Battery management method and device for target object, equipment and product

Abstract

The embodiment of the invention discloses a battery management method and device for a target object, equipment and a product. The method comprises the following steps: receiving time sequence data sent by a battery management module; performing time windowing on the time sequence data according to the length of first preset time to obtain a plurality of pieces of window data; determining the state of each piece of window data based on historical time sequence data, wherein the historical time sequence data is time sequence data sent by the battery management module at historical time; and uploading each piece of window data to a cloud server, so that the cloud server stores each piece of window data to a distributed storage system or a memory database according to the state of each piece of window data. According to the battery management method of the target object, provided by the invention, through combination of big data edge calculation and cloud and a lossy compression mode, uncompressed storage is carried out on abnormal data at the same time, and efficient compression and reliable and hybrid storage of battery data are realized.

Description

Technical Field

[0001] The present invention relates to the field of data processing, and in particular to a battery management method, device, equipment and product for a target object. Background Art

[0002] With the rapid development of flying car technology, batteries, as the core energy source of flying cars, play a crucial role in their endurance and safety. During operation, batteries generate a large amount of data with significant time-series characteristics, such as voltage, current, temperature, and charge and discharge rates. Efficiently collecting, storing, analyzing, and applying this time-series data has become a key research direction for flying car battery management systems.

[0003] Current battery management technologies primarily address four key areas: data acquisition, transmission, processing, and storage. During data acquisition, sensor networks monitor key battery parameters in real time, providing raw data for subsequent processing. Transmission relies on mature communication technologies such as 4G to ensure data flow between in-vehicle systems and the cloud. During the processing phase, onboard computing units and cloud servers conduct preliminary and in-depth analysis to unlock the data's value. For storage, relational or non-relational databases are used to address the storage needs of both structured and unstructured data. However, significant challenges exist in high-frequency real-time processing, long-term storage, and system reliability. For example, they cannot meet the efficient storage requirements of flying car battery data. Furthermore, existing technologies often utilize lossy compression, which cannot significantly reduce data storage space, resulting in low storage efficiency. Furthermore, data integrity and reliability cannot be guaranteed under abnormal circumstances.

[0004] Therefore, it is particularly important to develop a battery management method, device, equipment and product for target objects that can achieve efficient compression and storage of battery data while ensuring the integrity and reliability of abnormal data. Summary of the Invention

[0005] In order to solve the above technical problems, the present application provides a battery management method, device, equipment and product for the target object. By combining big data edge computing with the cloud, and using lossy compression, while performing non-compressed storage of abnormal data, it can solve the current problem of lack of battery management methods, devices, equipment and products for the target object that can realize efficient compression and storage of battery data while ensuring the integrity and reliability of abnormal data.

[0006] The technical solutions provided in this application are as follows:

[0007] In one aspect, the present application provides a battery management method for a target object, wherein the target object includes a battery management module, and the method includes:

[0008] Receive timing data sent by the battery management module;

[0009] Performing time windowing on the time series data according to a first preset time length to obtain a plurality of window data;

[0010] Determine the state of each window data based on historical time series data, where the historical time series data is time series data sent by the battery management module at a historical time;

[0011] Each of the window data is uploaded to the cloud server respectively, so that the cloud server stores each of the window data in a distributed storage system or a memory database according to the status of each of the window data.

[0012] In some optional implementations, the time series data is time-windowed by a first preset time length to obtain a plurality of window data, including:

[0013] Using the time series data as current time series data;

[0014] Windowing the current time series data by a first preset time length to obtain current window data;

[0015] Storing the current window data in a key-value storage library, and deleting the current window data from the current time series data;

[0016] Reuse the remaining time series data in the current time series data as the current time series data;

[0017] Repeat the step of windowing the current time series data by the length of the first preset time, and re-using the remaining time series data in the current time series data as the current time series data, until no time series data exists.

[0018] In some optional implementations, determining the state of each window data based on historical time series data includes:

[0019] Obtain historical time series data and preset models;

[0020] Based on the historical time series data and the state labels corresponding to the historical time series data, the preset model is trained for state prediction to obtain a target model;

[0021] Each of the window data is input into the target model for state prediction to obtain the state of each of the window data.

