Metadata-driven industrial energy storage intelligent diagnosis method and related device

Abstract

The invention discloses an industrial energy storage intelligent diagnosis method based on metadata driving and a related device, and belongs to the technical field of industrial energy storage technology and intelligent diagnosis. The method comprises the following steps: carrying out preliminary analysis on an original communication data flow to obtain analysis information; processing the analysis information by using a statistical analysis algorithm to obtain communication protocol features; generating a data analysis rule according to the communication protocol definition information, the message structure and the field distribution rule, and performing normalization processing on the original communication data flow according to the data analysis rule to obtain standardized data; performing comparison detection on the communication protocol definition information, the standardized data and the normal data mode to obtain an anomaly detection result; according to the anomaly detection result, anomaly inference is carried out based on a preset knowledge base, an anomaly diagnosis report is obtained, and intelligent diagnosis is completed. The problem that in the prior art, when equipment is accessed, codes are written depending on a specific protocol for analysis can be solved.

Description

Technical Field

[0001] This invention belongs to the field of industrial energy storage technology and intelligent diagnostic technology, specifically relating to a metadata-driven intelligent diagnostic method and related device for industrial energy storage. Background Technology

[0002] With the rapid development of integrated cloud platforms for industrial, commercial, and energy storage (industrial, commercial, and energy storage), an increasing number of energy storage devices (such as Battery Management Systems (BMS), Power Conversion Systems (PCS), Energy Management Systems (EMS), and photovoltaic inverters) need to be connected to a unified cloud platform for centralized monitoring and management. However, these devices often come from different manufacturers, employ diverse communication protocols (such as Modbus RTU / TCP, DL / T645, and various proprietary protocols), and there may be subtle differences in protocol versions, data point mappings, and data formats.

[0003] Traditional device access methods typically rely on manual configuration of point tables and parameters, or writing code for specific protocols for parsing. This approach exhibits certain limitations in practical applications. Regarding integration efficiency, when connecting a large number of heterogeneous devices or adding new device types, manual configuration is labor-intensive, time-consuming, and prone to errors, thus slowing down project deployment cycles. In terms of maintenance costs, once device protocols are updated or new device models need to be adapted, it often involves modifying underlying code or reconfiguring numerous parameters, resulting in complex and costly maintenance. Regarding fault diagnosis, when the system experiences data anomalies or communication failures, locating the root cause is complex, potentially involving hardware failures in the device itself, communication link problems, protocol mismatches, data parsing errors, and other factors, prolonging troubleshooting time and impacting system stability. Furthermore, existing systems typically use static configuration methods, lacking intelligent recognition capabilities for protocol differences, making adaptive adjustment and iterative optimization difficult.

[0004] Therefore, the industry urgently needs a solution that can efficiently and automatically handle communication protocols of heterogeneous devices and intelligently diagnose protocol mismatch issues in order to improve the integration efficiency, data reliability, and intelligent operation and maintenance level of industrial and commercial cloud storage platforms. Summary of the Invention

[0005] The purpose of this invention is to provide a metadata-driven intelligent diagnostic method and related device for industrial energy storage, so as to solve the problem that existing technologies rely on writing code based on specific protocols for parsing when connecting equipment.

[0006] To achieve the above objectives, the present invention adopts the following technical solution: Firstly, a metadata-driven intelligent diagnostic method for industrial energy storage includes the following steps: Obtain communication protocol definition information of heterogeneous devices, collect raw communication data streams at the device layer in real time, perform preliminary parsing of the raw communication data streams, and obtain parsed information; The parsed information is processed using statistical analysis algorithms to obtain communication protocol characteristics, which include: message structure, field distribution patterns, and normal data patterns. Data parsing rules are generated based on the communication protocol definition information, message structure and field distribution patterns, and the original communication data stream is normalized according to the data parsing rules to obtain standardized data; The anomaly detection results are obtained by comparing and detecting communication protocol definition information, standardized data, and normal data patterns. Based on the anomaly detection results and the anomaly inference based on the preset knowledge base, an anomaly diagnosis report is obtained, and intelligent diagnosis is completed.

[0007] In some implementations, the following steps are also included: Based on the anomaly diagnosis report, optimization suggestions are generated and data parsing rules and communication protocol definition information are adjusted.

[0008] In some implementations, the step of obtaining communication protocol characteristics by processing the parsed information using statistical analysis algorithms specifically includes: Extract message structure and field distribution patterns from parsed information; Clustering algorithms are used to identify field distribution characteristics in the patterns of field distribution. Based on the field distribution characteristics, time series analysis is used to extract the data change trends under normal operating conditions, forming a normal data pattern.

