Data processing method and device, equipment, medium and program product

Abstract

The invention provides a data processing method and device, equipment, a medium and a program product. The method comprises the steps of obtaining to-be-processed battery data and an analysis target for the to-be-processed battery data, wherein the analysis target comprises at least one battery index; determining a keyword corresponding to the analysis target according to the at least one battery index; according to the keyword, at least one battery analysis model used for processing the to-be-processed data is determined in a battery analysis model library, the battery analysis model library comprises a plurality of battery analysis models, and different battery analysis models are used for achieving different battery analysis functions; and processing the to-be-processed battery data through the at least one battery analysis model to obtain the value of each battery index in the analysis target. According to the battery data analysis method and device, various suitable battery analysis models can be automatically screened out from the battery analysis model library according to the analysis target corresponding to the battery data, rapid analysis processing of the battery data can be achieved, and the processing efficiency of a battery data analysis task is improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a data processing method, device, equipment, medium and program product. Background Art

[0002] In the field of battery technology and its applications, battery analysis models are widely used in the in-depth analysis of battery data, aiming to improve the accuracy and efficiency of battery performance evaluation. With the vigorous development of the battery industry, battery analysis models have undergone rapid updates and iterations. There may be multiple different battery analysis models to choose from for the same analysis indicator. Even for the same battery analysis model, when applied to different scenarios or needs, it is often necessary to adjust its model parameters according to actual conditions.

[0003] However, with the model library becoming increasingly large and the model iteration speed accelerating, related technologies still generally rely on manual methods to select suitable battery analysis models when processing battery data analysis tasks, which leads to low efficiency. Summary of the invention

[0004] The present application provides a data processing method, device, equipment, medium and program product to solve the defect of the prior art that a suitable battery analysis model is selected manually, which can realize rapid analysis and processing of battery data and improve the processing efficiency of battery data analysis tasks.

[0005] The present application provides a battery data processing method, comprising the following steps: Acquire battery data to be processed and an analysis target for the battery data to be processed, wherein the analysis target includes at least one battery indicator; Determining a keyword corresponding to the analysis target according to the at least one battery indicator; According to the keyword, determining at least one battery analysis model for processing the data to be processed in a battery analysis model library, wherein the battery analysis model library includes a plurality of battery analysis models, and different battery analysis models are used to implement different battery analysis functions; The battery data to be processed is processed by using the at least one battery analysis model to obtain the value of each of the battery indicators in the analysis target.

[0006] According to a battery data processing method provided by the present application, determining at least one battery analysis model for processing the data to be processed in a battery analysis model library according to the keyword includes: In the battery analysis model library, determining a battery analysis model whose model name contains any of the keywords as a first type of model; According to the constraint conditions of the input features and the output features of each of the battery analysis models in the first category of models, the first category of models is screened to obtain the at least one battery analysis model.

[0007] According to a battery data processing method provided by the present application, a battery analysis model includes at least one battery analysis sub-model, and different battery analysis sub-models in the same battery analysis model are used to implement different data processing functions. The first type of model is screened according to the constraint conditions and output characteristics of the input characteristics of each battery analysis model in the first type of model to obtain the at least one battery analysis model, including: According to the constraint conditions and output characteristics of the input characteristics of each of the battery analysis models in the first type of models, the first type of models are screened to obtain a second type of model; Determining the data processing function lacking in the second type of model according to the output characteristics of the battery analysis sub-models included in each of the battery analysis models in the second type of model; Determining a third type of model in the battery analysis model library for implementing the missing data processing function; The at least one battery analysis model is determined according to the second type of model and the third type of model.

[0008] According to a battery data processing method provided by the present application, the first type of models includes a battery analysis model that is mandatory to use. After screening the first type of models, the method further includes: If the constraint conditions of the input features of the battery analysis model that is forced to be used do not meet the preset conditions, the data to be processed is adjusted until the constraint conditions of the input features of the battery analysis model that is forced to be used meet the preset conditions.

[0009] According to a battery data processing method provided by the present application, a battery analysis model includes at least one battery analysis sub-model, and different battery analysis sub-models in the same battery analysis model are used to implement different data processing functions; the processing of the battery data to be processed by the at least one battery analysis model includes: Determining a connection relationship between each of the battery analysis sub-models according to an output feature of each of the battery analysis models and an output feature of a battery analysis sub-model included in each of the battery analysis models; Generate a table of events to be executed according to the output characteristics of each of the battery analysis sub-models and the connection relationship, wherein the table of events to be executed includes various events that need to be executed when achieving the analysis goal and the execution order of each of the events; Each event in the to-be-executed event table is executed through each of the battery analysis sub-models in accordance with the execution order of each of the events, so as to realize the processing of the to-be-processed battery data.

[0010] According to a battery data processing method provided by the present application, the connection relationship between each of the battery analysis sub-models is determined according to the output characteristics of each of the battery analysis models and the output characteristics of the battery analysis sub-models contained in each of the battery analysis models, including: Deduplication processing is performed on battery analysis sub-models having the same function in each of the battery analysis sub-models according to output characteristics of each of the battery analysis sub-models; The connection relationship between the battery analysis sub-models is determined according to the output characteristics of each battery analysis model and the output characteristics of each battery analysis sub-model obtained after deduplication.

