Battery fault detection system, method and apparatus

Abstract

This application provides a battery fault detection system, method, and device. The system includes: a first data warehouse for storing historical battery operating data; a fault detection module for extracting features from the historical battery operating data in the first data warehouse to obtain corresponding fault detection feature data; a batch calculation module for performing batch calculations on the fault detection feature data using a batch calculation engine to obtain corresponding intermediate parameters; and a stream computing platform for using a stream computing engine to acquire the intermediate parameters and real-time battery operating data in real time for fusion calculation to obtain the corresponding battery fault detection result.

Description

Technical Field

[0001] This application relates to the field of battery technology, and in particular to a battery fault detection system, method and device. Background Technology

[0002] As the main power source for new energy vehicles, batteries may frequently malfunction after prolonged operation, affecting the safe operation of these vehicles. Therefore, fault detection of batteries is particularly important.

[0003] Because power battery faults are strongly correlated with time-series factors, current power battery fault identification typically requires calculating data over a very long period to ensure the reliability of warning results. However, real-time detection is equally crucial for power battery fault detection; therefore, balancing reliability and real-time performance is a pressing issue that needs to be addressed in this field. Summary of the Invention

[0004] This application provides a battery fault detection system, method, and device that can balance the accuracy and real-time performance of battery fault detection.

[0005] In a first aspect, embodiments of this application propose a battery fault detection system, comprising: a first data warehouse for storing historical battery operating data; a fault detection module for extracting features from the corresponding historical battery operating data in the first data warehouse to obtain corresponding fault detection feature data; a batch calculation module for performing batch calculations on the fault detection feature data through a batch calculation engine to obtain corresponding intermediate parameters; and a stream computing platform for performing real-time fusion calculations on the intermediate parameters and real-time battery operating data through a stream computing engine to obtain the corresponding battery fault detection result.

[0006] According to the embodiments of this application, the first data warehouse can store massive amounts of historical operating data, allowing the fault detection module to extract features from historical operating data spanning a sufficiently long period, resulting in more accurate and reliable fault detection feature data. Then, the batch computing module performs batch calculations on this large amount of fault detection feature data using a batch computing engine to quickly obtain the corresponding intermediate parameters, ensuring that the subsequent stream computing platform can promptly acquire these intermediate parameters during real-time fusion calculations. The stream computing platform can acquire intermediate parameters and real-time battery operating data in real-time for fusion calculations to obtain the corresponding real-time fault detection result for the battery. Since the fault detection result is obtained from the battery's historical and real-time operating data through corresponding feature extraction and fusion calculations, the accuracy is higher, reducing the false alarm rate of battery risk identification and improving the reliability of battery fault detection. Furthermore, in this embodiment, the stream computing platform can acquire and process large-scale flowing data (i.e., real-time battery operating data) in real-time based on the stream computing engine, and the batch computing module provides rapid calculation capabilities for intermediate parameters before this processing, further ensuring the real-time execution of the calculation process. Therefore, the embodiments of this application, while taking into account the reliability of the detection results, can also ensure the real-time nature of fault detection, which is conducive to timely fault response to battery fault risks and provides a guarantee for the safe operation of power batteries.

[0007] In some embodiments, the system further includes a second data warehouse for storing intermediate parameters and real-time battery operating data in real time for the streaming computing platform to access.

[0008] According to the embodiments of this application, since the real-time operating data of the battery is continuously updated, in order to enable the stream computing platform to acquire and process the real-time operating data generated by the battery in real time, the system of this embodiment can store the intermediate parameters calculated by the batch computing module and the real-time operating data of the battery obtained from the terminal in real time through a data warehouse, thereby providing a guarantee for the stream computing platform to read data during the streaming fusion computing process.

[0009] In some embodiments, the streaming computing platform includes: a data interface for collecting real-time operating data of corresponding batteries from various terminals via an Internet of Things (IoT) protocol; a first transmission unit for transmitting real-time operating data to a second data warehouse to update existing real-time operating data of batteries stored in the second data warehouse; and a second transmission unit for transmitting real-time operating data to a first data warehouse for storage to update historical operating data of batteries.

[0010] According to embodiments of this application, the data platform, based on the Internet of Things (IoT) protocol, collects real-time battery operating data from various terminals via data interfaces, completing big data collection for batch battery risk identification, thereby facilitating large-scale battery risk identification and management. The acquired real-time operating data is transmitted to a second data warehouse via a first transmission unit to continuously update existing real-time battery operating data, thus providing a continuous, real-time data foundation for the stream computing platform's fusion computing, improving the real-time performance and reliability of large-scale battery risk identification and management. Furthermore, the real-time operating data is also transmitted to the first data warehouse via the second transmission unit for storage, updating historical battery operating data and providing the time-series data foundation required for feature extraction by the fault detection module. This improves the robustness of the fault detection module's computation based on massive amounts of data, while also enhancing the accuracy of battery fault detection.

[0011] In some embodiments, the fault detection module includes: a data acquisition unit, configured to acquire historical battery operation data corresponding to a sampling period from a first data warehouse at a preset sampling time, and input the data into a feature extraction unit in chronological order; and a feature extraction unit, configured to perform multi-dimensional feature extraction on the historical battery operation data using a preset first fault detection model, and output corresponding fault detection feature data.

[0012] According to an embodiment of this application, the fault detection module includes a data acquisition unit and a feature extraction unit. The data acquisition unit acquires historical battery operating data from a first data warehouse at a preset sampling time and inputs it into the feature extraction unit in chronological order. This allows the feature extraction unit to perform multi-dimensional feature extraction on the historical battery operating data in chronological order using a preset first fault detection model. This fully integrates the multidimensional and chronological influences of fault feature sources, resulting in more accurate fault detection feature data and thus facilitating more reliable fault detection results.

[0013] In some embodiments, the batch calculation module includes: a first calculation unit, used to perform batch calculations on fault detection feature data through a batch calculation engine to obtain corresponding intermediate parameters; a generation unit, used to generate an intermediate table corresponding to the time series based on the intermediate parameters; and a third transmission unit, used to transmit the intermediate table to a second data warehouse to update the existing intermediate table stored in the second data warehouse.

[0014] According to an embodiment of this application, the batch computing module includes a first computing unit, a generation unit, and a third transmission unit. The first computing unit, through a batch computing engine, performs batch calculations on a large amount of fault detection feature data obtained from massive historical battery operation data, quickly obtaining corresponding intermediate parameters, which is beneficial for improving data computing efficiency in big data processing scenarios. The generation unit generates intermediate tables corresponding to the time series based on the intermediate parameters, and then the third transmission unit transmits the intermediate tables to a second data warehouse for storage and use by the stream computing platform. The intermediate parameters are stored in the second data warehouse via intermediate tables before fusion computing, allowing for faster and more convenient retrieval, reducing the waiting time for real-time computing, reducing the data processing difficulty of stream computing, and improving the computing efficiency in the real-time fault detection process.

[0015] In some embodiments, the stream computing platform further includes: a real-time acquisition module, used to acquire intermediate parameters and real-time battery operating data from a second data warehouse in real time through a stream computing engine; a determination module, used to determine the corresponding real-time battery status data based on the real-time battery operating data; a splicing module, used to splice the real-time battery operating data with the intermediate parameters in an intermediate table; and a calculation module, used to calculate the spliced ​​data using a preset second fault detection model, and output the corresponding battery fault detection data and battery identity information.