[0009] In some implementations, the step of comparing and detecting communication protocol definition information, standardized data, and normal data patterns to obtain anomaly detection results specifically includes: Each data point in the standardized data is compared with the corresponding data type and numerical range in the communication protocol definition information to identify the first abnormal data point and filter the first normal data point. The first normal data point is compared with the corresponding time-series change distribution in the normal data pattern to identify the second abnormal data point; The logical relationships between multiple related standardized data points within the same time period are compared with the corresponding field association rules in the normal data pattern to identify the third abnormal data point; The first, second, and third abnormal data points are taken as the anomaly detection results.

[0010] In some implementations, the step of obtaining an anomaly diagnosis report based on the anomaly detection results and anomaly inference based on a preset knowledge base specifically includes: Based on the anomaly detection results, combined with the device failure modes, communication protocol version compatibility information and common problem solutions in the preset knowledge base, anomaly inference is made to obtain at least one of the following problems: protocol version incompatibility problem, parameter configuration error problem, device hardware failure problem and network link problem; An anomaly diagnostic report is generated based on at least one of the following issues: protocol version incompatibility, parameter configuration errors, device hardware failure, and network link problems.

[0011] In some implementations, the step of normalizing the original communication data stream according to data parsing rules to obtain standardized data specifically includes: According to the data parsing rules, the binary or text format data in the original communication data stream is converted into standardized physical quantity data; By defining the gain coefficient, offset, and unit conversion rules in the information through the communication protocol, the standardized physical quantity data is converted and calibrated to obtain standardized data.

[0012] Secondly, a metadata-driven intelligent diagnostic system for industrial energy storage includes: The data acquisition module is used to obtain the communication protocol definition information of heterogeneous devices, collect the raw communication data stream of the device layer in real time, perform preliminary parsing of the raw communication data stream, and obtain parsed information. The protocol feature learning module is used to process the parsed information using statistical analysis algorithms to obtain communication protocol features, which include: message structure, field distribution patterns, and normal data patterns. The dynamic adaptation module is used to generate data parsing rules based on communication protocol definition information, message structure and field distribution patterns, and to normalize the original communication data stream according to the data parsing rules to obtain standardized data. The anomaly detection module is used to compare and detect communication protocol definition information, standardized data and normal data patterns to obtain anomaly detection results. The intelligent diagnosis module is used to make anomaly inferences based on the anomaly detection results and a preset knowledge base, and to obtain an anomaly diagnosis report to complete the intelligent diagnosis.

[0013] In some implementations, it also includes: a metadata storage module, an optimization feedback module, a user interaction module, and a data transmission module; The metadata storage module, which communicates with the dynamic adaptation module, the anomaly detection module, the protocol feature learning module, the optimization feedback module, and the user interaction module, is used to store the communication protocol definition information of the heterogeneous devices in a structured form. The optimization feedback module is used to generate optimization suggestions and adjust the data parsing rules and communication protocol definition information based on the anomaly diagnosis report; The user interaction module is communicatively connected to the metadata storage module, the anomaly detection module, and the intelligent diagnosis module. It is used to provide a human-machine interaction interface and receive metadata configuration instructions, device status monitoring instructions, alarm information viewing instructions, and optimization suggestion confirmation instructions. The data transmission module communicates with the data acquisition module, metadata storage module, protocol feature learning module, dynamic adaptation module, anomaly detection module, intelligent diagnosis module, optimization feedback module, and user interaction module, and is used to asynchronously transmit data and instructions between the modules using a distributed message middleware.

[0014] Thirdly, a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of a metadata-driven intelligent diagnostic method for industrial energy storage.

[0015] Fourthly, a computer program product comprising a computer program that, when executed by a processor, implements steps of a metadata-driven intelligent diagnostic method for industrial energy storage.

[0016] Compared with the prior art, the present invention has the following beneficial effects: This invention provides a metadata-driven intelligent diagnostic method for industrial energy storage. By introducing the synergy of two core features—"acquiring communication protocol definition information" and "obtaining communication protocol features based on parsed information using statistical analysis algorithms"—a complete adaptive diagnostic process is constructed. Specifically, statistical analysis algorithms are used to process the parsed information to obtain communication protocol features, thereby dynamically generating data parsing rules. This fundamentally changes the problem of manually writing dedicated parsing code for each protocol in traditional access methods, significantly improving the automation level and efficiency of heterogeneous device access. Simultaneously, by comparing and detecting communication protocol definition information, standardized data, and normal data patterns, multi-dimensional anomaly detection is achieved. Combined with a pre-set knowledge base for anomaly inference, this shortens the overall fault diagnosis time and reduces operation and maintenance costs.

[0017] Furthermore, based on the anomaly diagnosis report, optimization suggestions are generated and data parsing rules and communication protocol definition information are adjusted. When the system has diagnostic deviations due to protocol version updates, configuration errors, or incomplete initial learning, it can assist in correcting the data parsing rules and communication protocol definition information, effectively solving the maintenance problem of having to manually modify the code again after the protocol changes. This significantly improves the long-term adaptability and stability of the system and reduces the maintenance burden throughout the entire lifecycle.