[0011] According to a battery data processing method provided by the present application, the step of obtaining the battery data to be processed includes: Obtain historical battery data corresponding to the battery to be analyzed; The historical battery data is preprocessed to obtain preprocessed battery data, where the preprocessed battery data is the battery data to be processed.

[0012] The present application also provides a battery data processing device, comprising the following modules: An acquisition module, used to acquire battery data to be processed and an analysis target for the battery data to be processed, wherein the analysis target includes at least one battery indicator; A first determination module, configured to determine a keyword corresponding to the analysis target according to the at least one battery indicator; A second determination module is used to determine, according to the keyword, at least one battery analysis model for processing the data to be processed in a battery analysis model library, wherein the battery analysis model library includes a plurality of battery analysis models, and different battery analysis models are used to implement different battery analysis functions; The processing module processes the battery data to be processed through the at least one battery analysis model to obtain the value of each battery indicator in the analysis target.

[0013] The present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the computer program, a battery data processing method as described in any one of the above is implemented.

[0014] The present application also provides a non-transitory computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, a battery data processing method as described in any one of the above is implemented.

[0015] The present application also provides a computer program product, including a computer program, wherein when the computer program is executed by a processor, the battery data processing method as described in any one of the above is implemented.

[0016] The present application provides a data processing method, device, equipment, medium and program product. To implement the battery data processing method of the present application, first obtain the battery data to be processed and the analysis target for the battery data to be processed, the analysis target includes at least one battery indicator; then, according to the at least one battery indicator, determine the keyword corresponding to the analysis target; then, according to the keyword, determine at least one battery analysis model for processing the data to be processed in the battery analysis model library, the battery analysis model library includes multiple battery analysis models, and different battery analysis models are used to implement different battery analysis functions; finally, the battery data to be processed is processed by at least one battery analysis model to obtain the value of each battery indicator in the analysis target. The present application can automatically select various suitable battery analysis models in the battery analysis model library according to the analysis target of the battery data, realize the rapid analysis and processing of the battery data, improve the processing efficiency of the battery data analysis task, and better meet the user's battery analysis needs and business needs in various battery fields. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the present application or the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 is a flow chart of a battery data processing method shown in an embodiment of the present application; Figure 2 is a schematic diagram of the relationship between a battery analysis model shown in an embodiment of the present application; Figure 3 is a functional architecture diagram of an electronic device shown in an embodiment of the present application; Figure 4 is a schematic diagram of a complete process of processing battery data shown in an embodiment of the present application; Figure 5 is a structural block diagram of a battery data processing device shown in an embodiment of the present application; Figure 6It is a schematic diagram of the physical structure of an electronic device shown in an embodiment of the present application. DETAILED DESCRIPTION

[0019] In order to make the purpose, technical solutions and advantages of this application clearer, the technical solutions in this application will be clearly and completely described below in conjunction with the drawings in this application. Obviously, the described embodiments are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0020] The execution subject of the battery data processing method of the present application is any type of electronic device, or a battery data processing device. The method of the present application will be described below by taking the execution subject as an electronic device as an example.

[0021] Figure 1 is a flow chart of a battery data processing method shown in an embodiment of the present application. Figure 1 , the battery data processing method of the present application specifically includes the following steps: Step 101: Obtain battery data to be processed and an analysis target for the battery data to be processed, wherein the analysis target includes at least one battery indicator.

[0022] The battery index may be an index representing battery performance parameters, including but not limited to energy density, power density, charge and discharge efficiency, etc., or an index representing life and reliability parameters, including but not limited to state of charge, depth of discharge, health status, etc. The battery index may be set according to actual needs, and this embodiment does not impose specific restrictions on this.

[0023] Specifically, obtaining the battery data to be processed may include: Obtain historical battery data corresponding to the battery to be analyzed; The historical battery data is preprocessed to obtain preprocessed battery data, and the preprocessed battery data is the battery data to be processed.

[0024] In this embodiment, after acquiring the historical battery data, the historical battery data may be cleaned to eliminate errors, redundancy, etc. in the data to ensure the accuracy and consistency of the data. Data cleaning operations may include missing value processing (deletion method, filling method, marking method, etc.), abnormal value processing, data consistency check and correction (ensuring consistency of data format, unit, naming, etc.), duplicate data processing, etc. Data preprocessing operations may include data integration (merging data from multiple data sources), data transformation (feature scaling, feature encoding, data conversion, etc.), data selection (identifying and removing irrelevant or redundant features), etc. Data cleaning operations and data preprocessing operations can be set according to actual needs, and this embodiment does not limit this.

[0025] Step 102: Determine a keyword corresponding to an analysis target according to at least one battery indicator.

[0026] In one implementation, each battery indicator may be directly used as a keyword. For example, when the battery indicators are state of charge and state of health, the state of charge and state of health may be directly used as keywords corresponding to the analysis target.

[0027] In another embodiment, if each battery indicator is directly used as a keyword, the battery analysis model screened subsequently may not be comprehensive enough, so the battery indicators need to be expanded to obtain expanded keywords. When expanding keywords, some intelligent analysis tools can be used for expansion. For example, each battery indicator is input into the intelligent analysis tool, and the intelligent analysis tool will output the relevant expanded keywords. Of course, other methods can also be used for expansion, such as pre-designing an expanded word table corresponding to each battery indicator. When expanding keywords, the expanded words corresponding to each battery indicator can be directly queried in the expanded word table to obtain the expanded keywords.