[0016] According to an embodiment of this application, the real-time acquisition module of the stream computing platform, through the stream computing engine, can continuously acquire intermediate parameters and real-time battery operating data from the second data warehouse in a stream computing manner. The determination module then determines the real-time battery status data corresponding to the real-time battery operating data. Next, the splicing module splices the real-time battery operating data with the intermediate parameters in the intermediate table to form a new data table corresponding to the time sequence. The calculation module then reads the data from this table, performs calculations using a preset second fault detection model, and outputs the corresponding fault detection data and the battery's identity information. This facilitates obtaining fault detection data for each battery in big data battery risk identification and management scenarios.

[0017] In some embodiments, the system further includes an early warning module and a service subsystem. The early warning module is used to generate early warning information and send it to the service subsystem based on the fault detection data and the battery's identity information. The service subsystem is used to query the target terminal corresponding to the battery's identity information based on the early warning information, so as to generate service information for the corresponding target terminal and fault detection data.

[0018] According to the embodiments of this application, when the fault detection data indicates that the battery has a fault or abnormal condition, an early warning message can be generated and sent to the service subsystem. The service subsystem can then query the target terminal corresponding to the battery identity information and send the relevant service information to the target terminal for the user to know, thereby realizing big data battery fault risk early warning.

[0019] Secondly, embodiments of this application provide a battery fault detection method, including:

[0020] The system retrieves historical battery operation data from the first data warehouse; extracts features from the historical battery operation data using a pre-set first fault detection model to obtain corresponding fault detection feature data; performs batch calculations on the fault detection feature data using a batch computing engine to obtain corresponding intermediate parameters; performs batch calculations on the fault detection feature data using a batch computing engine to obtain corresponding intermediate parameters; acquires intermediate parameters and real-time battery operation data in real time using a stream computing platform; and performs fusion calculations on the intermediate parameters and real-time battery operation data using a second fault detection model to obtain the corresponding battery fault detection result.

[0021] According to the embodiments of this application, the first data warehouse can store massive amounts of historical operating data. Based on this massive amount of historical operating data with a sufficiently long time span, feature extraction is performed through the first fault detection model to obtain more accurate fault detection feature data. Then, the batch computing engine performs batch calculations on a large amount of this fault detection feature data to quickly obtain the corresponding intermediate parameters, ensuring that the intermediate parameters can be obtained in a timely manner when the stream computing engine performs real-time fusion calculations. The stream computing engine can obtain intermediate parameters and real-time battery operating data in real time, and use the second fault detection model to perform fusion calculations on these data to obtain the battery fault detection result. Since the fault detection result is obtained from the battery's historical operating data and real-time operating data through corresponding feature extraction and fusion calculations, the accuracy is higher, reducing the false alarm rate of battery risk identification and improving the reliability of battery fault detection. Furthermore, in this embodiment of the application, the stream computing engine can perform real-time acquisition and calculation processing of large-scale flowing data (i.e., real-time battery operating data), and the batch computing engine provides the ability to quickly calculate intermediate parameters before the calculation processing, further ensuring the real-time execution of the calculation processing. Therefore, the embodiments of this application, while taking into account the reliability of the detection results, can also ensure the real-time nature of fault detection, which is conducive to timely fault response to battery fault risks and provides a guarantee for the safe operation of power batteries.

[0022] In some embodiments, before extracting features from the battery historical operation data using a preset first fault detection model, the method further includes: obtaining battery historical operation sample data within a target duration from a first data warehouse; performing multi-dimensional feature extraction on the battery historical operation sample data to construct a first fault detection model corresponding to fault features.

[0023] According to the embodiments of this application, a large amount of historical operating data within a target duration stored in the first data warehouse can be used as samples (i.e., historical battery operating sample data) to perform multi-dimensional feature extraction, thereby obtaining fault features in various dimensions and constructing a corresponding first fault detection model. Because a large amount of historical operating data is used as training samples, the constructed data model can obtain more accurate extraction results when applied to feature extraction.

[0024] In some embodiments, feature extraction is performed on historical battery operation data using a preset first fault detection model to obtain corresponding fault detection feature data, including: inputting historical battery operation data of the corresponding sampling period into the first fault detection model in chronological order at a preset sampling time; and performing multi-dimensional feature extraction on the historical battery operation data using the first fault detection model to obtain corresponding fault detection feature data.

[0025] According to an embodiment of this application, at a preset sampling time, the historical battery operation data corresponding to the sampling period is input into the first fault detection model in chronological order, enabling the first fault detection model to extract multi-dimensional features from the historical battery operation data in chronological order. This fully integrates the multidimensional and chronological influences of fault feature sources, resulting in more accurate fault detection feature data, and thus facilitating the acquisition of more reliable fault detection results.

[0026] In some embodiments, after performing batch calculations on fault feature data using a batch computing engine to obtain corresponding intermediate parameters, the method further includes: generating an intermediate table corresponding to the time series based on the intermediate parameters; and inputting the intermediate table into a preset second data warehouse to update the existing intermediate table stored in the second data warehouse, wherein the second data warehouse is a real-time data warehouse.

[0027] According to the embodiments of this application, intermediate parameters are stored in the second data warehouse in the form of intermediate tables before being used for fusion calculation. This allows them to be called more quickly and conveniently, reducing the waiting time for real-time calculation, reducing the data processing difficulty of streaming calculation, and improving the calculation efficiency in the real-time fault detection process.

[0028] In some embodiments, intermediate parameters and real-time battery operating data are acquired in real time through a stream computing engine, and fused together using a second fault detection model to obtain the corresponding battery fault detection result. This includes: acquiring intermediate parameters and real-time battery operating data from a second data warehouse in real time through a stream computing engine; determining the corresponding real-time battery status data based on the real-time battery operating data; concatenating the real-time battery status data with intermediate parameters in an intermediate table; calculating the concatenated data using a preset second fault detection model, and outputting the corresponding battery fault detection data and battery identity information.

[0029] According to the embodiments of this application, intermediate parameters and real-time battery operating data are continuously and in real-time obtained from the second data warehouse through a stream computing engine. Based on the real-time battery operating data, the corresponding real-time battery status data is continuously updated. Then, the real-time battery status data is concatenated with the intermediate parameters in the intermediate table to form a new data table corresponding to the time series. With the support of the stream computing engine, the second fault detection model continuously reads the data in the data table for calculation and outputs the fault detection data and battery identity information of the corresponding battery. This facilitates the real-time acquisition of fault detection data of each battery in the big data battery fault risk identification and management scenario, thereby improving the response capability to battery fault risks.

[0030] In some embodiments, after obtaining the fault detection result of the corresponding battery, the method further includes: generating warning information based on the fault detection data of the corresponding battery and the battery's identity information; sending the warning information to the service subsystem, so that the service subsystem can query the target terminal corresponding to the battery's identity information based on the warning information, and generate service information for the corresponding target terminal and fault detection data.

[0031] According to the embodiments of this application, when the fault detection data indicates that the battery has a fault or abnormal condition, an early warning message can be generated and sent to the service subsystem. The service subsystem can then query the target terminal corresponding to the battery identity information and send the relevant service information to the target terminal for the user to know, thereby realizing big data battery fault risk early warning.

[0032] Thirdly, embodiments of this application provide an electronic device, including a processor, a memory, and a program or instructions stored in the memory and executable on the processor. When the program or instructions are executed by the processor, they implement the steps of the battery fault detection method as described in any embodiment of the first aspect.

[0033] Fourthly, embodiments of this application provide a readable storage medium on which a program or instructions are stored, which, when executed by a processor, implement the steps of the battery fault detection method as described in any embodiment of the first aspect.