[0018] Furthermore, by using clustering algorithms to identify field distribution characteristics in the field distribution patterns, the clustering patterns of field values ​​can be automatically discovered from communication data in an unsupervised manner, and the field distribution characteristics can be objectively and quantitatively extracted. Based on the field distribution characteristics, the data change trends under normal operating conditions can be extracted through time series analysis, which can accurately characterize the fluctuation trends and reasonable ranges of physical quantities under normal operating conditions, thereby establishing a dynamic normal data model.

[0019] Furthermore, by comparing data points in standardized data with the data types and numerical ranges defined in communication protocol information, the first normal data point can be quickly and compliantly screened based on hard rules. By comparing the first normal data point with the temporal variation distribution in normal data patterns, hidden anomalies that are numerically compliant but whose behavior deviates from historical patterns can be identified. By comparing the logical relationships between multiple related data points with the predefined field association rules in normal data patterns, complex faults with system-level logical contradictions can be detected. This detection method can greatly expand the coverage of anomaly detection, effectively balance detection speed and depth, and significantly improve the detection rate and diagnostic accuracy of various types of faults. Attached Figure Description

[0020] Figure 1 A flowchart illustrating a metadata-driven intelligent diagnostic method for industrial energy storage provided in an embodiment of the present invention; Figure 2 A schematic diagram of the data processing process of a metadata-driven intelligent diagnostic method for industrial energy storage provided in an embodiment of the present invention; Figure 3 A schematic diagram of the specific structure of an industrial energy storage intelligent diagnostic system based on metadata-driven technology provided in an embodiment of the present invention; Figure 4 A flowchart illustrating the workflow of the dynamic adaptation module provided in an embodiment of the present invention; Figure 5 The diagram shows the operational logic of the anomaly detection module provided in this embodiment of the invention. Detailed Implementation

[0021] To enable those skilled in the art to better understand the present invention, the technical solution of the present invention will be further described in detail below with reference to the accompanying drawings. The content is for explanation rather than limitation of the present invention.

[0022] It should be noted that the terms "comprising" and "having" and any variations thereof in the specification and claims of this invention are intended to cover a non-exclusive inclusion. For example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units that are explicitly listed, but may include other steps or units that are not explicitly listed or that are inherent to such processes, methods, systems, products, or devices.

[0023] like Figure 1 and Figure 2 As shown, this embodiment provides a metadata-driven intelligent diagnostic method for industrial energy storage, including the following steps: S1, obtain the communication protocol definition information of heterogeneous devices, collect the raw communication data stream of the device layer in real time, perform preliminary parsing of the raw communication data stream, and obtain parsing information; Specifically, the communication protocol definition information of heterogeneous devices is centrally managed through the metadata storage unit, including device model, communication parameters, data point mapping relationship, numerical range, unit conversion rules and alarm thresholds; the raw communication data stream of the device layer is collected in real time and preliminarily parsed to extract message frame header, frame tail, check code and basic field information.

[0024] S2, using statistical analysis algorithms to process the parsed information, obtains communication protocol characteristics, including: message structure, field distribution patterns, and normal data patterns; Specifically, based on historical communication data and real-time collected data, statistical analysis algorithms are used to identify the communication protocol characteristics of the device, including message structure, field distribution patterns, and data patterns under normal operating conditions. S3 generates data parsing rules based on the communication protocol definition information, message structure, and field distribution patterns, and normalizes the original communication data stream according to the data parsing rules to obtain standardized data; Specifically, data parsing rules are dynamically generated or adjusted to convert binary or text format data in the original communication data stream into standardized physical quantity data, while applying gain coefficients, offsets, and unit conversion rules to complete data normalization processing. S4 compares and detects the communication protocol definition information, standardized data, and normal data patterns to obtain anomaly detection results; Specifically, the parsed standardized data is compared with the specification definition in the metadata storage unit and the normal data pattern constructed by the protocol feature learning module to detect data type matching, reasonableness of numerical range and logical consistency. S5, based on the anomaly detection results and the pre-set knowledge base, makes anomaly inferences, obtains an anomaly diagnosis report, and completes intelligent diagnosis.

[0025] Specifically, based on the anomaly detection results, combined with the built-in knowledge base and context information, the root cause of the anomaly is inferred, including protocol version incompatibility, parameter configuration errors, device hardware failure, or network link problems. Based on the diagnostic results, optimization suggestions are generated and relevant parameters or dynamic adaptation rules in the metadata storage unit are automatically adjusted to achieve self-iterative optimization of the system.

[0026] The communication protocol definition information of heterogeneous devices is centrally managed through a metadata storage unit, including: storing the message structure, field distribution patterns, and numerical ranges corresponding to the device model through a relational database or NoSQL database; supporting add, modify, and delete operations, and storing the communication protocol definition information in a structured form.