[0028] In this embodiment, other methods other than the above methods may be used to determine the keywords corresponding to the analysis target, and this embodiment does not impose any specific limitation on this.

[0029] Step 103: According to the keyword, at least one battery analysis model for processing the data to be processed is determined in the battery analysis model library, wherein the battery analysis model library includes a plurality of battery analysis models, and different battery analysis models are used to implement different battery analysis functions.

[0030] In this embodiment, a plurality of different types of battery analysis models are stored in the battery analysis model library, and for each type of battery analysis model, a plurality of different battery analysis models are specifically included. By way of example, for a battery analysis model of a type for calculating the state of charge, a plurality of different battery analysis models may be specifically included, and different battery analysis models respectively use different algorithms to calculate the state of charge, or the constraints of the input features of the models are different. By way of another example, for a battery analysis model of a type for calculating the state of health, a plurality of different battery analysis models may be specifically included, and different battery analysis models respectively use different algorithms to calculate the state of health, or the constraints of the input features of the models are different.

[0031] When executing step 103, a matching battery analysis model may be selected from the battery analysis model library according to the keyword. For example, when the keywords are state of charge and state of health, a battery analysis model for calculating the state of charge and a battery analysis model for calculating the state of health may be selected from the battery analysis model library to obtain at least one battery analysis model.

[0032] Step 104: Process the battery data to be processed by at least one battery analysis model to obtain the value of each battery indicator in the analysis target.

[0033] For example, if the analysis target includes two battery indicators, namely state of charge and state of health, after obtaining battery analysis model 1 (calculating state of charge) and battery analysis model 2 (calculating state of health), the value of the battery indicator of state of charge can be obtained by calling battery analysis model 1 to process the battery data to be processed, and the value of the battery indicator of state of health can be obtained by calling battery analysis model 2 to process the battery data to be processed.

[0034] To implement the battery data processing method of the present application, first obtain the battery data to be processed and the analysis target for the battery data to be processed, wherein the analysis target includes at least one battery indicator; then, determine the keyword corresponding to the analysis target based on the at least one battery indicator; then, determine at least one battery analysis model for processing the data to be processed in the battery analysis model library based on the keyword, wherein the battery analysis model library includes multiple battery analysis models, and different battery analysis models are used to implement different battery analysis functions; finally, process the battery data to be processed through at least one battery analysis model to obtain the value of each battery indicator in the analysis target. The present application can automatically select various suitable battery analysis models in the battery analysis model library according to the analysis target of the battery data, realize rapid analysis and processing of battery data, improve the processing efficiency of battery data analysis tasks, and better meet the battery analysis needs of users and business needs in various battery fields.

[0035] In combination with the above embodiments, in one implementation, at least one battery analysis model for processing the data to be processed is determined in the battery analysis model library according to the keywords, which may specifically include: In the battery analysis model library, the battery analysis models whose model names contain any keywords are determined to be first-class models; According to the constraint conditions of the input features and the output features of each battery analysis model in the first type of models, the first type of models are screened to obtain at least one battery analysis model.

[0036] In this embodiment, the first type of model is a battery analysis model obtained by preliminary screening from a battery analysis model library. The first type of model may include a battery analysis model whose model name contains any keyword.

[0037] In addition, the first type of models may also include battery analysis models that are forced to be used by users. The battery analysis models that are forced to be used by users may refer to battery analysis models that do not contain any keywords in the model name but are determined by the user to be necessary to achieve the analysis goal.

[0038] In the first type of model, each battery analysis model has input feature constraints and output features. Input features refer to the type of input data, such as battery charging data or battery discharge data. Output features refer to the type of output data, such as power density, charge and discharge efficiency, state of charge, depth of discharge, health status, etc. Constraints are restrictions on input features, such as the maximum value of the input feature cannot be greater than X, or the minimum value of the input data cannot be less than Y, or the input feature must be data generated between time T1 and time T2. Constraints can be set according to actual needs.

[0039] Next, at least one battery analysis model is further screened out in the first category of models according to the constraints of the input features and the output features of each battery analysis model in the first category of models. For example, the first category of models includes battery analysis models 1 to 10. If the constraints of battery analysis model 1 are not applicable to the battery data to be processed (i.e., the battery data to be processed cannot be used as the input features of battery analysis model 1), then battery analysis model 1 needs to be eliminated; if the output features of battery analysis model 2 do not match the expected output features (e.g., the expected output features are features 1 and 2, but the output features of battery analysis model 2 are features 1), then battery analysis model 2 needs to be eliminated. According to the same principle, at least one battery analysis model for final use is screened out in the first category of models.

[0040] In combination with the above embodiments, in one implementation, after screening the first type of models according to the constraints of the input features and the output features of each battery analysis model in the first type of models, the method of the present application may further include: If the constraint conditions of the input features of the battery analysis model that is forced to be used do not meet the preset conditions, the data to be processed is adjusted until the constraint conditions of the input features of the battery analysis model that is forced to be used meet the preset conditions.

