to the SOH attenuation value. Figure 6 From the color change trend of the middle block, it can be clearly seen that with the increase of the operating mileage and operating years of the battery pack, the battery pack SOH shows a trend of improvement.

[0178] It can be seen that in this optional embodiment, by constructing a third linear model (multiple linear regression model) and introducing new battery pack SOH data, new operating mileage and new operating years as the third benchmark data, the prediction accuracy of the battery pack state of health (SOH) is further improved; and, by continuously updating the regression coefficients and intercepts of the model, the target analysis model can better adapt to the complex laws of battery pack performance degradation, thereby improving the accuracy and reliability of the prediction; in addition, using the multivariate linear regression model as the basis of the third linear model, the target analysis model can simultaneously consider the impact of multiple influencing factors (such as operating mileage and operating years) on the battery pack SOH. This multi-dimensional data input method not only improves the prediction accuracy of the model, but also enhances the generalization ability of the model, so that it can be more widely applied to different types of battery packs and operating scenarios.

[0179] Embodiment 3

[0180] See also Figure 3 , Figure 3 : is a schematic diagram of the structure of a battery health status intelligent prediction device disclosed in an embodiment of the present invention. The battery health status intelligent prediction device can be a battery health status intelligent prediction terminal, device, system or server. The server can be a local server, a remote server, or a cloud server (also known as a cloud server). When the server is a non-cloud server, the non-cloud server can communicate with the cloud server, which is not limited in the embodiment of the present invention. Figure 3 As shown, the intelligent prediction device for the battery health status may include a data acquisition module 301, a model building module 302, a data generation module 303 and a model generation module 304, wherein:

[0181] The data acquisition module 301 is used to acquire the battery pack data to be analyzed, where the battery pack data includes the battery pack SOH data, the operating mileage corresponding to the battery pack SOH data, and the operating years corresponding to the battery pack SOH data.

[0182] The model building module 302 is used to build an initial analysis model corresponding to the battery pack data. The initial analysis model includes a first analysis model and a second analysis model. The first analysis model is an analysis model corresponding to the battery pack SOH data and operating mileage; the second analysis model is an analysis model corresponding to the battery pack SOH data and operating years.

[0183] The data generation module 303 is used to generate future battery pack data corresponding to the battery pack data according to the battery pack data and the initial analysis model, and to generate a target data set according to the battery pack data and the future battery pack data.

[0184] The model generation module 304 is used to generate a target analysis model corresponding to the initial analysis model based on the target data set. The target analysis model is a comprehensive analysis model for the three parameters of new battery pack SOH data, new operating mileage and new operating years in the target data set; and the target analysis model is used to determine the SOH prediction data of any target battery pack.

[0185] It can be seen that implementation Figure 3 The described intelligent prediction device for the battery health status, by constructing an initial analysis model including three key parameters of the battery pack SOH (health status), operating mileage and operating years, can more comprehensively consider the multi-dimensional factors affecting the battery performance degradation compared to the single-factor prediction model, which is conducive to improving the prediction accuracy of the future use status of the battery pack, and helps to identify the trend of battery performance degradation in advance, providing a scientific basis for battery maintenance and management; in addition, the constructed target analysis model can realize the intelligent prediction of the battery pack health status. This prediction capability enables the battery management system to dynamically adjust the maintenance strategy according to real-time data / predicted data, thereby optimizing the battery usage efficiency and extending the battery life.

[0186] In an optional embodiment, the model building module 302 builds the initial analysis model corresponding to the battery pack data in the following manner:

[0187] Taking the battery pack SOH data and the operating mileage as the first benchmark data, constructing a first linear model corresponding to the first benchmark data;

[0188] Taking the battery pack SOH data and the operating years as the second benchmark data, a second linear model corresponding to the second benchmark data is constructed;

[0189] Substituting the first benchmark data into the first linear model to obtain a first slope and a first intercept corresponding to the first linear model; and substituting the second benchmark data into the second linear model to obtain a second slope and a second intercept corresponding to the second linear model;

[0190] The first linear model is updated to a first analysis model according to the first slope and the first intercept; and the second linear model is updated to a second analysis model according to the second slope and the second intercept, and the first analysis model and the second analysis model are determined as initial analysis models.