[0027] The communication protocol characteristics of devices are identified using statistical analysis algorithms, including: extracting field distribution patterns from historical communication data and using clustering algorithms to identify field distribution characteristics; and extracting data change trends under normal operating conditions through time series analysis to form normal data patterns.

[0028] Dynamically generate or adjust data parsing rules, including: generating parsing rules based on the message structure and field distribution patterns provided by the metadata storage unit; and adjusting parsing rules based on the learning results of the protocol feature learning module to ensure that the parsing rules are adapted to the communication protocol of the current device.

[0029] The detection of data type matching, numerical range rationality, and logical consistency includes: checking whether a single data point conforms to the data type and numerical range defined in the metadata storage unit; and analyzing the logical relationships between multiple data points to determine whether there are any unreasonable associations.

[0030] Inferring the root cause of anomalies includes: combining device failure modes, communication protocol version compatibility information, and common problem solutions from the built-in knowledge base to distinguish between protocol version incompatibility, parameter configuration errors, device hardware failures, or network link problems.

[0031] like Figure 3 As shown, this embodiment provides a metadata-driven intelligent diagnostic system for industrial energy storage, including: The data acquisition module is used to acquire raw communication data streams from the device layer in real time and perform preliminary parsing and processing to ensure the integrity and correctness of the data frames; The metadata storage module is used to centrally manage the communication protocol definition information of heterogeneous devices, supports adding, modifying and deleting operations, and stores it in a structured form; The protocol feature learning module is used to extract the communication protocol features of the device from historical communication data or real-time acquired data, including message structure, field distribution patterns, and data patterns under normal operating conditions. The dynamic adaptation module is used to dynamically generate or adjust data parsing rules and complete data normalization processing based on the standard definitions provided by the metadata storage module and the learning results of the protocol feature learning module. The anomaly detection module is used to compare the standardized data output by the dynamic adaptation module with the specification definitions in the metadata storage module and the normal data pattern constructed by the protocol feature learning module to detect data anomalies. The intelligent diagnosis module is used to infer the root cause of the anomaly and generate a diagnostic report based on the results of the anomaly detection module, combined with the built-in knowledge base and contextual information. The optimization feedback module is used to generate optimization suggestions based on the diagnostic results of the intelligent diagnostic module and automatically adjust the relevant parameters in the metadata storage module or the parsing rules of the dynamic adaptation module. The user interaction module provides a user interface and supports applications such as metadata configuration, device status monitoring, alarm viewing, and optimization suggestions. The data transmission module, serving as the core communication channel between modules, employs a distributed message middleware to achieve asynchronous data transmission, ensuring decoupling and high reliability between modules.

[0032] In practical applications, industrial energy storage devices typically consist of multiple heterogeneous devices that connect to the industrial and commercial energy storage cloud platform via different communication protocols. To address the issues of low efficiency in handling protocol differences and frequent data parsing errors when heterogeneous devices connect, this embodiment provides a complete solution. First, the data acquisition module is responsible for acquiring raw communication data streams from the device layer in real time and performing preliminary parsing to ensure the integrity and correctness of the data frames. The data acquisition module connects to the device layer via physical interfaces, such as RS485, Modbus, or TCP / IP, to acquire the raw message data sent by the devices. After preliminary parsing, the acquired data extracts the message frame header, frame trailer, checksum, and basic field information. This information is then transmitted to the data transmission module as the basis for subsequent processing.

[0033] The metadata storage module centrally manages the communication protocol definition information of heterogeneous devices, including device model, communication parameters, data point mapping relationships, numerical ranges, unit conversion rules, and alarm thresholds. The metadata storage module uses a relational database or NoSQL database for data storage, supports add, modify, and delete operations, and stores the above information in a structured format. The metadata storage module communicates with other modules through the data transmission module, ensuring that each module can efficiently access the required metadata information. For example, when the dynamic adaptation module needs to generate parsing rules based on the communication protocol of a certain device, it will request relevant parameters from the metadata storage module through the data transmission module, such as the message structure or field distribution patterns corresponding to the device model.

[0034] The protocol feature learning module extracts communication protocol features of the device from historical communication data or real-time acquired data, including message structure, field distribution patterns, and data patterns under normal operating conditions. The module processes the data using statistical analysis algorithms, such as clustering to identify field distribution patterns or time series analysis to extract data change trends under normal operating conditions. Connected to the data transmission module, the module receives raw data from the data acquisition module and outputs the learning results to the dynamic adaptation module, providing a basis for generating parsing rules.

[0035] The dynamic adaptation module dynamically generates or adjusts data parsing rules based on the standard definitions provided by the metadata storage module and the learning results of the protocol feature learning module. Figure 4 The workflow of the dynamic adaptation module is demonstrated. First, it retrieves communication protocol definition information from the metadata storage module. Then, combining the learning results from the protocol feature learning module, it generates parsing rules suitable for the current device. These parsing rules include how to convert binary or text-formatted data in the raw communication data stream into standardized physical quantity data, while applying gain coefficients, offsets, and unit conversion rules to complete data normalization. The dynamic adaptation module then transmits the standardized data to the anomaly detection module for further processing via the data transmission module.