[0041] In this embodiment, if a battery analysis model is a battery analysis model that is forced to be used, and the constraints of the battery analysis model are not applicable to the battery data to be processed (that is, the battery data to be processed cannot be used as input features of the battery analysis model), then the input part of the data corresponding to the battery analysis model in the data to be processed is reprocessed, for example, by using interpolation or completion methods, etc., until the constraints of the battery analysis model are applicable to the battery data to be processed.

[0042] Through this embodiment, a suitable battery analysis model can be quickly screened out in the battery analysis model library, so as to facilitate subsequent rapid analysis and processing of battery data.

[0043] In combination with the above embodiments, in one implementation manner, a battery analysis model includes at least one battery analysis sub-model, and different battery analysis sub-models in the same battery analysis model are used to implement different data processing functions.

[0044] In this embodiment, a battery analysis model is actually composed of at least one battery analysis sub-model. A battery analysis sub-model can be understood as an algorithm, and therefore, the battery analysis model includes one or more different algorithms. For example, battery analysis model X is used to calculate battery index Y, and battery analysis model X includes battery analysis sub-model 1-battery analysis sub-model 4, battery analysis sub-model 1 is used to calculate parameter 1 based on input features, battery analysis sub-model 2 is used to calculate parameter 2 based on input features, battery analysis sub-model 3 is used to calculate parameter 3 based on parameter 1 and parameter 2, and battery analysis sub-model 4 is used to calculate battery index Y based on parameters 1-parameter 4. Battery analysis sub-model 1-battery analysis sub-model 4 represent different algorithms respectively.

[0045] Accordingly, according to the constraints of the input features and the output features of each battery analysis model in the first type of models, the first type of models are screened to obtain at least one battery analysis model, which may include: According to the constraint conditions and output characteristics of the input characteristics of each battery analysis model in the first type of model, the first type of model is screened to obtain the second type of model; According to the output characteristics of the battery analysis sub-models included in each battery analysis model in the second type of model, the data processing function lacking in the second type of model is determined; Identify a third type of model in the battery analysis model library for implementing the missing data processing functions; At least one battery analysis model is determined according to the second type of model and the third type of model.

[0046] Among them, the output characteristics of the battery analysis sub-model refer to the type of output data of the battery analysis sub-model, for example, the output characteristics can be parameter 1-parameter 10. The data processing function of the battery analysis sub-model can be determined according to the output characteristics of the battery analysis sub-model. For example, if the output characteristic of a battery analysis sub-model is volume energy density (the energy that can be released by the battery per unit volume), then it means that the data processing function of the battery analysis sub-model is to calculate the volume energy density of the battery. For another example, if the output characteristic of a battery analysis sub-model is mass power density (the maximum power that can be output by the battery per unit mass), then it means that the data processing function of the battery analysis sub-model is to calculate the mass power density of the battery.

[0047] In this embodiment, first, the second type of model is screened in the first type of model according to the constraints of the input features and the output features of each battery analysis model in the first type of model, and the specific screening process is as described above. Then, according to the output features of the battery analysis sub-models contained in each battery analysis model, the data processing function of each battery analysis sub-model is determined, and it is compared with the expected data processing function to find out the data processing function lacking in the second type of model. For example, the battery analysis model M (used to calculate the battery index Y) in the second type of model includes battery analysis sub-model 1-battery analysis sub-model 3, battery analysis sub-model 1 is used to calculate parameter 1 according to the input features, battery analysis sub-model 2 is used to calculate parameter 2 according to the input features, and battery analysis sub-model 3 is used to calculate the battery index Y according to parameter 1 and parameter 2. Although the battery analysis model M can realize the calculation of the battery index Y, the final expected result is the average value of the battery index Y calculated multiple times, or the battery index Y in a certain specific data format, then the battery analysis model M lacks the data processing function of calculating the average value, or the data processing function of converting the data into a specific data format.

[0048] After determining the missing data processing function, continue to determine the battery analysis model (third type model) used to implement the missing data processing function in the battery analysis model library, then combine the second type model and the third type model, and determine the individual battery analysis models obtained after the combination as at least one battery analysis model.

[0049] In summary, in this embodiment, firstly, the first type of model is screened out in the battery analysis model library according to the keywords, and the second type of model is screened out in the first type of model; then, the output characteristics of the battery analysis sub-models contained in each battery analysis model in the second type of model are analyzed to determine whether there is a lack of data processing function, and if there is a lack of data processing function, then the battery analysis model (third type of model) that can realize the lack of data processing function is screened out in the battery analysis model library according to the lack of data processing function, and at least one battery analysis model is obtained by combining the second type of model and the third type of model. Then, the output characteristics of each sub-model (the battery analysis sub-model is composed of at least one algorithm, and these algorithms represent the sub-models included in the battery analysis sub-model) further included in each battery analysis sub-model in at least one battery analysis model can be analyzed to determine whether there is a lack of data processing function, and then the battery analysis model that can realize the lack of data processing function is screened out in the battery analysis model library according to the lack of data processing function. Repeat the above steps and perform multiple levels of analysis until a perfect battery analysis model is obtained.

[0050] Through this embodiment, a suitable battery analysis model can be accurately screened out from the battery analysis model library to avoid the phenomenon of missing a battery analysis model, thereby providing technical support for the subsequent accurate analysis of battery data.