[0191] In this optional embodiment, the model formula corresponding to the first linear model is specifically:

[0192] Soh mile =k1*mile1+b1

[0193] Among them, Soh mileis the first linear model; mile1 is the operating mileage corresponding to the battery pack SOH data; k1 is the first slope; b1 is the first intercept;

[0194] The model formula corresponding to the second linear model is as follows:

[0195] Soh days =k2*days1+b2

[0196] Among them, Soh days is the second linear model; days1 is the operating life corresponding to the battery pack SOH data, and the operating life is in days; k2 is the second slope; b2 is the second intercept.

[0197] It can be seen that in this optional embodiment, by respectively constructing a first linear model corresponding to the battery pack SOH data and operating mileage, and a second linear model corresponding to the battery pack SOH data and operating years, this modeling method based on linear relationships can more accurately capture the linear relationship between battery performance degradation and operating mileage and operating years, which is conducive to improving the reliability of subsequent data prediction analysis based on the initial analysis model.

[0198] In another optional embodiment, the data generation module 303 generates future battery pack data corresponding to the battery pack data according to the battery pack data and the initial analysis model, specifically including:

[0199] determining a numerical filter interval for performing data filtering;

[0200] Determine target screening data matching the numerical screening interval from the battery pack SOH data corresponding to the battery pack data;

[0201] Input the target screening data into the first analysis model and the second analysis model to obtain the future operating mileage corresponding to the first analysis model and the future operating years corresponding to the second analysis model;

[0202] Determine the future operating mileage and its corresponding target screening data, the future operating years and its corresponding target screening data as the future battery pack data corresponding to the battery pack data;

[0203] Furthermore, the data generation module 303 generates the target data set according to the battery pack data and the future battery pack data in the following manner:

[0204] The battery pack data is spliced ​​with the future battery pack data execution data to obtain the target data set.

[0205] It can be seen that in this optional embodiment, firstly, by determining the numerical screening interval, the target screening data that better matches the target analysis model is screened out from the battery pack data, ensuring that the data input into the first analysis model and the second analysis model have higher relevance and accuracy, thereby improving the accuracy of future battery pack data generation; at the same time, through targeted data screening, the interference of irrelevant data on the prediction results is effectively avoided, thereby enhancing the targeted nature of the prediction; the target screening data is input into the first analysis model and the second analysis model respectively, so as to obtain future battery pack data related to operating mileage and operating years. This multi-dimensional prediction method not only takes into account the decline of battery pack performance with operating mileage, but also takes into account the natural aging over time (operating years), which is conducive to improving the comprehensiveness and accuracy of the prediction of the future state of the battery pack.

[0206] In yet another optional embodiment, the model generation module 304 generates a target analysis model corresponding to the initial analysis model according to the target data set in a manner that specifically includes:

[0207] Taking the new battery pack SOH data, new operating mileage and new operating years in the target data set as the third benchmark data, a third linear model corresponding to the third benchmark data is constructed, and the third linear model is a multivariate linear regression model; the third linear model includes a first regression coefficient, a second regression coefficient and a third intercept;

[0208] Generate a coefficient vector according to the first regression coefficient, the second regression coefficient and the third intercept;

[0209] Generate a target matrix based on new operating mileage and new operating years;

[0210] According to the target matrix and the new battery pack SOH data, the preset data calculation is performed in combination with the least squares method to obtain the solution value corresponding to the coefficient vector;

[0211] Update the first regression coefficient, the second regression coefficient and the third intercept according to the solution value corresponding to the coefficient vector;

[0212] The third linear model is updated to a target analysis model according to the first regression coefficient, the second regression coefficient and the third intercept.