[0036] The anomaly detection module compares the standardized data output by the dynamic adaptation module with the specification definitions in the metadata storage module and the normal data patterns constructed by the protocol feature learning module to detect data type matching, reasonableness of numerical range, and logical consistency. Figure 5The operational logic of the anomaly detection module is demonstrated. First, it performs multi-dimensional comparisons on individual data points, such as checking whether a data point exceeds a preset numerical range or conforms to the data type defined in the metadata. Second, the anomaly detection module analyzes the logical relationships between multiple data points, such as determining whether there is a reasonable correlation between the temperature data and power data of a device. The anomaly detection module then transmits the detection results to the intelligent diagnostic module via the data transmission module.

[0037] Based on the results from the anomaly detection module, the intelligent diagnostic module, combined with its built-in knowledge base and contextual information, infers the root cause of the anomaly and generates a diagnostic report. The built-in knowledge base includes device failure modes, communication protocol version compatibility information, and solutions to common problems. The intelligent diagnostic module uses rule-based reasoning and pattern matching techniques to distinguish between different types of anomalies, such as protocol version incompatibility, incorrect parameter configuration, device hardware failure, or network link problems. The intelligent diagnostic module then transmits the diagnostic results to the optimization feedback module and sends alarm information to the user interaction module via the data transmission module.

[0038] Based on the diagnostic results from the intelligent diagnostic module, the optimization feedback module generates optimization suggestions and automatically adjusts relevant parameters in the metadata storage module or the parsing rules of the dynamic adaptation module. For example, when the diagnostic results indicate that a device's communication protocol version is incompatible, the optimization feedback module will suggest updating the device's protocol definition information and write the updated information to the metadata storage module through the data transmission module. Furthermore, the optimization feedback module will also adjust the parsing rules of the dynamic adaptation module to adapt to the new protocol version. This closed-loop feedback mechanism enables the system to continuously learn and self-optimize.

[0039] The user interaction module provides a user-friendly interface, supporting flexible configuration of metadata and intuitive viewing of diagnostic results. Users can configure metadata, monitor device status, view alarms, and apply optimization suggestions through this interface. The user interaction module communicates with other modules via the data transmission module, ensuring that user actions are reflected in the system in real time. For example, when a user modifies an alarm threshold for a device through the user interaction module, the modified information is transmitted to the metadata storage module for updating via the data transmission module.

[0040] The data transmission module, serving as the core communication channel between modules, employs a distributed message middleware to achieve asynchronous data transmission, ensuring decoupling and high reliability between modules. The data transmission module supports multiple communication protocols, such as Kafka or RabbitMQ, avoiding system interruption risks due to single points of failure through asynchronous communication. For example, when the data acquisition module collects new raw data, the data transmission module distributes the data to the protocol feature learning module and the dynamic adaptation module for processing, while ensuring data integrity and timeliness.

[0041] In a specific application scenario, suppose an industrial energy storage system contains multiple heterogeneous devices, some using the Modbus protocol and others using the CAN bus protocol. The data acquisition module connects to these devices via physical interfaces, acquiring raw communication data streams in real time and transmitting the data to the data transmission module. The protocol feature learning module extracts Modbus protocol message structure features from historical data, such as the position and length of the start address field, and the distribution pattern of CAN bus protocol frame identifiers. The dynamic adaptation module generates parsing rules based on these features, converting the raw data into standardized physical quantity data and performing unit conversion and normalization. The anomaly detection module performs multi-dimensional comparisons of the standardized data, discovering that the temperature data of a certain device exceeds the preset range, and that there is an unreasonable relationship between its power data and temperature data. The intelligent diagnosis module, combining a built-in knowledge base and contextual information, infers that the root cause of the anomaly is a hardware failure and generates a diagnostic report. The optimization feedback module suggests replacing the faulty device based on the diagnostic results and writes the updated metadata information to the metadata storage module via the data transmission module. The user interaction module displays alarm information and optimization suggestions through an interface, allowing users to perform device replacement operations based on the prompts.

[0042] As can be seen from the above implementation methods, this embodiment forms a complete closed-loop feedback mechanism through the collaborative work of modules such as multi-level metadata management, communication protocol feature learning, dynamic adaptation rule generation, anomaly detection and intelligent diagnosis, which significantly improves the efficiency and intelligence level of the industrial and commercial cloud storage platform in terms of heterogeneous device access.

[0043] To enable those skilled in the art to fully understand and implement this embodiment, the specific implementation principle of this embodiment will be further explained below in conjunction with a specific application scenario.