[0051] In combination with the above embodiments, in one implementation, processing the battery data to be processed by at least one battery analysis model may include: Determine the connection relationship between each battery analysis sub-model according to the output characteristics of each battery analysis model and the output characteristics of the battery analysis sub-model contained in each battery analysis model; Generate a table of pending events based on the output characteristics and connection relationships of each battery analysis sub-model, which includes each event that needs to be executed to achieve the analysis goal and the execution order of each event; Through each battery analysis sub-model, each event in the to-be-executed event table is executed in the execution order of each event, so as to realize the processing of the battery data to be processed.

[0052] Among them, executing each event in the to-be-executed event table according to the execution order of each event can execute the events in parallel to the greatest extent, so as to achieve the fastest processing of the battery data to be processed.

[0053] In this embodiment, each battery analysis model determined in step 103 is recorded in the model table to be loaded. The electronic device obtains a feature registration table based on the model table to be loaded, and the feature registration table records the output features of each battery analysis model and the output features of the battery analysis sub-models contained in each battery analysis model.

[0054] Next, the electronic device performs intelligent analysis based on the analysis target and each feature in the feature registration table to obtain the connection relationship between each battery analysis sub-model, as shown in the following figure: Figure 2 shown. Figure 2 It is a schematic diagram of the relationship between a battery analysis model shown in an embodiment of the present application.

[0055] exist Figure 2 In the figure, the battery analysis model includes a first battery analysis model A, a second battery analysis model B and a third battery analysis model C. The first battery analysis model A includes sub-models 1 to 4, the second battery analysis model B includes sub-models 5 to 14, and the third battery analysis model C includes sub-models 15 to 16. The connection relationship (data flow relationship) between the sub-models is as follows: Figure 2 As shown by the arrow in Figure 2 In the table of models to be loaded, all battery analysis models are divided into models of multiple levels. The first battery analysis model A is a model of the first level, the second battery analysis model B is a model of the second level, and the third battery analysis model C is a model of the third level. Models of different levels are in series relationship. Submodel 1-submodel 16 are different task units (basic calculation units). Different task units can form a task block. For example, the task unit corresponding to submodel 3 and the task unit corresponding to submodel 4 can form a task block, and the task unit corresponding to submodel 5 and the task unit corresponding to submodel 6 can form a task block. Multiple different task units can form multiple different calculation sequences. The calculation process composed of submodel 1, submodel 2, submodel 3, submodel 5, submodel 11, submodel 14 and submodel 15 is a calculation sequence; the calculation process composed of submodel 1, submodel 2, submodel 4, submodel 6, submodel 8 and submodel 16 is a calculation sequence.

[0056] Next, according to Figure 2 The output features and connection relationships of each battery analysis sub-model shown generate a table of pending events, which includes various events that need to be executed to achieve the analysis goals. For example, an event can be calculation parameter 1, calculation parameter 2, calculation parameter 3, etc. Finally, by calling each battery analysis sub-model, each event in the pending event table is executed to realize the processing of the battery data to be processed. For example, the function of battery analysis sub-model 1 is to calculate parameter 1, and the function of battery analysis sub-model 2 is to calculate parameter 2. Then, battery analysis sub-model 1 can be called to realize the calculation of parameter 1, and battery analysis sub-model 2 can be called to realize the calculation of parameter 2. Until all events in the pending event table are executed, it means that the processing of the data to be processed is completed.

[0057] In combination with the above embodiments, in one implementation, the connection relationship between the battery analysis sub-models is determined according to the output characteristics of each battery analysis model and the output characteristics of the battery analysis sub-models contained in each battery analysis model, including: According to the output characteristics of each battery analysis sub-model, the battery analysis sub-models with the same functions in each battery analysis sub-model are deduplicated; According to the output characteristics of each battery analysis model and the output characteristics of each battery analysis sub-model obtained after deduplication, the connection relationship between each battery analysis sub-model, as well as the serial execution order and parallel execution order between each battery analysis sub-model are determined. These serial execution orders and parallel execution orders help to more accurately and reasonably determine the execution order of each event in the event table to be executed, thereby realizing the rapid processing of the battery data to be processed.

[0058] In this embodiment, different battery analysis models may have battery analysis sub-models with the same functions. In order to avoid repeated calculation of input features by models of the same type, this embodiment can perform deduplication processing on battery analysis sub-models with the same functions in various battery analysis sub-models, thereby speeding up data processing efficiency. In this embodiment, rapid analysis and processing of battery data can be achieved by calling various pre-matched battery analysis models, better meeting the user's battery analysis needs and business needs in various battery fields.

[0059] Figure 3 1 is a functional architecture diagram of an electronic device shown in an embodiment of the present application. Figure 3 The processing architecture for battery data in this application generally includes a data calculation module, a model library management model, a model loading rule module and an exception handling module.

[0060] The functions implemented by the data calculation module mainly include data cleaning, preprocessing, reprocessing, model loading and calculation. Data cleaning implements the cleaning operation of battery data. Preprocessing implements the preprocessing of cleaned battery data. Reprocessing is used to improve data quality, usually using methods such as interpolation and completion. Model loading is used to determine at least one battery analysis model in the model library and load it. Calculation is used to call the loaded battery analysis model and perform data calculation.