[0034] Fifthly, embodiments of this application provide a chip, the chip including a processor and a communication interface, the chip including a processor and a communication interface, the communication interface and the processor being coupled, the processor being used to run programs or instructions to implement the steps of the battery fault detection method as described in any embodiment of the first aspect.

[0035] Sixthly, embodiments of this application provide a computer program product in which instructions, when executed by a processor of an electronic device, cause the electronic device to perform the steps of the battery fault detection method as described in any embodiment of the first aspect.

[0036] The above description is only an overview of the technical solution of this application. In order to better understand the technical means of this application and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of this application more obvious and understandable, the following are specific embodiments of this application. Attached Figure Description

[0037] To more clearly illustrate the technical solutions of the embodiments of this application, the accompanying drawings used in the embodiments of this application will be briefly described below:

[0038] Figure 1 This is a schematic diagram of the structure of a battery fault detection system disclosed in an embodiment of this application;

[0039] Figure 2 This is a schematic diagram of a battery fault detection system performing fault detection in a specific embodiment of this application;

[0040] Figure 3 This is a schematic flowchart of a battery fault detection method disclosed in an embodiment of this application;

[0041] Figure 4 This is a schematic diagram of the hardware structure of an electronic device disclosed in an embodiment of this application. Detailed Implementation

[0042] The embodiments of this application will be described in further detail below with reference to the accompanying drawings and examples. The detailed description of the following embodiments and the accompanying drawings are used to illustrate the principles of this application by way of example, but should not be used to limit the scope of this application, that is, this application is not limited to the described embodiments.

[0043] In the description of this application, it should be noted that, unless otherwise stated, "a plurality of" means two or more; the terms "upper," "lower," "left," "right," "inner," and "outer," etc., indicating orientation or positional relationships, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this application. Furthermore, the terms "first," "second," and "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. "Vertical" is not vertical in the strict sense, but within the allowable tolerance range. "Parallel" is not parallel in the strict sense, but within the allowable tolerance range.

[0044] In this application, the reference to "embodiment" means that a specific feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a mutually exclusive, independent, or alternative embodiment. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described in this application can be combined with other embodiments.

[0045] In the field of power battery fault detection, due to the strong correlation between power battery faults and time series, current power battery fault identification typically requires calculating data over a very long period to ensure the reliability of early warning results. Therefore, current risk battery early warning systems based on big data fault detection platforms often require calculating and processing data from hundreds of thousands or even millions of vehicles over one to two years. To meet the requirements of big data storage and processing capabilities, most systems currently use big data platforms for power battery fault detection. However, even with powerful computing resources, processing such large volumes of data can be slow, taking days or even weeks to complete. A common solution is to shorten the data acquisition cycle to reduce the amount of data used for fault detection, thereby improving real-time performance. However, this comes at the cost of lower accuracy, because power battery faults are strongly correlated with time series, and short-term data often fails to reflect the true state of the battery.

[0046] To address some or all of the problems existing in the aforementioned related technologies, embodiments of this application provide a battery fault detection system, method, and device. Based on massive amounts of historical battery operating data stored in an offline warehouse, feature extraction and batch calculation are performed, enabling rapid acquisition of intermediate parameters for fault detection features corresponding to the historical operating data. A stream computing engine then acquires these intermediate parameters in real time and fuses them with the battery's real-time operating data for calculation, resulting in more accurate real-time fault detection results. This approach combines accuracy and real-time performance in large-scale battery risk identification and management scenarios. The battery fault detection system provided in this application embodiment will be described below.

[0047] Figure 1 The diagram shown is a structural schematic of a fault detection system provided in an embodiment of this application. Figure 1 As shown, the system 100 includes:

[0048] First data warehouse 101 is used to store historical battery operating data;

[0049] The fault detection module 102 is used to obtain the corresponding historical battery operation data from the first data warehouse for feature extraction to obtain the corresponding fault detection feature data.

[0050] Batch calculation module 103: Used to perform batch calculations on fault detection feature data through the batch calculation engine to obtain the corresponding intermediate parameters;

[0051] The stream computing platform 104 is used to obtain intermediate parameters and real-time battery operating data in real time through the stream computing engine, perform fusion calculations, and obtain the corresponding battery fault detection results.

[0052] According to the embodiments of this application, the first data warehouse 101 can store massive amounts of historical operating data, allowing the fault detection module 102 to extract features from historical operating data spanning a sufficient time period, thereby obtaining more accurate and reliable fault detection feature data. Then, the batch calculation module 103 performs batch calculations on a large amount of this fault detection feature data using a batch calculation engine to quickly obtain the corresponding intermediate parameters, ensuring that the subsequent stream computing platform 104 can promptly acquire these intermediate parameters during real-time fusion calculations. The stream computing platform 104 can acquire intermediate parameters and real-time battery operating data in real-time through the stream computing engine for fusion calculations, obtaining the corresponding real-time fault detection result for the battery. Since the fault detection result is obtained from the battery's historical operating data and real-time operating data through corresponding feature extraction and fusion calculations, the accuracy is higher, reducing the false alarm rate of battery risk identification and improving the reliability of battery fault detection. Furthermore, in this embodiment, the stream computing platform 104 can perform real-time acquisition and calculation processing of large-scale flowing data (i.e., real-time battery operating data) based on the stream computing engine, and the batch calculation module 103 provides rapid calculation capabilities for intermediate parameters before this calculation processing, further ensuring the real-time execution of the calculation processing. Therefore, the embodiments of this application, while taking into account the reliability of the detection results, can also ensure the real-time nature of fault detection, which is conducive to timely fault response to battery fault risks and provides a guarantee for the safe operation of power batteries.

[0053] In this embodiment, the first data warehouse 101 can be an offline data warehouse, such as a Hive warehouse. As battery usage time increases, the risk of battery failure also increases. Based on the strong temporal correlation of battery failure risk, this embodiment stores massive amounts of historical battery operation data in the offline data warehouse, enabling the system to obtain accurate fault detection results by calculating historical operation data over a sufficiently long time span, thereby improving the reliability of fault warning.

[0054] In some examples, for scenarios where terminal battery failure risk monitoring is conducted through a big data platform, the first data warehouse can store vehicle power battery operation data obtained by the big data platform from the terminals (such as vehicle terminals) of hundreds of thousands or even millions of vehicles within 1 to 2 years, in order to provide sufficient data samples for subsequent extraction and detection of fault characteristics.

[0055] For example, historical battery operating data may include, but is not limited to, the battery's voltage, current, temperature, state of charge (SOC), and state of health values ​​at historical times.

[0056] In some examples, the first data warehouse 101 can clean and normalize the acquired massive amounts of battery data before storing it, so that it can be accessed by the fault detection module 102.

[0057] In some embodiments, reference Figure 2 As shown, the system 100 also includes a second data warehouse 105, which is used to store intermediate parameters and real-time battery operation data in real time for the streaming computing platform 104 to obtain.

[0058] In this embodiment, since the real-time operating data of the battery is continuously updated, in order to enable the streaming computing platform to acquire and process the real-time operating data generated by the battery in real time, the system of this embodiment can store the intermediate parameters calculated by the batch computing module and the real-time operating data of the battery obtained from the terminal in real time through a data warehouse, thereby providing a guarantee for the streaming computing platform to read data during the streaming fusion computing process.