[0044] In an industrial energy storage system, assuming it contains multiple heterogeneous devices, some communicating via the Modbus protocol and others via the CAN bus protocol, these devices connect to a data acquisition module via physical interfaces. The data acquisition module acquires the raw communication data stream in real time and performs preliminary parsing. First, the data acquisition module extracts basic fields such as the message frame header, frame trailer, and checksum to ensure the integrity and correctness of the data frame. Then, the preliminarily parsed data is transmitted to the data transmission module, serving as the basis for subsequent collaborative work among the modules.

[0045] The protocol feature learning module extracts the communication protocol features of the device from historical communication data or real-time acquired data. For example, for the Modbus protocol, the module identifies the position and length of the start address field through statistical analysis algorithms; for the CAN bus protocol, it extracts the distribution pattern of frame identifiers and field distribution patterns. This feature information provides the basis for the dynamic adaptation module to generate parsing rules. Based on the standard definitions provided by the metadata storage module and the learning results of the protocol feature learning module, the dynamic adaptation module dynamically generates parsing rules applicable to the current device. Specifically, the dynamic adaptation module converts binary or text format data in the original communication data stream into standardized physical quantity data and applies gain coefficients, offsets, and unit conversion rules to complete data normalization. For example, temperature data may be stored in degrees Celsius, while some devices may output in Kelvin; the dynamic adaptation module automatically performs unit conversion to ensure data consistency.

[0046] The anomaly detection module receives standardized data output from the dynamic adaptation module and compares it with the specification definitions in the metadata storage module and the normal data patterns constructed by the protocol feature learning module. For example, the anomaly detection module checks whether the temperature data of a device exceeds a preset range, while simultaneously analyzing the logical relationship between its power data and temperature data. If the temperature data of a device deviates significantly from the normal range and there is an unreasonable relationship between it and its power data, the anomaly detection module determines that the device is abnormal. This multi-dimensional comparison not only improves the accuracy of anomaly detection but also reduces the possibility of false alarms and false negatives.

[0047] Based on the results from the anomaly detection module, the intelligent diagnostic module, combined with its built-in knowledge base and contextual information, infers the root cause of the anomaly. For example, when a device's temperature data is abnormal and its power data is unreasonable, the intelligent diagnostic module uses rule-based reasoning and pattern matching techniques, combined with the device's operating status and communication link information, to infer that the anomaly is caused by a hardware failure. Subsequently, the intelligent diagnostic module generates a diagnostic report and sends the relevant information to the optimization feedback module via the data transmission module.

[0048] Based on the diagnostic results from the intelligent diagnostic module, the optimization feedback module generates optimization suggestions and automatically adjusts relevant parameters in the metadata storage module or the parsing rules of the dynamic adaptation module. For example, when the diagnostic results indicate that a device's communication protocol version is incompatible, the optimization feedback module suggests updating the device's protocol definition information and writes the updated information to the metadata storage module through the data transmission module. Furthermore, the optimization feedback module also adjusts the parsing rules of the dynamic adaptation module to adapt to new protocol versions. This closed-loop feedback mechanism enables the system to continuously learn and self-optimize.

[0049] The user interaction module displays alarm information and optimization suggestions through a user interface. For example, when a device is diagnosed with a hardware failure, the user interaction module will display specific alarm information on the interface and provide optimization suggestions for replacing the device. Users can intuitively view device status, modify metadata configurations, or apply optimization suggestions through the interface. When a user modifies the alarm threshold of a device through the user interaction module, the modified information will be transmitted to the metadata storage module via the data transmission module for updating, thus reflecting it in the system in real time.

[0050] The data transmission module acts as the core communication channel throughout the process, employing a distributed message middleware (such as Kafka or RabbitMQ) to achieve asynchronous data transmission, ensuring decoupling and high reliability between modules. For example, when the data acquisition module collects new raw data, the data transmission module distributes the data to the protocol feature learning module and the dynamic adaptation module for processing, while ensuring data integrity and timeliness. In this way, even if a module experiences a brief failure, it will not affect the normal operation of the entire system.

[0051] As can be seen from the above steps, this embodiment forms a complete closed-loop feedback mechanism through the collaborative work of modules such as multi-level metadata management, communication protocol feature learning, dynamic adaptation rule generation, anomaly detection, and intelligent diagnosis. This mechanism significantly improves the efficiency and intelligence level of the industrial and commercial cloud storage platform in heterogeneous device access, while also realizing the system's self-iterative optimization capabilities.

[0052] In the technical solution provided in this embodiment, the centralized management function of the metadata storage module enables unified storage and efficient access to the communication protocol definition information of heterogeneous devices, avoiding errors and omissions that may occur during manual configuration. The protocol feature learning module utilizes statistical analysis algorithms to extract communication protocol features from historical and real-time data, forming a deep understanding of device communication behavior and providing accurate parsing basis for the dynamic adaptation module. The dynamic adaptation module dynamically generates or adjusts data parsing rules based on the standard definitions of the metadata storage module and the learning results of the protocol feature learning module, significantly improving the system's adaptability to diverse communication protocols. The anomaly detection module comprehensively detects data anomalies through multi-dimensional comparisons, including data type matching, numerical range rationality, and logical consistency, reducing the possibility of false alarms and missed alarms.