[0022] In some optional implementations, uploading each window data to the cloud server separately includes:

[0023] When the state of each window data is normal, compressing each window data and uploading it to the cloud server;

[0024] When the state of each of the window data is an abnormal state, each of the window data, the preceding window data of each of the window data, and the succeeding window data of each of the window data are used as window abnormality storage data, and the window abnormality storage data are uploaded to the cloud server; the preceding window data of each of the window data is the window data of the previous time sequence of the time sequence of each of the window data; the succeeding window data of each of the window data is the window data of the next time sequence of the time sequence of each of the window data.

[0025] In some optional implementations, when the state of each window data is normal, compressing each window data and uploading it to the cloud server includes:

[0026] When the state of each window data is normal, dividing each window data by a second preset time interval to obtain sub-window data corresponding to each window data, wherein the second preset time is less than the first preset time;

[0027] determining an average value of sub-window data corresponding to each window data to compress each window data;

[0028] The average value of the sub-window data corresponding to each window data is used as the normal storage data of the window;

[0029] The window stores data normally and uploads it to the cloud server.

[0030] In some optional embodiments, the method further comprises:

[0031] The cloud server is used for storing the normal storage data of each window in the distributed storage system when the state of the window data is normal;

[0032] And when the state of each window data is an abnormal state, the abnormal storage data of the window is stored in the memory database.

[0033] On the other hand, the present application provides a battery management device for a target object, the battery management device for the target object comprising:

[0034] A data acquisition module, configured to receive timing data sent by the battery management module;

[0035] an edge computing module, the edge computing module being configured to time-window the time series data by a first preset time length to obtain a plurality of window data; and being configured to determine a state of each window data based on historical time series data, the historical time series data being time series data historically sent by the battery management module;

[0036] A data transmission module, the data transmission module is used to upload each window data to the cloud server respectively;

[0037] A cloud storage module is used to store each window data in a distributed storage system or a memory database according to the status of each window data.

[0038] On the other hand, the present application provides an electronic device, which includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement a battery management method for a target object as described in any one of the above embodiments.

[0039] On the other hand, the present application provides a computer-readable storage medium, which stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by a processor to implement a battery management method for a target object as described in any one of the above embodiments.

[0040] On the other hand, the present application provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the battery management method of the target object as described in any one of the above embodiments.

[0041] The battery management method of the target object provided by the present application includes: receiving time series data sent by a battery management module; time-windowing the time series data with a first preset time length to obtain a plurality of window data; determining the status of each window data based on historical time series data, wherein the historical time series data is the time series data sent by the battery management module at a historical time; uploading each window data to a cloud server respectively, so that the cloud server stores each window data in a distributed storage system or a memory database according to the status of each window data. By time-windowing the time series data and determining the status of each window data, and combining with the cloud server, efficient compression of the battery data is achieved, and the data is stored in a distributed storage system or a memory database according to the status of the data, thereby achieving reliable and hybrid storage of the battery data. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] In order to more clearly illustrate the technical solutions in the embodiments of the present application, 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 application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0043] Figure 1 is a step diagram of a battery management method for a target object according to an embodiment of the present invention;

[0044] Figure 2 is a schematic diagram of a battery management device for a target object according to an embodiment of the present invention;

[0045] Figure 3 FIG. 4 is a flow chart of a battery management method for a target object according to an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making any creative work are within the scope of protection of this application.

[0047] References to "one embodiment" or "an embodiment" herein refer to specific features, structures, or characteristics that may be included in at least one implementation of the present application. Throughout the description of this application, it should be understood that the terms "upper," "lower," "top," and "bottom," etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings and are intended solely for ease of description and simplification. They do not indicate or imply that the devices or components referred to must have, be constructed, or operate in a specific orientation, and are therefore not to be construed as limiting the present application. Furthermore, the terms "first" and "second" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly specifying the number of the technical features referred to. Thus, a feature designated "first" or "second" may explicitly or implicitly include one or more of the features. Furthermore, the terms "first," "second," etc. are used to distinguish similar objects and are not necessarily used to describe a specific order or sequential sequence. It should be understood that the terms used in this manner are interchangeable where appropriate, such that the embodiments of the present application described herein can be implemented in an order other than that illustrated or described herein.