[0059] For example, the second data warehouse 105 can be a real-time data warehouse, such as an HBase warehouse. The second data warehouse 105 can acquire and store each frame of real-time operating data from the battery, updating the corresponding real-time operating data for use by the stream computing platform 104. In this way, the stream computing platform 104 can continuously fuse and calculate feature data related to the battery's historical operating data with each frame of real-time operating data through its stream computing engine, obtaining more accurate and reliable real-time fault detection results for the battery, thereby facilitating real-time fault risk prediction.

[0060] For example, real-time battery operating data may include, but is not limited to, real-time battery voltage, current, temperature, state of charge (SOC), and state of health (SOH) values. Correspondingly, the real-time battery status data stored in the second data warehouse 105 may include voltage, current, temperature, SOC, and SOH values.

[0061] To ensure the real-time performance of data processing, optionally, in this embodiment, the stream computing platform 104 may include:

[0062] The data interface is used to collect real-time operating data of the corresponding battery from various terminals via the Internet of Things protocol;

[0063] The first transmission unit is used to transmit real-time operating data to the second data warehouse to update the existing real-time operating data of the battery stored in the second data warehouse.

[0064] The second transmission unit is used to transmit real-time operating data to the first data warehouse for storage, so as to update the battery's historical operating data.

[0065] In some examples, reference Figure 2 As shown, the stream computing platform 104 can be a processing platform including the Flink stream computing engine 106 (hereinafter referred to as the "Flink platform"). The Flink platform provides an open-source stream processing framework and a distributed stream computing engine 106, which can execute streaming data programs in a data-parallel and pipelined manner, thereby continuously and in real-time acquiring the terminal's real-time battery operation data and storing this data in the first data warehouse 101 and the second data warehouse 105 respectively. In the second data warehouse 105, new real-time operation data overwrites existing data, achieving continuous data updates. In the first data warehouse 101, both new real-time operation data and existing historical battery operation data are stored in the first data warehouse 101, allowing data to accumulate continuously and achieving overall data updates.

[0066] Specifically, during a real-time fault detection process, the fault detection module 102 retrieves historical operating data from the first data warehouse 101, extracts features, obtains corresponding fault detection feature data, performs batch calculations, generates intermediate parameters for this historical operating data, and stores them in the second data warehouse 105. These parameters are then used in fusion calculations together with the real-time operating data retrieved from the second data warehouse 105. Simultaneously, the real-time operating data retrieved in this instance is stored in the first data warehouse 101 in preparation for the next real-time fault detection.

[0067] Furthermore, based on the Flink platform's stream computing engine, during the fusion computing process, it can continuously and systematically acquire the aforementioned real-time operational data and intermediate parameters updated over time from the second data warehouse using a stream computing approach, and then perform data fusion computing. In this way, the results of the fusion computing can retain the temporal attributes of both historical and real-time battery operational data, making the fault detection results strongly correlated with the time sequence of the data. This aligns with the objective laws governing changes in battery operating status over time, thereby improving the accuracy of fault detection results.

[0068] In this example, the Flink platform demonstrates its advantages in supporting high throughput, low latency, and high performance, ensuring real-time data processing in large-scale battery fault detection scenarios. The Flink platform also efficiently processes data according to continuous events, and its superior fault tolerance based on lightweight distributed snapshots guarantees the reliability and durability of the Flink platform.

[0069] In other examples, the stream computing platform may also be a real-time platform employing other types of stream computing engines to achieve the functions of the stream computing platform described in this application.

[0070] The Flink platform establishes communication connections with terminals based on IoT protocols using data interfaces. It is understood that the IoT protocol can be one or more of the following: Message Queue Telemetry Transport (MQTT), Constrained Application Protocol (CoAP), Lightweight Machine-to-Machine (LwM2M), HyperText Transfer Protocol (HTTP), Narrow Band Internet of Things (NB-IoT), etc. This application does not specifically limit the specific protocols used.

[0071] The terminal monitors the real-time operating data of the battery and uploads the monitored data to the Flink platform in real time. After receiving the data from the terminal, the Flink platform transmits it to the first data warehouse 101 via the second transmission unit to update the historical battery operating data stored in the first data warehouse 101. Simultaneously, the Flink platform transmits the data received from the terminal to the second data warehouse 105 via the first transmission unit to update the existing real-time battery status data stored in the second data warehouse 105.

[0072] According to an embodiment of this application, the data platform 105, based on the Internet of Things (IoT) protocol, collects real-time battery operating data from various terminals through a data interface, completing big data collection for batch battery risk identification, thereby facilitating large-scale battery risk identification and management. The acquired real-time operating data is transmitted to the second data warehouse 105 via a first transmission unit to continuously update existing real-time battery operating data, thus providing a continuous real-time data foundation for the stream computing platform 104 to perform fusion computing, improving the real-time performance and reliability of large-scale battery risk identification and management. Furthermore, the real-time operating data is also transmitted to the first data warehouse 101 via the second transmission unit for storage, updating historical battery operating data and providing the time-sensitive data foundation required for feature extraction by the fault detection module 102. This improves the robustness of the fault detection module 102's calculations based on massive amounts of data and also helps to improve the accuracy of battery fault detection.

[0073] In some embodiments, the fault detection module 102 may include:

[0074] The data acquisition unit is used to acquire the battery historical operation data of the corresponding sampling period from the first data warehouse 101 at a preset sampling time, and input it into the feature extraction unit in time sequence;

[0075] The feature extraction unit is used to extract multi-dimensional features from the battery's historical operating data using a preset first fault detection model, and output the corresponding fault detection feature data.

[0076] In this embodiment, the sampling period can be a time measure such as year, month, day, hour, minute, etc., and the preset sampling time can also be set at intervals of day, hour, minute, second, etc. The fault detection module 102 obtains the battery historical operation data within the corresponding sampling period from the first data warehouse 101 through the data acquisition unit at the preset sampling time, and inputs it into the feature extraction unit according to the time of the obtained battery historical operation data in time sequence. Multi-dimensional feature extraction is performed through the preset first fault detection model to obtain fault detection characteristic data.

[0077] For example, before extracting features from the battery's historical operating data, the feature extraction unit can first obtain a sufficient number of battery historical operating data samples from the first data warehouse 101. From these samples, voltage characteristics, temperature characteristics, current characteristics, and corresponding fault codes (such as thermal runaway fault codes) are extracted to construct a multi-dimensional first fault detection model. This first fault detection model can be used to acquire battery historical operating data for prediction and identification, outputting corresponding fault codes and other fault detection feature data. In some specific examples, the constructed first fault detection model can be a matrix model, a neural network model, or a decision tree model, etc.; this embodiment does not impose a unique limitation.

[0078] Based on the first fault detection model, the feature extraction unit acquires historical battery operating data within the sampling period to perform multi-dimensional feature extraction and identification, outputting fault detection characteristic data. This fully integrates the multi-dimensional and temporal influences of fault feature sources, resulting in more accurate fault detection feature data, which in turn facilitates obtaining more reliable fault detection results.

[0079] In some embodiments, the batch calculation module 103 may specifically include:

[0080] The first calculation unit is used to perform batch calculations on fault detection feature data through a batch calculation engine to obtain the corresponding intermediate parameters;

[0081] The generation unit is used to generate intermediate tables for the corresponding timing based on intermediate parameters;

[0082] The third transmission unit is used to transmit the intermediate table to the second data warehouse to update the existing intermediate table stored in the second data warehouse.