[0061] The functions implemented by the model library management module mainly include new model identification, model modification, model deletion and model viewing. New model identification is used to automatically identify new models and dynamically load new models. Model modification includes model label modification, feature parameter and threshold modification, model basic information update and available status modification. Model deletion is used to delete a model. Model viewing is used to view detailed information of a model.

[0062] The functions implemented by the model loading rule module include rule addition, deletion, modification and query, and rule type management, among which rule type management mainly includes blacklist and whitelist management and model availability status management.

[0063] The exception handling module is used to handle various abnormal events that occur. The type and processing method of the abnormal events can be set according to actual needs.

[0064] For the identification of new models, all model files are scanned regularly. When changes are detected in model files, the model deletion process, model modification process, or new model process is entered. When entering the new model process, the model information of the new model is first read to determine whether there is an abnormality in the model information. If there is no abnormality, the model information table is updated according to the model information, and the new model event is recorded. If there is an abnormality, the exception is handled according to the pre-set exception handling strategy.

[0065] To modify the model, on the one hand, the model file can be modified, and on the other hand, the model information table can be modified. When modifying the model information table, first determine whether the modified parameters exceed the default boundary of the model to be modified (the constraints of the input features). If not, update the model information table according to the modified parameters, and then record the update event. When modifying the model information table, the modified content can include model labels, feature methods, available status, basic information, etc.

[0066] Figure 4 FIG. 1 is a schematic diagram of a complete process of processing battery data in an embodiment of the present application. The main operation process of the electronic device in the present application is as follows: Figure 4 As shown in the figure: First, read the battery data to be processed, then clean the battery data to be processed, and pre-process the cleaned battery data. Then, according to the analysis target, select a battery analysis model from the battery analysis model library to obtain the model table to be loaded. Then, extract and summarize the features according to the model table to be loaded (obtain the feature registration table and the event table to be executed), and perform feature calculation in descending order (according to Figure 2 The hierarchical order of the model shown in the figure is used to call the model for calculation in turn). If the calculation is normal, the result is output and the data is saved until the calculation of the model at all levels is completed.

[0067] exist Figure 4 In the process of model matching, a model table to be loaded is created, and the model table to be loaded records each battery analysis model in a hierarchical manner, such as Figure 2 As shown in the figure, different levels are connected in series to realize the connection calculation of multiple models.

[0068] When extracting and summarizing features, the input features that require the same output features are merged through the feature registration table, thereby avoiding repeated calculations of the same output features and speeding up data processing efficiency. Secondly, the various features in the feature registration table are intelligently analyzed to obtain a table of pending events. Calculations with dependencies are assigned to a calculation sequence, and different parameters in the same output feature are divided into a task block. The calculation sequence consists of task blocks. The position of the task block in the calculation sequence is consistent with the level of the output feature in the task block. When multiple levels contain the same output feature, the position of the corresponding task block in the calculation sequence is consistent with the highest level of the output feature. In this way, the independence of the calculation tasks at the same level is ensured, so that the calculation tasks at the same level can be calculated in parallel, improving the calculation efficiency.

[0069] In summary, the method of the present application includes at least the following technical features: (1) Automatic matching of battery analysis models. Different battery analysis models use different boundary conditions (constraints on input features) and cannot satisfy the calculation of any input features. This application can automatically filter and load suitable battery analysis models from the model library for calculation according to the analysis objectives.

[0070] (2) Support reprocessing of input features. When a battery analysis model is forced to participate in the calculation, and the input data required by the battery analysis model does not meet the boundary conditions of the current battery analysis model, the required input data can be reprocessed, such as interpolating or completing the required input data.

[0071] (3) Supports automatic identification of new or modified models. When a new battery analysis model developed offline is added to the model library, the model library is automatically initialized and updated, and the boundary conditions, calculation indicators and other information of the new model are identified. The model is deployed in the form of a file and can be updated online.

[0072] (4) Supports multiple models in series calculation. Some battery analysis models use raw data to directly calculate the results, while some battery analysis models need to use intermediate results, which requires models to calculate the intermediate results, and some models are used to recalculate the calculation results. Therefore, the model screening process needs to load multiple models in a loop multiple times to obtain the final complete battery analysis model.

[0073] (5) Model loading is configurable. The computing framework of the electronic device supports model loading rules, and the model to be loaded is selected from the matching models according to the loading rules.

[0074] (6) Support model characteristic parameter and threshold maintenance. When the model has characteristic parameters or thresholds, they can be modified manually or automatically, and support a model with multiple sets of characteristic parameters or thresholds to adapt to different types of batteries or application scenarios.

[0075] (7) When extracting and summarizing features, the input features that require the same output features are merged through the feature registration table, thereby avoiding repeated calculations of the same output features and speeding up data processing efficiency. (8) A single calculation process supports parallel calculations. Calculations that have a front-and-back relationship, such as dependent models, are not placed in a single calculation process to ensure that there is no dependency between all calculations in a single calculation process, so that parallel or distributed computing can be used when necessary.

[0076] (9) Model calculations are independent. The calculations of each model feature and algorithm are independent. If the feature calculation of a model is abnormal or the calculation time is too long, it will not affect the calculation of other models.

[0077] A battery data processing device provided by the present application is described below. The battery data processing device described below and the battery data processing method described above can be referenced to each other.