[0053] The intelligent diagnostic module, combining a built-in knowledge base and contextual information, accurately infers the root cause of anomalies, distinguishing between different types of anomalies such as protocol version incompatibility, parameter configuration errors, device hardware failures, or network link problems, significantly shortening troubleshooting time. The optimization feedback module generates optimization suggestions based on the diagnostic results and automatically adjusts relevant parameters of the metadata storage module or dynamically adapts the parsing rules of the module, forming a closed-loop feedback mechanism that enables the system to continuously learn and self-optimize. The user interaction module provides a user-friendly interface, supporting flexible configuration of metadata and intuitive viewing of diagnostic results, further enhancing the system's usability. The data transmission module employs a distributed message middleware, ensuring high concurrency, low latency, and high reliability in data transmission between modules, providing a solid guarantee for the stable operation of the system.

[0054] In its implementation, this embodiment leverages the deep learning capabilities of the protocol feature learning module to construct an accurate model of device communication behavior, providing a scientific basis for the generation of parsing rules in the dynamic adaptation module. The dynamic adaptation module incorporates gain coefficients, offsets, and unit conversion rules during the parsing process to ensure the accuracy of data normalization. The anomaly detection module, through multi-dimensional comparison, not only detects anomalies in single data points but also analyzes the logical relationships between multiple data points, further improving the comprehensiveness and accuracy of detection. The intelligent diagnosis module, combining contextual information and a built-in knowledge base, comprehensively utilizes rule-based reasoning, pattern matching, and machine learning classification algorithms to accurately infer the root cause of anomalies, providing a clear direction for subsequent optimization and adjustments.

[0055] The optimization feedback module, through a closed-loop feedback mechanism, transforms diagnostic results into specific optimization suggestions and automatically adjusts relevant parameters of the metadata storage module or the parsing rules of the dynamic adaptation module, achieving self-iterative optimization of the system. This closed-loop mechanism not only improves the system's robustness but also significantly reduces the need for manual intervention. The user interaction module, with its intuitive interface design, allows users to easily complete operations such as metadata configuration, device status monitoring, and alarm handling, further enhancing the user experience. The data transmission module employs distributed message middleware, such as Kafka or RabbitMQ, ensuring efficient collaboration between modules through asynchronous communication and avoiding the risk of system outages due to single points of failure.

[0056] In summary, this embodiment forms a complete closed-loop feedback mechanism through the collaborative work of modules such as multi-level metadata management, communication protocol feature learning, dynamic adaptation rule generation, anomaly detection, and intelligent diagnosis, which significantly improves the efficiency and intelligence level of the industrial and commercial cloud storage platform in terms of heterogeneous device access.

[0057] The module division in this embodiment of the invention is illustrative and represents only one logical functional division. In actual implementation, other division methods may be used. Furthermore, the functional modules in the various embodiments of the invention can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0058] This embodiment also provides a storage medium, specifically a computer-readable storage medium (Memory). A computer-readable storage medium is a memory device in a computer device used to store programs and data. It is understood that the computer-readable storage medium here can include both the built-in storage medium in the computer device and extended storage media supported by the computer device. The computer-readable storage medium provides storage space that stores the terminal's operating system. Furthermore, this storage space also stores one or more instructions suitable for loading and execution by a processor. These instructions can be one or more computer programs (including program code). It should be noted that the computer-readable storage medium here can be high-speed RAM or non-volatile memory, such as at least one disk storage device. The processor can load and execute one or more instructions stored in the computer-readable storage medium to implement the corresponding steps of the metadata-driven intelligent diagnostic method for industrial energy storage in the above embodiment.

[0059] This embodiment also provides a computer program product, which includes a computer program that, when executed by a processor, implements the corresponding steps of the metadata-driven intelligent diagnostic method for industrial energy storage described in the above embodiment.

[0060] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0061] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0062] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0063] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0064] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the scope of protection of the claims of the present invention.

Claims

1. A metadata-driven intelligent diagnostic method for industrial energy storage, characterized in that, Includes the following steps: Obtain communication protocol definition information of heterogeneous devices, collect raw communication data streams at the device layer in real time, perform preliminary parsing of the raw communication data streams, and obtain parsing information; The parsed information is processed using statistical analysis algorithms to obtain communication protocol characteristics, which include: message structure, field distribution patterns, and normal data patterns. Data parsing rules are generated based on the communication protocol definition information, message structure, and field distribution patterns. The original communication data stream is then normalized according to the data parsing rules to obtain standardized data. The communication protocol definition information, standardized data, and normal data patterns are compared and detected to obtain anomaly detection results. Based on the anomaly detection results and anomaly inference based on a preset knowledge base, an anomaly diagnosis report is obtained, and intelligent diagnosis is completed.