[0083] For example, the batch computing engine can be the Spark batch computing engine. The Spark batch computing engine has the advantages of speed and versatility, enabling batch computation of massive amounts of fault detection feature data to obtain corresponding intermediate parameters and generate intermediate tables for the corresponding time series. In a specific example, intermediate parameters may include battery ID, detection time, report time, current value, voltage value, SOC value, SOH value, fault code, battery status, etc. This data is used to generate intermediate tables and stored in the second data warehouse 105 for use by the stream computing platform 104 during real-time fault detection, reducing the data computation complexity during stream computing.

[0084] In this embodiment, in the scenario of battery fault risk early warning using a big data platform, system 100 typically needs to process 1-2 years of historical operating data from hundreds of thousands or even millions of vehicles, with the data volume reaching the petabyte level. If a big data platform from related technologies is used to process such a massive amount of data, the processing speed would be extremely slow, generally requiring several days or even weeks to complete. Furthermore, if triggering early warning measures is required after processing, a reaction time of more than one day is needed. Thus, if the battery fails within two days of triggering the early warning in a practical application, related technologies cannot meet the requirement for early warning of fault risks. However, in this embodiment, system 100 periodically obtains sufficient historical battery operating data from the offline first data warehouse 101 through the fault detection module 102 for fault detection. The fault detection feature data is then batch-calculated to generate an intermediate table, which is saved to the second data warehouse 105. This table can then be directly accessed during real-time fault detection by the stream computing platform 104, eliminating the need for additional calculation of historical operating data features and improving the rapid response capability in real-time fault risk identification and early warning scenarios.

[0085] According to the embodiments of this application, the system 100, based on the Spark batch computing engine of the batch computing module 106, supports Spark batch computing during the feature extraction and identification process of massive data in the fault detection module 102. This enables the rapid acquisition of corresponding intermediate parameters, which is beneficial for improving data computing efficiency in big data processing scenarios. The intermediate table generated by the intermediate parameters is transmitted to the second data warehouse 105 for storage, allowing the stream computing platform 104 to access it more quickly and conveniently. This reduces the waiting time for real-time computing, lowers the data processing difficulty of stream computing, and further improves the computing efficiency in the real-time fault detection process.

[0086] In some embodiments, the stream computing platform 104 may include:

[0087] The real-time acquisition module is used to acquire intermediate parameters and real-time battery operation data from the second data warehouse in real time through the stream computing engine;

[0088] The determination module is used to determine the corresponding real-time battery status data based on the real-time battery operating data.

[0089] The splicing module is used to splice real-time battery status data with intermediate parameters in the intermediate table;

[0090] The calculation module is used to calculate the spliced ​​data using a preset second fault detection model, and output the corresponding battery fault detection data and battery identification information.

[0091] In this embodiment of the application, reference is made to Figure 2 As shown, the real-time acquisition module of the stream computing platform, through the stream computing engine 106, can continuously and systematically acquire intermediate parameters and real-time battery operating data from the second data warehouse in a stream computing manner. The determination module then determines the corresponding real-time battery status data based on the real-time battery operating data. The real-time battery status data can be stored in the form of a table. Therefore, during the stitching process, the stitching module can stitch together the table of real-time battery status data and the aforementioned intermediate table in the second data warehouse 105 to form a new table. Based on this new table, the calculation module reads the corresponding data and inputs it into a preset second fault detection model. This model can comprehensively analyze the characteristics of historical operating data and the faults in real-time operating data to perform identification calculations, obtaining the fault detection data and identity information (such as Identity document, id) for each battery. Furthermore, based on the support of the stream computing engine, the temporal attributes of historical and real-time battery operating data can be preserved, making the fault detection results strongly correlated with the time sequence of the data. This aligns with the objective law of battery operating status changes over time, improving the accuracy of the fault detection results.

[0092] In some specific examples, the second fault detection model can be a pre-built matrix model, neural network model, or decision tree model. The calculation module inputs the concatenated real-time battery status data and intermediate parameters into the corresponding second fault detection model for processing, and outputs fault detection data and identification information for each battery. For example, the concatenated historical and real-time battery IDs, current values, voltage values, fault codes, SOC values, etc., are input into the model, and the model outputs data such as whether each battery ID has a fault risk, the fault code of the fault risk, etc., thereby obtaining the corresponding fault detection data for each battery in a big data battery risk identification and management scenario.

[0093] To achieve early warning, optionally, in some embodiments, system 100 may further include an early warning module 107 and a service subsystem 108, wherein:

[0094] The early warning module 107 is used to generate early warning information and send it to the service subsystem 108 based on the fault detection data and the battery's identity information. The service subsystem 108 is used to query the target terminal corresponding to the battery's identity information based on the early warning information, so as to generate service information for the corresponding target terminal and fault detection data.

[0095] For example, the service subsystem 108 can be the after-sales service system of the operator. For instance, if the data output by the stream computing platform 104 indicates that the power battery with battery ID "123" is at risk of thermal runaway, then the early warning model 107 can generate early warning information based on this data. The early warning information may include the battery ID and the identifier of the fault risk, and is sent to the service subsystem 108 in real time.

[0096] After receiving the warning information, service subsystem 108 can locate the target terminal corresponding to the battery ID and send relevant service information to the target terminal for user awareness. This allows the terminal to perform real-time risk interception of the corresponding battery and instruct risk intervention, such as battery replacement. This achieves a closed-loop process in the detection and warning of battery failure risks using big data, resulting in a low false alarm rate and high reliability.

[0097] The fault detection system provided in this application is particularly suitable for fault risk identification and early warning of power batteries in new energy vehicles based on a big data platform. It does not require additional hardware costs for the Battery Management System (BMS) and is easy to deploy. Furthermore, based on an offline batch processing engine, it can train historical operating data, optimize fault detection feature data parameters, improve the accuracy of fault feature identification, and sample each frame of real-time operating data using stream computing for fusion calculation to obtain more reliable fault detection results, providing early warning protection for the safety of users and vehicle batteries.

[0098] This application also provides a fault detection method. Figure 3 The diagram shown is a flowchart illustrating a fault detection method provided in an embodiment of this application. Figure 3 As shown, the method may include S101 to S104:

[0099] S101 retrieves the corresponding historical battery operation data from the first data warehouse;

[0100] S102 extracts features from the battery's historical operating data using a preset first fault detection model to obtain the corresponding fault detection feature data;

[0101] The S103 uses a batch processing engine to perform batch calculations on fault detection feature data to obtain the corresponding intermediate parameters.

[0102] S104 uses a stream computing engine to acquire intermediate parameters and real-time battery operating data in real time, and uses a second fault detection model to perform fusion calculations to obtain the corresponding battery fault detection results.

[0103] According to the embodiments of this application, the first data warehouse can store massive amounts of historical operating data. Based on this massive amount of historical operating data with a sufficiently long time span, feature extraction is performed through the first fault detection model to obtain more accurate fault detection feature data. Then, the batch computing engine performs batch calculations on a large amount of this fault detection feature data to quickly obtain the corresponding intermediate parameters, ensuring that the intermediate parameters can be obtained in a timely manner when the stream computing engine performs real-time fusion calculations. The stream computing engine can obtain intermediate parameters and real-time battery operating data in real time, and use the second fault detection model to perform fusion calculations on these data to obtain the battery fault detection result. Since the fault detection result is obtained from the battery's historical operating data and real-time operating data through corresponding feature extraction and fusion calculations, the accuracy is higher, reducing the false alarm rate of battery risk identification and improving the reliability of battery fault detection. Furthermore, in this embodiment of the application, the stream computing engine can perform real-time acquisition and calculation processing of large-scale flowing data (i.e., real-time battery operating data), and the batch computing engine provides the ability to quickly calculate intermediate parameters before the calculation processing, further ensuring the real-time execution of the calculation processing. Therefore, the embodiments of this application, while taking into account the reliability of the detection results, can also ensure the real-time nature of fault detection, which is conducive to timely fault response to battery fault risks and provides a guarantee for the safe operation of power batteries.