[0213] In this optional embodiment, the model formula corresponding to the third linear model is specifically:

[0214] Soh=k3*mile2+k4*days2+b

[0215] Among them, Soh is the third linear model; mile2 is the new operating mileage; days2 is the new operating years; k3 is the first regression coefficient; k4 is the second regression coefficient; b is the third intercept;

[0216] According to the target matrix and the new battery pack SOH data, the preset data calculation is performed in combination with the least squares method to obtain the solution value corresponding to the coefficient vector. The calculation formula used is as follows:

[0217] X T Xβ=X T y

[0218] Among them, X is the target matrix; y is the new battery pack SOH data; β is the coefficient vector;

[0219] Then we can get the variant:

[0220] β=(X T X) -1 X T y

[0221] The β calculated by this variant is used to reversely determine the first regression coefficient, the second regression coefficient and the third intercept.

[0222] It can be seen that in this optional embodiment, by constructing a third linear model (multiple linear regression model) and introducing new battery pack SOH data, new operating mileage and new operating years as the third benchmark data, the prediction accuracy of the battery pack state of health (SOH) is further improved; and, by continuously updating the regression coefficients and intercepts of the model, the target analysis model can better adapt to the complex laws of battery pack performance degradation, thereby improving the accuracy and reliability of the prediction; in addition, using the multivariate linear regression model as the basis of the third linear model, the target analysis model can simultaneously consider the impact of multiple influencing factors (such as operating mileage and operating years) on the battery pack SOH. This multi-dimensional data input method not only improves the prediction accuracy of the model, but also enhances the generalization ability of the model, so that it can be more widely applied to different types of battery packs and operating scenarios.

[0223] In another alternative embodiment, see Figure 4 , Figure 4 FIG. 1 is a schematic diagram of the structure of another intelligent prediction device for battery health status disclosed in an embodiment of the present invention. Figure 4 As shown, the device also includes a determination module 305 and a prediction module 306, wherein:

[0224] The determination module 305 is used to determine the parameters to be predicted according to the usage demand when the usage demand for the target analysis model is detected, and the parameters to be predicted include the operating mileage to be predicted and the operating years to be predicted;

[0225] The data generation module 303 is further used to generate a target data combination according to the operating mileage to be predicted and the operating years to be predicted, wherein the target data combination includes a plurality of data pairs, each data pair consisting of a mileage value of the operating mileage to be predicted and a year value of the operating years to be predicted;

[0226] Prediction module 306, used to combine target data and output target analysis model to obtain target SOH prediction data;

[0227] Among them, the target SOH prediction data is the SOH change information associated with the battery pack and the parameters to be predicted.

[0228] It can be seen that in this optional embodiment, by generating a target data combination including multiple data pairs, each data pair consists of a mileage value of the operating mileage to be predicted and a year value of the operating life to be predicted, which can fully consider the joint impact of the operating mileage and the operating life on the battery pack SOH; this multi-dimensional data input method enables the target analysis model to more accurately capture the complex laws of battery performance degradation, which is conducive to improving the prediction accuracy; in addition, the target SOH prediction data is the SOH change information associated with the battery pack and the parameters to be predicted, which helps users understand the health status of the battery pack under different operating mileages and operating years in the future. This long-term and continuous prediction capability is conducive to improving the convenience of formulating battery maintenance plans and helps to achieve long-term health management of battery packs.

[0229] Embodiment 4

[0230] See also Figure 5 , Figure 5 FIG. 1 is a schematic diagram of the structure of another intelligent prediction device for battery health status disclosed in an embodiment of the present invention. Figure 5 As shown, the intelligent prediction device for the battery health status may include:

[0231] A memory 401 storing executable program codes;

[0232] a processor 402 coupled to the memory 401;

[0233] The processor 402 calls the executable program code stored in the memory 401 to execute the steps of the intelligent prediction method for the battery health status described in the first embodiment of the present invention or the second embodiment of the present invention.

[0234] Embodiment 5

[0235] An embodiment of the present invention discloses a computer storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute the steps in the intelligent prediction method of battery health status described in Embodiment 1 or Embodiment 2 of the present invention.

[0236] Embodiment 6

[0237] An embodiment of the present invention discloses a computer program product, which includes a non-transitory computer storage medium storing a computer program, and the computer program is operable to enable a computer to execute the steps in the intelligent prediction method of the battery health status described in Example 1 or Example 2.

[0238] The device embodiments described above are only illustrative, wherein the modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical modules, i.e., they may be located in one place, or they may be distributed on multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Those of ordinary skill in the art may understand and implement it without creative work.