[0048] When a numerical range is disclosed herein, the above range is considered to be continuous and includes the minimum and maximum values ​​of the range, as well as every value between such minimum and maximum values. Further, when a range refers to an integer, every integer between the minimum and maximum values ​​of the range is included. In addition, when multiple ranges are provided to describe a feature or characteristic, the ranges can be merged. In other words, unless otherwise indicated, all ranges disclosed herein should be understood to include any and all subranges included therein. For example, a specified range from "1 to 10" should be considered to include any and all subranges between a minimum of 1 and a maximum of 10. Exemplary subranges of the range 1 to 10 include, but are not limited to, 1 to 6.1, 3.5 to 7.8, 5.5 to 10, etc.

[0049] Existing battery management technologies cannot meet the efficient storage requirements of flying car battery data. Furthermore, existing technologies often use lossy compression, which cannot significantly reduce data storage space, resulting in low storage efficiency. Furthermore, they cannot guarantee data integrity and reliability under abnormal circumstances. Therefore, to achieve efficient compression and storage of battery data while ensuring the integrity and reliability of abnormal data, this application provides a target battery management method.

[0050] See also Figure 1 , Figure 1 1 is a step diagram of a battery management method for a target object according to an embodiment of the present invention. On the one hand, the present application provides a battery management method for a target object, wherein the target object includes a battery management module, and the method includes:

[0051] S101. Receive timing data sent by the battery management module.

[0052] Optionally, the target object may be a flying car, which is provided with a battery management module and a central data computer. The battery management module is used to store and send battery data information, and the central data computer is used to receive battery data sent by the battery management module.

[0053] Optionally, the battery data sent by the battery management module includes time series data with obvious time series characteristics, such as voltage, current, temperature, and charge and discharge rate. The central data computer receives the time series data through canFD, and the processing unit in the central data computer sends the above time series data to the avionics system of the flying car. At the same time, the time series data is sent to the Kafka message queue through the processing unit.

[0054] Optionally, the data acquisition module can perform preliminary processing on the collected time series data, such as removing duplicate data, performing simple data cleaning, etc., to reduce the amount of data and improve the efficiency of subsequent processing.

[0055] Optionally, the flying car's central data computer can quickly and accurately collect key data during battery operation by receiving time series data through canFD, providing a reliable basis for subsequent data processing and analysis, and realizing real-time processing and analysis of battery data; at the same time, the collected time series data is not only sent to the avionics system, but also sent to the Kafka data source operator message queue, providing a timely and accurate data source for subsequent edge computing, data transmission and cloud storage, ensuring the efficiency and consistency of the entire data management process.

[0056] S102. Time-window the time series data according to a first preset time length to obtain a plurality of window data.

[0057] Optionally, the real-time big data processing module in the central data computer receives the time series data in the message queue of the Kafka data source operator, and the streaming computing window (flink) can time-window the time series data.

[0058] In an optional embodiment, the time series data is time-windowed by a first preset time length to obtain a plurality of window data, including:

[0059] Using the time series data as current time series data;

[0060] Windowing the current time series data by a first preset time length to obtain current window data;

[0061] Storing the current window data in a key-value storage library, and deleting the current window data from the current time series data;

[0062] Reuse the remaining time series data in the current time series data as the current time series data;

[0063] Repeat the step of windowing the current time series data by the length of the first preset time, and re-using the remaining time series data in the current time series data as the current time series data, until no time series data exists.

[0064] Optionally, the length of the first preset time can be set according to actual business needs and is not limited here. For example, the length of the first preset time is 3 minutes. The flink window can divide the data into time windows and perform data calculations according to a 3-minute window.

[0065] Optionally, the key-value storage library is RocksDB. Every time the Flink window is divided into 3-minute window data, the window data is temporarily stored in RocksDB.

[0066] Optionally, you can also divide the Flink window into several 3-minute window data and then temporarily store the window data in RocksDB.