2. The metadata-driven intelligent diagnostic method for industrial energy storage according to claim 1, characterized in that, It also includes the following steps: Based on the anomaly diagnosis report, optimization suggestions are generated and the data parsing rules and communication protocol definition information are adjusted.

3. The metadata-driven intelligent diagnostic method for industrial energy storage according to claim 1, characterized in that, The step of processing the parsed information using a statistical analysis algorithm to obtain the communication protocol features specifically includes: Extract message structure and field distribution patterns from the parsed information; Clustering algorithms are used to identify field distribution features in the field distribution patterns. Based on the field distribution characteristics, time series analysis is used to extract the data change trends under normal operating conditions to form a normal data pattern.

4. The metadata-driven intelligent diagnostic method for industrial energy storage according to claim 1, characterized in that, The step of comparing and detecting the communication protocol definition information, standardized data, and normal data patterns to obtain anomaly detection results specifically includes: Each data point in the standardized data is compared with the corresponding data type and value range in the communication protocol definition information to identify the first abnormal data point and filter the first normal data point. The first normal data point is compared with the corresponding time-series change distribution in the normal data pattern to identify the second abnormal data point; The logical relationship between multiple related standardized data points within the same time period is compared with the corresponding field association rules in the normal data pattern to identify the third abnormal data point; The first abnormal data point, the second abnormal data point, and the third abnormal data point are taken as the abnormal detection results.

5. The metadata-driven intelligent diagnostic method for industrial energy storage according to claim 1, characterized in that, The step of obtaining an anomaly diagnosis report based on the anomaly detection results and anomaly inference based on a preset knowledge base specifically includes: Based on the anomaly detection results, combined with the device failure modes, communication protocol version compatibility information and common problem solutions in the preset knowledge base, anomaly inference is performed to obtain at least one of the following problems: protocol version incompatibility problem, parameter configuration error problem, device hardware failure problem and network link problem. An anomaly diagnostic report is generated based on at least one of the following issues: protocol version incompatibility, parameter configuration errors, device hardware failure, and network link problems.

6. The metadata-driven intelligent diagnostic method for industrial energy storage according to claim 1, characterized in that, The step of normalizing the original communication data stream according to the data parsing rules to obtain standardized data specifically includes: According to the data parsing rules, the binary data or text format data in the original communication data stream is converted into standardized physical quantity data; The standardized physical quantity data is converted and calibrated by using the gain coefficient, offset, and unit conversion rules defined in the communication protocol to obtain standardized data.

7. A metadata-driven intelligent diagnostic system for industrial energy storage, characterized in that, include: The data acquisition module is used to acquire communication protocol definition information of heterogeneous devices, acquire raw communication data streams at the device layer in real time, perform preliminary parsing on the raw communication data streams, and obtain parsed information. The protocol feature learning module is used to process the parsed information using statistical analysis algorithms to obtain communication protocol features, which include: message structure, field distribution patterns, and normal data patterns. The dynamic adaptation module is used to generate data parsing rules based on the communication protocol definition information, message structure and field distribution rules, and to normalize the original communication data stream according to the data parsing rules to obtain standardized data. The anomaly detection module is used to compare and detect the communication protocol definition information, standardized data, and normal data patterns to obtain anomaly detection results. The intelligent diagnosis module is used to perform anomaly inference based on the anomaly detection results and a preset knowledge base, obtain an anomaly diagnosis report, and complete the intelligent diagnosis.

8. The metadata-driven intelligent diagnostic system for industrial energy storage according to claim 7, characterized in that, Also includes: Metadata storage module, optimization feedback module, user interaction module, and data transmission module; The metadata storage module, which communicates with the dynamic adaptation module, the anomaly detection module, the protocol feature learning module, the optimization feedback module, and the user interaction module, is used to store the communication protocol definition information of the heterogeneous devices in a structured form. The optimization feedback module is used to generate optimization suggestions and adjust the data parsing rules and communication protocol definition information based on the anomaly diagnosis report; The user interaction module is communicatively connected to the metadata storage module, the anomaly detection module, and the intelligent diagnosis module. It is used to provide a human-machine interaction interface and receive metadata configuration instructions, device status monitoring instructions, alarm information viewing instructions, and optimization suggestion confirmation instructions. The data transmission module communicates with the data acquisition module, metadata storage module, protocol feature learning module, dynamic adaptation module, anomaly detection module, intelligent diagnosis module, optimization feedback module, and user interaction module, and is used to asynchronously transmit data and instructions between the modules using a distributed message middleware.

9. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the metadata-driven intelligent diagnostic method for industrial energy storage as described in any one of claims 1 to 6.

10. A computer program product, the computer program product comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the steps of the metadata-driven intelligent diagnostic method for industrial energy storage as described in any one of claims 1 to 6.

Images