[0104] In some embodiments, the first data warehouse can be an offline data warehouse, such as a Hive warehouse, capable of storing massive amounts of historical battery operating data. This provides historical operating data with a sufficiently long time span for relevant calculations in fault detection, resulting in accurate fault detection results and improving the reliability of fault warnings. The historical battery operating data may include, but is not limited to, battery voltage, current, temperature, state of charge (SOC), and state of health values ​​at historical times.

[0105] In some examples, before S101 retrieves the corresponding historical battery operation data from the first data warehouse, the method may also include: collecting massive amounts of historical battery operation data, cleaning and normalizing it, and storing it in the first data warehouse so that it can be called by the first fault detection model.

[0106] In some embodiments, to ensure the effective execution of feature extraction, before S102 performs feature extraction on the battery historical operating data using a preset first fault detection model, the method may further include:

[0107] Obtain historical battery operation sample data within the target duration from the first data warehouse;

[0108] Multi-dimensional feature extraction is performed on historical battery operation sample data to construct a first fault detection model with corresponding fault features.

[0109] For example, historical battery operation sample data refers to using a sufficient amount of historical battery operation data as samples to extract voltage characteristics, temperature characteristics, current characteristics, and corresponding fault codes (such as thermal runaway fault codes), etc., to construct a multi-dimensional first fault detection model. This first fault detection model can be used to acquire historical battery operation data for prediction and identification, outputting corresponding fault codes and other fault detection feature data corresponding to different fault risks. In some specific examples, the constructed first fault detection model can be a matrix model, a neural network model, or a decision tree model, etc., and this embodiment is not limited to these. Furthermore, it should be understood that constructing matrix models, neural network models, or decision tree models based on sample data are all mature technologies in this field and will not be elaborated upon here.

[0110] According to the embodiments of this application, a large amount of historical operating data within a target duration stored in the first data warehouse can be used as samples (i.e., historical battery operating sample data) to perform multi-dimensional feature extraction, thereby obtaining fault features in various dimensions and constructing a corresponding first fault detection model. Because a large amount of historical operating data is used as training samples, the constructed data model can ensure the effective execution of feature extraction when applied to feature extraction, resulting in more accurate extraction results.

[0111] For example, step 102 extracts features from the battery's historical operating data using a preset first fault detection model to obtain corresponding fault detection feature data, which may specifically include:

[0112] At the preset sampling time, the historical battery operation data of the corresponding sampling period is input into the first fault detection model in chronological order;

[0113] The first fault detection model is used to extract multi-dimensional features from the battery's historical operating data to obtain the corresponding fault detection feature data.

[0114] In this embodiment, the sampling period can be a time measure such as year, month, day, hour, minute, etc., and the preset sampling time can also be set at intervals of days, hours, minutes, seconds, etc. In this embodiment, at the preset sampling time, the battery historical operation data corresponding to the sampling period is input into the first fault detection model in chronological order, so that the first fault detection model can extract multi-dimensional features from the battery historical operation data in chronological order. This can fully integrate the multidimensional and chronological influence of the fault feature sources to obtain more accurate fault detection feature data, thereby facilitating the obtaining of more reliable fault detection results.

[0115] In large-scale battery fault risk detection scenarios, it is necessary to acquire and process massive amounts of historical battery operation data. To improve data processing efficiency and thus enhance the response capability to fault risks, in some embodiments of this application, step S103 is performed to batch calculate fault detection feature data using a batch computing engine. This improves the system's ability to process massive amounts of data. Thus, in large-scale battery fault risk identification scenarios, fault detection feature data extracted from massive historical battery operation data can be efficiently processed, quickly obtaining intermediate parameters about the fault detection features for the stream computing engine to call. To ensure the effective execution of stream computing by the stream computing engine, for example, after obtaining the corresponding intermediate parameters by batch calculating the fault feature data using the batch computing engine in step 103, the method may further include:

[0116] Based on the intermediate parameters, generate the corresponding timing table;

[0117] The intermediate table is input into the preset second data warehouse to update the existing intermediate table stored in the second data warehouse, which is a real-time data warehouse.

[0118] For example, the batch computing engine can be the Spark batch computing engine. The Spark batch computing engine has the advantages of speed and versatility, enabling batch computation of massive amounts of fault detection feature data to obtain corresponding intermediate parameters and generate intermediate tables for the corresponding time series. In a specific example, intermediate parameters may include battery ID, detection time, report time, current value, voltage value, SOC value, SOH value, fault code, battery status, etc. This data is used to generate intermediate tables and stored in the second data warehouse 105, so that the stream computing engine can call them during real-time fault detection calculations, reducing the data computation complexity during stream computing.

[0119] In this embodiment, the method uses a first fault detection model to periodically obtain sufficient historical operating data of the battery from an offline first data warehouse for fault detection, obtains fault detection feature data, performs batch calculations to generate an intermediate table, and saves it to a second data warehouse. This allows the stream computing engine to directly call the data when performing real-time fault detection without having to calculate the features of the historical operating data separately, reducing the waiting time for real-time calculations, improving the computational efficiency in the real-time fault detection process, and enhancing the rapid response capability in real-time fault risk identification and early warning scenarios.

[0120] For example, the second data warehouse is a real-time data warehouse, such as an HBase warehouse. Since the real-time operating data of the battery is continuously updated, in order to enable the stream computing platform to acquire and process the real-time operating data generated by the battery in real time, the system of this embodiment can store the intermediate parameters calculated by the batch computing engine and the real-time operating data of the battery obtained from the terminal in real time through the second data warehouse, thereby providing a guarantee for the stream computing engine to read data during the fusion computing process.

[0121] Therefore, the second data warehouse uses an HBase warehouse to obtain real-time operational data for each frame of the battery, updating the corresponding real-time status data for the streaming computing engine to use. This allows the second fault detection model to fuse relevant feature data based on historical battery operational data with each frame of real-time operational data to obtain more accurate and reliable real-time fault detection results, thus facilitating real-time fault risk prediction. For example, real-time battery operational data may include, but is not limited to, real-time battery voltage, current, temperature, state of charge (SOC), and state of health (SOH) values. Correspondingly, the real-time battery status data stored in the second data warehouse may include voltage, current, temperature, SOC, and SOH values.

[0122] For example, the data acquired by the first and second data warehouses can be collected by a stream computing platform. This platform can be a processing platform including the Flink stream computing engine (hereinafter referred to as the "Flink platform"). This enables the continuous and real-time acquisition of the terminal's real-time battery operation data, which is then stored in the first and second data warehouses respectively. In the second data warehouse, new real-time operation data overwrites existing data, achieving continuous data updates. In the first data warehouse, both new real-time operation data and existing historical battery operation data are stored, allowing for continuous data accumulation and overall data updates.

[0123] Specifically, in a real-time fault detection process, the first fault detection model retrieves historical operational data from the first data warehouse for feature extraction, obtains corresponding fault detection feature data, performs batch calculations, and generates intermediate parameters for this historical operational data, which are then stored in the second data warehouse for fusion calculation together with the real-time operational data retrieved from the second data warehouse. Simultaneously, the real-time operational data retrieved in this instance is stored in the first data warehouse for use in the next real-time fault detection.