[0067] Dividing time series data into three-minute windows enables real-time processing and rapid response to battery data. Timely data processing can quickly capture abnormal fluctuations in battery performance, such as sudden voltage changes and sharp temperature increases, providing immediate decision support for the safe operation of flying cars. Furthermore, by windowing time series data, large amounts of continuous data can be split into multiple smaller batches for processing, enabling more rational allocation of data processing and storage resources, effectively reducing data processing complexity and improving data processing efficiency.

[0068] Optionally, a machine learning model can be used to optimize the window partitioning strategy and dynamically adjust the window size to adapt to different data change frequencies.

[0069] S103. Determine the state of each window data based on historical time series data, where the historical time series data is time series data sent by the battery management module at historical times.

[0070] In an optional embodiment, determining the state of each window data based on historical time series data includes:

[0071] Obtain historical time series data and preset models;

[0072] Based on the historical time series data and the state labels corresponding to the historical time series data, the preset model is trained for state prediction to obtain a target model;

[0073] Each of the window data is input into the target model for state prediction to obtain the state of each of the window data.

[0074] Optionally, after receiving the window data, Flink uses AsyncIO to asynchronously call an anomaly detection model and combines it with historical data to determine the status of the window data. The anomaly detection model is a data processing and prediction model.

[0075] Optionally, based on the historical time series data and the state labels corresponding to the historical time series data, the preset model is trained for state prediction to obtain a target model, which may specifically include the following steps: collecting historical operating data of the flying car battery, including key parameters such as voltage, current, temperature, and charge and discharge rate, which have obvious time series characteristics; labeling each section of historical time series data with a corresponding state label; and performing preprocessing operations such as cleaning and normalization on the collected historical time series data to ensure the quality and consistency of the data and make it suitable for model training.

[0076] Furthermore, a machine learning model suitable for processing time series data is selected, and the model's training objective is defined, namely, to enable the model to accurately predict the corresponding state label based on the input time series data. The preprocessed historical time series data is then fed into the preset model for training. During training, the model learns the time series features and patterns in the data and establishes a mapping relationship between these features and state labels. During training, the model is used to predict the state of the input historical time series data, obtaining a predicted state. The difference between the predicted state and the actual state label is then calculated, typically using a loss function to quantify this difference, resulting in loss data. Based on the loss data, an optimization algorithm is used to adjust the parameters of the preset model to reduce the difference between the predicted state and the actual state label. This process is repeated until the model converges, meaning that the loss data no longer decreases significantly and the model's predictive performance reaches a relatively high level, thus obtaining the target model.

[0077] Optionally, the status labels may include normal status, abnormal status (such as overvoltage, undervoltage, overheating, etc.), and the status labels corresponding to the historical time series data may be marked based on actual battery operation conditions or expert experience.

[0078] Optionally, during the operation of the flying car, the battery's time series data is acquired in real time through the data acquisition module, and after dividing the window data into 3-minute time intervals, the collected window data can be subjected to the same preprocessing operations as during training, including data cleaning and normalization, to ensure that the data format and quality are consistent with the training data.

[0079] The anomaly detection model can quickly identify anomalies in the current battery operating status based on features extracted from a large amount of historical data, thereby improving system reliability. In addition, the use of model prediction can also combine data features to accurately locate the source of the fault and accurately lock the faulty module, which greatly facilitates subsequent troubleshooting and maintenance work, shortens maintenance time, reduces maintenance costs, and improves the operational efficiency of flying cars.

[0080] Optionally, use AsyncIO to asynchronously call a trained anomaly detection model and feed the preprocessed window data into the anomaly detection model. The anomaly detection model then determines the state of the window data based on the learned time series data features and corresponding state labels, and outputs a predicted state. The predicted state output by the model is compared with the set threshold or rule to determine whether the battery state is normal. If the predicted state is abnormal, an alarm signal is promptly issued, prompting the system to take appropriate measures, such as adjusting the flight state or performing troubleshooting.