[0078] Figure 5 1 is a structural block diagram of a battery data processing device according to an embodiment of the present application. Figure 5 , a battery data processing device of the present application may include: An acquisition module 501 is used to acquire battery data to be processed and an analysis target for the battery data to be processed, wherein the analysis target includes at least one battery indicator; A first determining module 502, configured to determine a keyword corresponding to the analysis target according to the at least one battery indicator; A second determination module 503 is used to determine, according to the keyword, at least one battery analysis model for processing the data to be processed in a battery analysis model library, wherein the battery analysis model library includes a plurality of battery analysis models, and different battery analysis models are used to implement different battery analysis functions; The processing module 504 processes the battery data to be processed by using the at least one battery analysis model to obtain the value of each battery indicator in the analysis target.

[0079] According to a battery data processing device 500 provided by the present application, the second determination module 503 includes: A first determination submodule, configured to determine, in the battery analysis model library, a battery analysis model whose model name contains any of the keywords as a first type of model; The first screening submodule is used to screen the first type of models according to the constraint conditions and output characteristics of the input characteristics of each of the battery analysis models in the first type of models to obtain the at least one battery analysis model.

[0080] According to a battery data processing device 500 provided by the present application, a battery analysis model includes at least one battery analysis sub-model, and different battery analysis sub-models in the same battery analysis model are used to implement different data processing functions; the first screening sub-module includes: A second screening submodule, configured to screen the first type of models according to the constraints of the input features and the output features of each of the battery analysis models in the first type of models to obtain a second type of models; A second determination submodule, configured to determine the data processing function lacking in the second type of model according to the output characteristics of the battery analysis submodels included in each of the battery analysis models in the second type of model; A third determination submodule, configured to determine a third type of model in the battery analysis model library for implementing the missing data processing function; The fourth determination submodule is used to determine the at least one battery analysis model according to the second type of model and the third type of model.

[0081] According to a battery data processing device 500 provided by the present application, the first type of model includes a battery analysis model that is mandatory to use, and the battery data processing device 500 further includes: The adjustment module is used to adjust the data to be processed if the constraint conditions of the input features of the battery analysis model that is forced to be used do not meet the preset conditions until the constraint conditions of the input features of the battery analysis model that is forced to be used meet the preset conditions.

[0082] According to a battery data processing device 500 provided by the present application, a battery analysis model includes at least one battery analysis sub-model, and different battery analysis sub-models in the same battery analysis model are used to implement different data processing functions; the processing module 504 includes: A fifth determination submodule, configured to determine a connection relationship between each of the battery analysis sub-models according to an output feature of each of the battery analysis models and an output feature of a battery analysis sub-model included in each of the battery analysis models; A generating submodule, configured to generate a table of events to be executed according to the output characteristics of each of the battery analysis submodels and the connection relationship, wherein the table of events to be executed includes each event to be executed when achieving the analysis target and the execution order of each event; The execution submodule is used to execute each event in the to-be-executed event table according to the execution order of each event through each battery analysis submodel, so as to realize the processing of the to-be-processed battery data.

[0083] According to a battery data processing device 500 provided by the present application, the fifth determination submodule includes: A deduplication submodule, used for deduplicating battery analysis submodels with the same functions in each of the battery analysis submodels according to the output characteristics of each of the battery analysis submodels; The sixth determination submodule is used to determine the connection relationship between each of the battery analysis sub-models according to the output characteristics of each of the battery analysis models and the output characteristics of each of the battery analysis sub-models obtained after deduplication.

[0084] According to a battery data processing device 500 provided by the present application, the acquisition module 501 includes: The acquisition submodule is used to obtain the historical battery data corresponding to the battery to be analyzed; The preprocessing submodule is used to preprocess the historical battery data to obtain preprocessed battery data, where the preprocessed battery data is the battery data to be processed.

[0085] Figure 6 is a schematic diagram of the physical structure of an electronic device shown in an embodiment of the present application, such as Figure 6 As shown, the electronic device may include: a processor 610, a communication interface 620, a memory 630 and a communication bus 640, wherein the processor 610, the communication interface 620 and the memory 630 communicate with each other through the communication bus 640. The processor 610 may call the logic instructions in the memory 630 to execute a battery data processing method, the method comprising: obtaining battery data to be processed and an analysis target for the battery data to be processed, wherein the analysis target includes at least one battery indicator;

[0086] In addition, the logic instructions in the above-mentioned memory 630 can be implemented in the form of a software functional unit and can be stored in a computer-readable storage medium when it is sold or used as an independent product. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc. Various media that can store program codes.

[0087] On the other hand, the present application further provides a computer program product, the computer program product includes a computer program, the computer program can be stored on a non-transitory computer-readable storage medium, when the computer program is executed by a processor, the computer can execute a battery data processing method provided by the above methods, the method comprising: obtaining battery data to be processed and an analysis target for the battery data to be processed, the analysis target including at least one battery indicator; Determining a keyword corresponding to the analysis target according to the at least one battery indicator; According to the keyword, determining at least one battery analysis model for processing the data to be processed in a battery analysis model library, wherein the battery analysis model library includes a plurality of battery analysis models, and different battery analysis models are used to implement different battery analysis functions; The battery data to be processed is processed by using the at least one battery analysis model to obtain the value of each of the battery indicators in the analysis target.