[0124] Furthermore, based on the Flink platform's stream computing engine, during the fusion computing process, it can continuously and systematically acquire the aforementioned real-time operational data and intermediate parameters updated over time from the second data warehouse using a stream computing approach, and then perform data fusion computing. In this way, the results of the fusion computing can retain the temporal attributes of both historical and real-time battery operational data, making the fault detection results strongly correlated with the time sequence of the data. This aligns with the objective laws governing changes in battery operating status over time, thereby improving the accuracy of fault detection results.

[0125] In some embodiments, step S104 uses a stream computing engine to acquire the intermediate parameters and real-time battery operating data in real time, and uses a second fault detection model to perform fusion calculation on the real-time battery status data and intermediate parameters to obtain the corresponding battery fault detection result, which may specifically include:

[0126] The streaming computing engine retrieves intermediate parameters and real-time battery operating data from the second data warehouse in real time.

[0127] Based on the real-time battery operating data, determine the corresponding real-time battery status data;

[0128] The real-time battery status data is concatenated with the intermediate parameters in the intermediate table;

[0129] The pre-set second fault detection model is used to calculate the spliced ​​data and output the corresponding battery fault detection data and battery identification information.

[0130] In this embodiment, a stream computing engine continuously and in real-time acquires intermediate parameters and real-time battery operating data from a second data warehouse using a stream computing approach. Based on this real-time battery operating data, the corresponding real-time battery status data is determined. The real-time battery status data can be stored in the second data warehouse in the form of a table. Therefore, during concatenation, the table of real-time battery status data and the aforementioned intermediate table in the second data warehouse can be combined to form a new data table corresponding to the time sequence. Based on this new data table, with the support of the stream computing engine, the second fault detection model can continuously read the data in this table for calculation, obtaining the fault detection data and identity information (such as IdentityDocument, ID) for each battery. This facilitates obtaining the corresponding fault detection data for each battery in real-time in big data battery fault risk identification and management scenarios, improving the response capability to battery fault risks. Furthermore, based on the support of the stream computing engine, the time sequence attributes of historical and real-time battery operating data can be preserved, making the fault detection results strongly correlated with the time sequence of the data, conforming to the objective law of battery operating status changes over time, and improving the accuracy of the fault detection results.

[0131] In some specific examples, the second fault detection model can be a pre-built matrix model, neural network model, or decision tree model, etc. The construction method is similar to that of the first fault detection model, and will not be repeated here.

[0132] In this embodiment, the spliced ​​real-time battery status data and intermediate parameters are input into the corresponding second fault detection model for training and decision-making. This model can output fault detection data and identification information for each battery. For example, the spliced ​​historical and real-time battery IDs, current values, voltage values, fault codes, SOC values, etc., are input into the model, which outputs data such as whether each battery ID has a fault risk, the fault code for any fault risk that has occurred, etc. This allows for the acquisition of fault detection data for each battery in a big data battery risk identification and management scenario.

[0133] To achieve early warning, optionally, in some embodiments, after obtaining the fault detection result of the corresponding battery, the method may further include:

[0134] Based on the fault detection data and battery identification information of the corresponding battery, a warning message is generated;

[0135] The warning information is sent to the service subsystem, which then queries the target terminal corresponding to the battery's identity information based on the warning information and generates service information for the corresponding target terminal and fault detection data.

[0136] For example, the service subsystem can be the operator's after-sales service system. For instance, if the data output by the second fault detection model indicates that a power battery with battery ID "123" is at risk of thermal runaway, the generated warning model information can include the battery ID and a fault risk identifier. The warning information is sent to the service subsystem in real time. Upon receiving the warning information, the service subsystem can locate the target terminal corresponding to the battery ID and send relevant service information to the target terminal for user awareness. This allows the terminal to intercept the corresponding battery risk in real time and instruct risk intervention, such as replacement. This achieves a closed-loop process in the detection and warning of big data battery fault risks, resulting in a low false alarm rate and high reliability.

[0137] The fault detection method provided in this application is particularly suitable for fault risk identification and early warning of power batteries in new energy vehicles based on a big data platform. It requires minimal modification to the Battery Management System (BMS) hardware, resulting in lower costs for fault risk identification and early warning. Furthermore, the offline batch processing engine can train historical operating data, optimize fault detection feature data parameters, and improve the accuracy of fault feature identification. By sampling each frame of real-time operating data using stream computing and performing fusion calculations, more reliable fault detection results are obtained, providing early warning protection for the safety of users and vehicle batteries.

[0138] This application also proposes an electronic device, such as Figure 4 The diagram illustrates the structure of an electronic device according to an embodiment of this application. The electronic device includes a processor 401 and a memory 402 storing computer program instructions.

[0139] When the processor 401 executes computer program instructions, it implements the fault detection method as described in any of the above embodiments.

[0140] Specifically, the processor 401 may include a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits that can be configured to implement the embodiments of this application.

[0141] Memory 402 may include mass storage for data or instructions. For example, and not limitingly, memory 402 may include a hard disk drive (HDD), floppy disk drive, flash memory, optical disk, magneto-optical disk, magnetic tape, or Universal Serial Bus (USB) drive, or a combination of two or more of these. Where appropriate, memory 402 may include removable or non-removable (or fixed) media. Where appropriate, memory 402 may be internal or external to the integrated gateway disaster recovery device. In a particular embodiment, memory 402 is non-volatile solid-state memory.

[0142] Memory 402 may include read-only memory (ROM), random access memory (RAM), disk storage media device, optical storage media device, flash memory device, electrical, optical, or other physical / tangible memory storage device. Therefore, typically, memory includes one or more tangible (non-transitory) computer-readable storage media (e.g., memory devices) encoded with software including computer-executable instructions, and when the software is executed (e.g., by one or more processors), it is operable to perform the operations described with reference to the method according to one aspect of this application.

[0143] In one example, the electronic device may also include a communication interface 403 and a bus 410. For example, Figure 4 As shown, the processor 401, memory 402, and communication interface 403 are connected through bus 410 and complete communication with each other.

[0144] The communication interface 403 is mainly used to realize communication between various modules, devices, units and / or equipment in the embodiments of this application.

[0145] Bus 410 includes hardware, software, or both, that couples components of an electronic device together. For example, and not limitingly, the bus may include an Accelerated Graphics Port (AGP) or other graphics bus, an Enhanced Industry Standard Architecture (EISA) bus, a Front Side Bus (FSB), HyperTransport (HT) interconnect, an Industry Standard Architecture (ISA) bus, an Infinite Bandwidth Interconnect, a Low Pin Count (LPC) bus, a memory bus, a Microchannel Architecture (MCA) bus, a Peripheral Component Interconnect (PCI) bus, a PCI-Express (PCI-X) bus, a Serial Advanced Technology Attachment (SATA) bus, a Video Electronics Standards Association Local (VLB) bus, or other suitable buses, or combinations of two or more of these. Where appropriate, bus 410 may include one or more buses. Although specific buses are described and illustrated in embodiments of this application, this application contemplates any suitable bus or interconnect.

[0146] In addition, in conjunction with the fault detection method in the above embodiments, this application also provides a readable storage medium on which a program or instruction is stored. When the program or instruction is executed by a processor, it implements the steps of the battery fault detection method as described in the above embodiments.

[0147] Furthermore, this application embodiment also provides a chip, which includes a processor and a communication interface. The communication interface and the processor are coupled, and the processor is used to run programs or instructions to implement the steps of the battery fault detection method as described in the above embodiments.