[0081] Asynchronous calls allow the program to continue executing other tasks, such as data collection and preprocessing, while waiting for model prediction results. Once the model prediction is complete, the results are immediately available for analysis and decision-making, significantly shortening the total time from data collection to status judgment. This enables a response to battery status changes in seconds or even milliseconds, allowing for timely detection and resolution of abnormal fluctuations in battery operation to ensure flight safety. At the same time, asynchronous calls allow the system to fully utilize idle resources to handle other tasks, such as data collection and preprocessing, while waiting for prediction results, avoiding resource waste, improving overall resource utilization, and enhancing system performance.

[0082] S104. Upload each window data to the cloud server respectively, so that the cloud server stores each window data in a distributed storage system or a memory database according to the status of each window data.

[0083] In an optional embodiment, uploading each window data to the cloud server separately includes:

[0084] When the state of each window data is normal, compressing each window data and uploading it to the cloud server;

[0085] When the state of each of the window data is an abnormal state, each of the window data, the preceding window data of each of the window data, and the succeeding window data of each of the window data are used as window abnormality storage data, and the window abnormality storage data are uploaded to the cloud server; the preceding window data of each of the window data is the window data of the previous time sequence of the time sequence of each of the window data; the succeeding window data of each of the window data is the window data of the next time sequence of the time sequence of each of the window data.

[0086] Optionally, when the state of the current window data is an abnormal state, the current window data, the previous window data of the current window data, the subsequent window data of the current window data, and the abnormal signal recognized by the model are all sent down and uploaded to the cloud server, that is, when the state of the window data is an abnormal state, the data is not compressed and stored.

[0087] Uncompressed anomaly data accurately reflects fault characteristics, providing key clues for fault diagnosis and location. By comparing the current anomaly data with normal data and analyzing the correlation between the preceding and subsequent window data, technicians can quickly trace the anomaly's starting point and development path, effectively locating the root cause of the fault and implementing targeted solutions. This data also provides high-quality training data for machine learning models. This uncompressed anomaly data, along with its corresponding preceding and subsequent window data, helps diagnostic models better learn fault patterns and characteristics, improving their accuracy and reliability.

[0088] In an optional embodiment, when the state of each window data is normal, compressing each window data and uploading it to the cloud server includes:

[0089] When the state of each window data is normal, dividing each window data by a second preset time interval to obtain sub-window data corresponding to each window data, wherein the second preset time is less than the first preset time;

[0090] determining an average value of sub-window data corresponding to each window data to compress each window data;

[0091] The average value of the sub-window data corresponding to each window data is used as the normal storage data of the window;

[0092] The window stores data normally and uploads it to the cloud server.

[0093] Optionally, the length of the second preset time can be set according to actual business needs and is not limited here. For example, the length of the second preset time can be 10 seconds. When the status of the current window data is normal, the average value of the current window data is calculated every 10 seconds and uploaded to the cloud server to compress each of the window data.

[0094] Optionally, the current window data contains data corresponding to multiple parameters such as temperature, voltage, current, etc. When the current window data is in a normal state, the average value of each parameter data is calculated every 10 seconds and uploaded to the cloud server.

[0095] For normal window data, by calculating the average value every 10 seconds, the amount of data that needs to be stored and processed can be greatly reduced, saving storage space. Reducing the amount of data can also significantly reduce the computational burden of the data processing module and improve storage efficiency.

[0096] In an optional embodiment, the method further includes:

[0097] The cloud server is used for storing the normal storage data of each window in the distributed storage system when the state of the window data is normal;

[0098] And when the state of each window data is an abnormal state, the abnormal storage data of the window is stored in the memory database.

[0099] Optionally, the cloud server receives the abnormal window data in the message queue, stores it in the memory database, and displays the alarm on the data alarm interface, while displaying the current abnormal data details; performs lossy compression on the normal window data and stores it in distributed storage.

[0100] Alternatively, in-memory databases offer fast read and write speeds, but their storage capacity is relatively limited and their cost is high. Storing abnormal window data in an in-memory database fully leverages its high performance, storing the most critical data and achieving efficient use of storage resources. Distributed storage systems, on the other hand, are typically built using inexpensive storage devices, offering high storage capacity and scalability. Storing lossy compressed normal window data in distributed storage effectively reduces storage costs while meeting the storage needs of large amounts of normal data.