[0239] Through the specific description of the above embodiments, those skilled in the art can clearly understand that each implementation method can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on such an understanding, the above technical solution can be essentially or partly contributed to the prior art in the form of a software product, and the computer software product can be stored in a computer storage medium, and the storage medium includes a read-only memory (ROM), a random access memory (RAM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), a one-time programmable read-only memory (OTPROM), an electronically erasable rewritable read-only memory (EEPROM), a compact disc (CD-ROM) or other optical disc storage, magnetic disk storage, magnetic tape storage, or any other computer-readable medium that can be used to carry or store data.

[0240] Finally, it should be noted that the intelligent prediction method and device for the battery health status disclosed in the embodiment of the present invention discloses only the preferred embodiments of the present invention, which are only used to illustrate the technical scheme of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, it should be understood by those skilled in the art that the technical schemes described in the aforementioned embodiments can still be modified, or some of the technical features therein can be replaced by equivalents. However, these modifications or replacements do not deviate the essence of the corresponding technical schemes from the spirit and scope of the technical schemes of the embodiments of the present invention.

Claims

1. An intelligent prediction method for battery health status, characterized in that: The method comprises: Acquire battery pack data to be analyzed, the battery pack data including battery pack SOH data, operating mileage corresponding to the battery pack SOH data, and operating years corresponding to the battery pack SOH data; Constructing an initial analysis model corresponding to the battery pack data, the initial analysis model comprising a first analysis model and a second analysis model, the first analysis model being an analysis model corresponding to the battery pack SOH data and the operating mileage; the second analysis model being an analysis model corresponding to the battery pack SOH data and the operating years; generating future battery pack data corresponding to the battery pack data according to the battery pack data and the initial analysis model, and generating a target data set according to the battery pack data and the future battery pack data; According to the target data set, a target analysis model corresponding to the initial analysis model is generated, and the target analysis model is a comprehensive analysis model for the three parameters of new battery pack SOH data, new operating mileage and new operating years in the target data set; and the target analysis model is used to determine the SOH prediction data of any target battery pack.

2. The intelligent prediction method for battery health status according to claim 1, characterized in that: The constructing an initial analysis model corresponding to the battery pack data includes: Taking the battery pack SOH data and the operating mileage as first benchmark data, constructing a first linear model corresponding to the first benchmark data; Taking the battery pack SOH data and the operating years as second benchmark data, constructing a second linear model corresponding to the second benchmark data; Substituting the first benchmark data into the first linear model to obtain a first slope and a first intercept corresponding to the first linear model; and substituting the second benchmark data into the second linear model to obtain a second slope and a second intercept corresponding to the second linear model; The first linear model is updated to a first analysis model according to the first slope and the first intercept; and the second linear model is updated to a second analysis model according to the second slope and the second intercept, and the first analysis model and the second analysis model are determined as initial analysis models.

3. The intelligent prediction method for battery health status according to claim 1 or 2, characterized in that: The generating future battery pack data corresponding to the battery pack data according to the battery pack data and the initial analysis model includes: determining a numerical filter interval for performing data filtering; Determining target screening data matching the numerical screening interval from the battery pack SOH data corresponding to the battery pack data; Inputting the target screening data into the first analysis model and the second analysis model to obtain the future operating mileage corresponding to the first analysis model and the future operating years corresponding to the second analysis model; Determine the future operating mileage and the corresponding target screening data, the future operating years and the corresponding target screening data as future battery pack data corresponding to the battery pack data; And, generating a target data set according to the battery pack data and the future battery pack data includes: The battery pack data and the future battery pack data are performed data splicing to obtain a target data set.