[0148] Furthermore, this application also provides a computer program product in which the instructions are executed by the processor of an electronic device, causing the electronic device to perform the steps of the battery fault detection method as described in the above embodiments.

[0149] It should be clarified that this application is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of this application is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of this application.

[0150] The functional modules shown in the above-described block diagram can be implemented as hardware, software, firmware, or a combination thereof. When implemented in hardware, they can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this application are programs or code segments used to perform the required tasks. Programs or code segments can be stored on a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried on a carrier wave. "Machine-readable medium" can include any medium capable of storing or transmitting information. Examples of machine-readable media include electronic circuits, semiconductor memory devices, ROM, flash memory, erasable ROM (EROM), floppy disks, CD-ROMs, optical disks, hard disks, fiber optic media, radio frequency (RF) links, etc. Code segments can be downloaded via computer networks such as the Internet, intranets, etc.

[0151] It should also be noted that the exemplary embodiments mentioned in this application describe methods or systems based on a series of steps or apparatus. However, this application is not limited to the order of the above steps; that is, the steps can be performed in the order mentioned in the embodiments, or in a different order, or several steps can be performed simultaneously.

[0152] Although this application has been described with reference to preferred embodiments, various modifications can be made thereto and components can be replaced with equivalents without departing from the scope of this application. In particular, the technical features mentioned in the various embodiments can be combined in any manner, provided there is no structural conflict. This application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A battery fault detection system, comprising: The first data warehouse is used to store historical battery operating data; The fault detection module is used to obtain the corresponding historical battery operation data from the first data warehouse for feature extraction to obtain the corresponding fault detection feature data. Batch calculation module: used to perform batch calculations on the fault detection feature data through a batch calculation engine to obtain the corresponding intermediate parameters; The stream computing platform is used to obtain the intermediate parameters and real-time battery operating data in real time through the stream computing engine, perform fusion calculations, and obtain the corresponding battery fault detection results. The system also includes a second data warehouse. The second data warehouse is used to store the intermediate parameters and the real-time operating data of the battery in real time, so that the streaming computing platform can obtain them; The stream computing platform also includes: The real-time acquisition module is used to acquire the intermediate parameters and real-time battery operation data from the second data warehouse in real time through the stream computing engine; The determination module is used to determine the corresponding real-time battery status data based on the real-time battery operating data. The splicing module is used to splice the real-time operating data of the battery with intermediate parameters in an intermediate table, wherein the intermediate table is an intermediate table with a corresponding time sequence generated based on the intermediate parameters. The calculation module is used to calculate the spliced ​​data using a preset second fault detection model, and output the corresponding battery fault detection data and the battery's identity information.

2. The system according to claim 1, wherein, The stream computing platform includes: The data interface is used to collect real-time operating data of the corresponding battery from various terminals via the Internet of Things protocol; The first transmission unit is used to transmit the real-time running data to the second data warehouse to update the existing real-time running data of the battery stored in the second data warehouse; The second transmission unit is used to transmit the real-time operating data to the first data warehouse for storage, so as to update the battery historical operating data.

3. The system according to claim 1, wherein, The fault detection module includes: The data acquisition unit is used to acquire the battery historical operation data corresponding to the sampling period from the first data warehouse at a preset sampling time, and input it into the feature extraction unit in time sequence; The feature extraction unit is used to extract multi-dimensional features from the battery's historical operating data using a preset first fault detection model, and output the corresponding fault detection feature data.

4. The system according to claim 3, wherein, The batch calculation module includes: The first calculation unit is used to perform batch calculations on the fault detection feature data through a batch calculation engine to obtain the corresponding intermediate parameters. The generation unit is used to generate an intermediate table corresponding to the timing based on the intermediate parameters. The third transmission unit is used to transmit the intermediate table to the second data warehouse to update the existing intermediate table stored in the second data warehouse.

5. The system according to claim 1, wherein, The system also includes an early warning module and a service subsystem. The early warning module is used to generate early warning information and send it to the service subsystem based on the fault detection data of the corresponding battery and the battery's identity information. The service subsystem is used to query the target terminal corresponding to the battery's identity information based on the warning information, so as to generate service information corresponding to the target terminal and the fault detection data.

6. A battery fault detection method, comprising: Retrieve the corresponding historical battery operation data from the first data warehouse; The battery's historical operating data is used to extract features using a preset first fault detection model to obtain corresponding fault detection feature data. The fault detection feature data is processed in batches using a batch computing engine to obtain the corresponding intermediate parameters. The intermediate parameters and real-time battery operating data are obtained in real time through the stream computing engine, and the second fault detection model is used for fusion calculation to obtain the fault detection results of the corresponding battery. The process involves using a stream computing engine to acquire intermediate parameters and real-time battery operating data in real time, and then using a second fault detection model for fusion calculation to obtain the corresponding battery fault detection results, including: The intermediate parameters and real-time battery operating data are obtained in real time from the second data warehouse through the stream computing engine. The second data warehouse is used to store the intermediate parameters and real-time battery operating data in real time. Based on the real-time battery operating data, determine the corresponding real-time battery status data; The real-time battery status data is concatenated with intermediate parameters in an intermediate table, which is an intermediate table with a corresponding time sequence generated based on the intermediate parameters. The pre-set second fault detection model is used to calculate the spliced ​​data and output the corresponding battery fault detection data and the battery's identity information.

7. The method according to claim 6, wherein, Before performing feature extraction on the battery's historical operating data using a preset first fault detection model, the method further includes: Obtain historical battery operation sample data within the target duration from the first data warehouse; Multi-dimensional feature extraction is performed on the historical operating sample data of the battery to construct the first fault detection model corresponding to the fault features.

8. The method according to claim 6, wherein, The step of extracting features from the battery's historical operating data using a preset first fault detection model to obtain corresponding fault detection feature data includes: At the preset sampling time, the historical battery operation data of the corresponding sampling period is input into the first fault detection model in chronological order; The first fault detection model is used to extract multi-dimensional features from the battery's historical operating data to obtain corresponding fault detection feature data.

9. The method according to claim 8, wherein, After performing batch calculations on the fault detection feature data using a batch computing engine to obtain the corresponding intermediate parameters, the method further includes: Based on the intermediate parameters, an intermediate table corresponding to the timing sequence is generated; The intermediate table is input into a preset second data warehouse to update the existing intermediate table stored in the second data warehouse, which is a real-time data warehouse.

10. The method according to claim 8, wherein, After obtaining the fault detection result of the corresponding battery, the method further includes: Based on the fault detection data of the corresponding battery and the battery's identity information, a warning message is generated; The warning information is sent to the service subsystem, so that the service subsystem can query the target terminal corresponding to the battery's identity information based on the warning information, and generate service information corresponding to the target terminal and the fault detection data.

11. An electronic device comprising a processor, a memory, and a program or instructions stored in the memory and executable on the processor, wherein the program or instructions, when executed by the processor, implement the steps of the battery fault detection method as described in any one of claims 6-10.

12. A readable storage medium storing a program or instructions that, when executed by a processor, implement the steps of the battery fault detection method as described in any one of claims 6-10.

13. A chip, the chip comprising a processor and a communication interface, the communication interface being coupled to the processor, the processor being configured to run a program or instructions to implement the steps of the battery fault detection method as described in any one of claims 6-10.

14. A computer program product, wherein instructions in the computer program product, when executed by a processor of an electronic device, cause the electronic device to perform the steps of the battery fault detection method as described in any one of claims 6-10.

Images