[0101] Storing abnormal window data in an in-memory database ensures the integrity and originality of this critical data, helping technicians accurately diagnose problems and take effective repair measures. Lossy compression of normal window data before storing it in distributed storage reduces the amount of data while retaining its main features and trends, significantly saving storage space. This hybrid storage method improves the overall performance and reliability of the system.

[0102] On the other hand, this application provides a battery management device for a target object, see Figure 2 , Figure 2 1 is a schematic diagram of a battery management device for a target object according to an embodiment of the present invention, wherein the battery management device for the target object includes:

[0103] A data acquisition module 310, configured to receive timing data sent by the battery management module;

[0104] an edge computing module 320 configured to time-window the time series data by a first preset time length to obtain a plurality of window data; and to determine a state of each window data based on historical time series data, where the historical time series data is time series data historically sent by the battery management module;

[0105] A data transmission module 330, the data transmission module 330 is used to upload each window data to the cloud server;

[0106] The cloud storage module 340 is used to store each window data in a distributed storage system or a memory database according to the status of each window data.

[0107] Optionally, the data transmission module uploads the processed window data to the cloud server by utilizing the fifth-generation mobile communication technology, low-orbit satellites or dedicated wireless communication protocols to achieve transmission encryption and ensure data security.

[0108] The following is an overall description of the battery management method for the target object provided in the embodiment of the present application:

[0109] See also Figure 3 , Figure 3 This is a flow chart of a target object battery management method according to an embodiment of the present invention. The target object battery management method includes: a Kafka data source operator receives time series data sent by a battery management module; a Flink window divides the data into time windows, performs data calculations in 3-minute windows, and temporarily stores the window data in RocksDB;

[0110] Use AsyncIO to asynchronously call the trained anomaly detection model and input the window data into the anomaly detection model. The anomaly detection model will judge the state of the window data based on the learned time series data features and corresponding state labels and output the predicted state.

[0111] For normal window data, the average value is calculated every 10 seconds and uploaded to the cloud server; for abnormal window data, the current window data, the previous window data of the current window data, and the subsequent window data of the current window data are all uploaded to the cloud server; the cloud server receives the abnormal window data, stores it in the memory database, and displays an alarm on the data alarm interface, while displaying the current abnormal data details; the normal window data is losslessly compressed and stored in the distributed storage system.

[0112] The battery management method for the target object provided by the present application includes: receiving the timing data sent by the battery management module; time-windowing the timing data with the length of a first preset time to obtain a plurality of window data; determining the status of each of the window data based on the historical timing data, the historical timing data being the timing data sent by the battery management module at a historical time; uploading each of the window data to the cloud server respectively, so that the cloud server stores each of the window data in a distributed storage system or a memory database according to the status of each of the window data. By combining big data edge computing with the cloud, efficient compression of battery data is achieved, and the data is stored in a distributed storage system or a memory database according to the status of the data, thereby achieving reliable and hybrid storage of battery data. The battery management method for the target object provided by the present application has the following beneficial effects:

[0113] (1) By combining big data edge computing with cloud processing, millisecond-level data collection and processing are achieved. This efficient real-time performance ensures that flying car battery data can be monitored and analyzed in real time, thereby quickly responding to changes in battery status, promptly discovering and handling potential abnormalities, and ensuring the safe operation of flying cars.

[0114] (2) Normal window data is stored in a distributed storage system after lossy compression, which significantly reduces the amount of data and reduces storage costs. At the same time, abnormal window data is stored in an uncompressed form in the memory database, ensuring the integrity and rapid accessibility of key data. The use of data compression and hybrid storage solutions achieves efficient data storage.

[0115] (3) While compressing and storing normal window data, edge computing technology is used to perform non-compressed storage of abnormal data, ensuring the integrity and reliability of the abnormal data.

[0116] In an optional embodiment, the present application provides an electronic device comprising a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement a battery management method for a target object as described in any one of the above embodiments.