4. The intelligent prediction method for battery health status according to claim 3, characterized in that: The step of generating a target analysis model corresponding to the initial analysis model according to the target data set includes: Taking the new battery pack SOH data, new operating mileage and new operating years in the target data set as the third benchmark data, a third linear model corresponding to the third benchmark data is constructed, and the third linear model is a multivariate linear regression model; the third linear model includes a first regression coefficient, a second regression coefficient and a third intercept; generating a coefficient vector according to the first regression coefficient, the second regression coefficient and the third intercept; Generate a target matrix according to the new operating mileage and the new operating years; According to the target matrix and the new battery pack SOH data, a preset data calculation is performed in combination with a least squares method to obtain a solution value corresponding to the coefficient vector; Update the first regression coefficient, the second regression coefficient and the third intercept according to the solution value corresponding to the coefficient vector; The third linear model is updated to a target analysis model according to the first regression coefficient, the second regression coefficient and the third intercept.

5. The intelligent prediction method for battery health status according to claim 1, 2 or 4, characterized in that: The method further comprises: When a use demand for the target analysis model is detected, a parameter to be predicted is determined according to the use demand, and the parameter to be predicted includes an operating mileage to be predicted and an operating life to be predicted; Generate a target data combination according to the to-be-predicted operating mileage and the to-be-predicted operating years, wherein the target data combination includes a plurality of data pairs, each of which is composed of a mileage value of the to-be-predicted operating mileage and a year value of the to-be-predicted operating years; Combining the target data and outputting the target analysis model to obtain target SOH prediction data; The target SOH prediction data is SOH change information associated with the battery pack and the parameter to be predicted.

6. The intelligent prediction method of battery health status according to claim 2, characterized in that: The model formula corresponding to the first linear model is specifically: Soh mile =k1*mile1+b1 Among them, Soh mile is the first linear model; mile1 is the operating mileage corresponding to the battery pack SOH data; k1 is the first slope; b1 is the first intercept; The model formula corresponding to the second linear model is specifically: <h2 style=";text-align:left;direction:ltr">Soh<h2 style=";text-align:left;direction:ltr"> days <h2 style=";text-align:left;direction:ltr"> =k2*days1+b2 Among them, Soh days is the second linear model; days1 is the operating life corresponding to the battery pack SOH data, and the operating life is in days; k2 is the second slope; b2 is the second intercept.

7. The intelligent prediction method of battery health status according to claim 4, characterized in that: The model formula corresponding to the third linear model is as follows: Soh=k3*mile2+k4*days2+b in, Soh is the third linear model; mile2 is the new operating mileage; days2 is the new operating years; k3 is the first regression coefficient; k4 is the second regression coefficient; b is the third intercept; The calculation formula used for performing preset data calculation based on the target matrix and the new battery pack SOH data in combination with the least squares method to obtain the solution value corresponding to the coefficient vector is specifically: X T Xβ=X T y Wherein, X is the target matrix; y is the new battery pack SOH data; β is the coefficient vector; Then we can get the variant: β=(X T X) -1 X T y The β calculated by this variant is used to reversely determine the first regression coefficient, the second regression coefficient and the third intercept.

8. An intelligent prediction device for battery health status, characterized in that: The device comprises: A data acquisition module, used to acquire battery pack data to be analyzed, wherein the battery pack data includes battery pack SOH data, operating mileage corresponding to the battery pack SOH data, and operating years corresponding to the battery pack SOH data; A model building module, used to build an initial analysis model corresponding to the battery pack data, the initial analysis model comprising a first analysis model and a second analysis model, the first analysis model being an analysis model corresponding to the battery pack SOH data and the operating mileage; the second analysis model being an analysis model corresponding to the battery pack SOH data and the operating years; a data generation module, configured to generate future battery pack data corresponding to the battery pack data according to the battery pack data and the initial analysis model, and to generate a target data set according to the battery pack data and the future battery pack data; A model generation module is used to generate a target analysis model corresponding to the initial analysis model based on the target data set, wherein the target analysis model is a comprehensive analysis model for three parameters in the target data set: new battery pack SOH data, new operating mileage, and new operating years; and the target analysis model is used to determine the SOH prediction data of any target battery pack.

9. An intelligent prediction device for battery health status, characterized in that: The device comprises: A memory storing executable program code; a processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the intelligent prediction method for the battery health status as described in any one of claims 1-7.

10. A computer storage medium, characterized in that: The computer storage medium stores computer instructions, which, when called, are used to execute the intelligent prediction method for the battery health status as described in any one of claims 1 to 7.

Images