[0088] In another aspect, the present application further provides a non-transitory computer-readable storage medium having a computer program stored thereon, wherein the computer program is implemented when executed by a processor to execute a battery data processing method provided by the above methods, the method comprising: obtaining battery data to be processed and an analysis target for the battery data to be processed, wherein the analysis target includes at least one battery indicator; Determining a keyword corresponding to the analysis target according to the at least one battery indicator; According to the keyword, determining at least one battery analysis model for processing the data to be processed in a battery analysis model library, wherein the battery analysis model library includes a plurality of battery analysis models, and different battery analysis models are used to implement different battery analysis functions; The battery data to be processed is processed by using the at least one battery analysis model to obtain the value of each of the battery indicators in the analysis target.

[0089] The device embodiments described above are merely illustrative, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. Ordinary technicians in this field can understand and implement it without paying creative labor.

[0090] Through the description of the above implementation methods, 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, can also be implemented by hardware. Based on this understanding, the above technical solution is essentially or the part that contributes to the prior art can be embodied in the form of a software product, and the computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a disk, an optical disk, etc., including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0091] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A battery data processing method, characterized in that: include: Acquire battery data to be processed and an analysis target for the battery data to be processed, wherein the analysis target includes at least one battery indicator; Determining a keyword corresponding to the analysis target according to the at least one battery indicator; According to the keyword, determining at least one battery analysis model for processing the data to be processed in a battery analysis model library, wherein the battery analysis model library includes a plurality of battery analysis models, and different battery analysis models are used to implement different battery analysis functions; The battery data to be processed is processed by using the at least one battery analysis model to obtain the value of each of the battery indicators in the analysis target.

2. The battery data processing method according to claim 1, characterized in that: The step of determining, according to the keyword, at least one battery analysis model for processing the data to be processed in a battery analysis model library comprises: In the battery analysis model library, determining a battery analysis model whose model name contains any of the keywords as a first type of model; According to the constraint conditions of the input features and the output features of each of the battery analysis models in the first category of models, the first category of models is screened to obtain the at least one battery analysis model.

3. The battery data processing method according to claim 2, characterized in that: One of the battery analysis models includes at least one battery analysis sub-model, and different battery analysis sub-models in the same battery analysis model are used to implement different data processing functions. The first type of model is screened according to the constraints and output characteristics of the input characteristics of each of the battery analysis models in the first type of model to obtain the at least one battery analysis model, including: According to the constraint conditions and output characteristics of the input characteristics of each of the battery analysis models in the first type of models, the first type of models are screened to obtain a second type of model; Determining the data processing function lacking in the second type of model according to the output characteristics of the battery analysis sub-models included in each of the battery analysis models in the second type of model; Determining a third type of model in the battery analysis model library for implementing the missing data processing function; The at least one battery analysis model is determined according to the second type of model and the third type of model.

4. The battery data processing method according to claim 2, characterized in that: The first type of models includes a battery analysis model that is mandatory to use. After screening the first type of models, the method further includes: If the constraint conditions of the input features of the battery analysis model that is forced to be used do not meet the preset conditions, the data to be processed is adjusted until the constraint conditions of the input features of the battery analysis model that is forced to be used meet the preset conditions.

5. The battery data processing method according to claim 1, characterized in that: One of the battery analysis models includes at least one battery analysis sub-model, and different battery analysis sub-models in the same battery analysis model are used to implement different data processing functions; The processing of the battery data to be processed by using the at least one battery analysis model includes: Determining a connection relationship between each of the battery analysis sub-models according to an output feature of each of the battery analysis models and an output feature of a battery analysis sub-model included in each of the battery analysis models; Generate a table of events to be executed according to the output characteristics of each of the battery analysis sub-models and the connection relationship, wherein the table of events to be executed includes various events that need to be executed when achieving the analysis goal and the execution order of each of the events; Each event in the to-be-executed event table is executed in the execution order of each event through each of the battery analysis sub-models, so as to realize the processing of the to-be-processed battery data.

6. The battery data processing method according to claim 5, characterized in that: Determining the connection relationship between each of the battery analysis sub-models according to the output characteristics of each of the battery analysis models and the output characteristics of the battery analysis sub-models contained in each of the battery analysis models includes: Deduplication processing is performed on battery analysis sub-models having the same function in each of the battery analysis sub-models according to output characteristics of each of the battery analysis sub-models; The connection relationship between the battery analysis sub-models is determined according to the output characteristics of each battery analysis model and the output characteristics of each battery analysis sub-model obtained after deduplication.

7. A battery data processing device, characterized in that: include: An acquisition module, used to acquire battery data to be processed and an analysis target for the battery data to be processed, wherein the analysis target includes at least one battery indicator; A first determination module, configured to determine a keyword corresponding to the analysis target according to the at least one battery indicator; A second determination module is used to determine, according to the keyword, at least one battery analysis model for processing the data to be processed in a battery analysis model library, wherein the battery analysis model library includes a plurality of battery analysis models, and different battery analysis models are used to implement different battery analysis functions; The processing module processes the battery data to be processed through the at least one battery analysis model to obtain the value of each battery indicator in the analysis target.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the battery data processing method according to any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, a battery data processing method according to any one of claims 1 to 6 is implemented.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, a battery data processing method according to any one of claims 1 to 6 is implemented.

Images