[0117] In an optional embodiment, the present application provides a computer-readable storage medium, which stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by a processor to implement a battery management method for a target object as described in any of the above embodiments.

[0118] In an optional embodiment, the present application provides a computer program product, which includes a computer program. When the computer program is executed by a processor, it implements the battery management method for the target object as described in any one of the above embodiments.

[0119] The above description is merely an optional embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A battery management method for a target object, characterized in that: The target object includes a battery management module, and the method includes: Receive timing data sent by the battery management module; Performing time windowing on the time series data according to a first preset time length to obtain a plurality of window data; Determine the state of each window data based on historical time series data, where the historical time series data is time series data sent by the battery management module at a historical time; Each of the window data is uploaded to the cloud server respectively, so that the cloud server stores each of the window data in a distributed storage system or a memory database according to the status of each of the window data.

2. The battery management method of the target object according to claim 1, characterized in that: The time series data is time-windowed according to the length of the first preset time to obtain a plurality of window data, including: Using the time series data as current time series data; Windowing the current time series data by a first preset time length to obtain current window data; Storing the current window data in a key-value storage library, and deleting the current window data from the current time series data; Reuse the remaining time series data in the current time series data as the current time series data; Repeat the step of windowing the current time series data by the length of the first preset time, and re-using the remaining time series data in the current time series data as the current time series data, until no time series data exists.

3. The battery management method of the target object according to claim 1, characterized in that: The determining the state of each window data based on the historical time series data includes: Obtain historical time series data and preset models; Based on the historical time series data and the state labels corresponding to the historical time series data, the preset model is trained for state prediction to obtain a target model; Each of the window data is input into the target model for state prediction to obtain the state of each of the window data.

4. The battery management method of the target object according to claim 1, characterized in that: The step of uploading each window data to the cloud server comprises: When the state of each window data is normal, compressing each window data and uploading it to the cloud server; When the state of each of the window data is an abnormal state, each of the window data, the preceding window data of each of the window data, and the succeeding window data of each of the window data are used as window abnormality storage data, and the window abnormality storage data are uploaded to the cloud server; the preceding window data of each of the window data is the window data of the previous time sequence of the time sequence of each of the window data; the succeeding window data of each of the window data is the window data of the next time sequence of the time sequence of each of the window data.

5. The battery management method of the target object according to claim 4, characterized in that: When the state of each window data is normal, compressing each window data and uploading it to the cloud server includes: When the state of each window data is normal, dividing each window data by a second preset time interval to obtain sub-window data corresponding to each window data, wherein the second preset time is less than the first preset time; determining an average value of sub-window data corresponding to each window data to compress each window data; The average value of the sub-window data corresponding to each window data is used as the normal storage data of the window; The window stores data normally and uploads it to the cloud server.

6. The battery management method of the target object according to claim 5, characterized in that: The method further comprises: The cloud server is used for storing the normal storage data of each window in the distributed storage system when the state of the window data is normal; And when the state of each window data is an abnormal state, the abnormal storage data of the window is stored in the memory database.

7. A battery management device for a target object, characterized in that: The battery management device of the target object includes: A data acquisition module, configured to receive timing data sent by the battery management module; an edge computing module, the edge computing module being configured to time-window the time series data by a first preset time length to obtain a plurality of window data; and being configured to determine a state of each window data based on historical time series data, the historical time series data being time series data historically sent by the battery management module; A data transmission module, the data transmission module is used to upload each window data to the cloud server respectively; A cloud storage module is used to store each window data in a distributed storage system or a memory database according to the status of each window data.

8. An electronic device, characterized in that: The electronic device includes a processor and a memory, wherein the memory stores at least one instruction or at least one program, and the at least one instruction or at least one program is loaded and executed by the processor to implement the battery management method for the target object according to any one of claims 1 to 7.

9. A computer-readable storage medium, characterized in that The computer-readable storage medium stores at least one instruction or at least one program, and the at least one instruction or the at least one program is loaded and executed by a processor to implement the battery management method for the target object according to any one of claims 1 to 7.

10. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the battery management method for a target object according to any one of claims 1 to 7 is implemented.

Images