Equipment health management system and method based on artificial intelligence and computer equipment

Abstract

The invention discloses an equipment health management system and method based on artificial intelligence and computer equipment, and the system comprises an equipment data collection module which is used for obtaining equipment environment data transmitted by a sensor disposed in an influence area corresponding to each piece of electrical equipment, and obtaining equipment state data corresponding to each piece of electrical equipment; the data preprocessing module is used for carrying out standardized preprocessing on the equipment environment data and the equipment state data; the fault diagnosis module is used for performing fault diagnosis on the electrical equipment based on the preprocessed equipment environment data and equipment state data so as to identify faulty electrical equipment and generate fault information of the faulty electrical equipment; the research and judgment processing module is used for constructing a fault analysis prompt word according to the fault information, inputting the fault analysis prompt word into a research and judgment large model, and generating fault processing information of the fault electrical equipment through the research and judgment large model; and the fault handling module is used for carrying out fault handling on the fault electrical equipment based on the fault handling information.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to an artificial intelligence-based equipment health management system, method and computer equipment. Background Art

[0002] With the rapid development of intelligent manufacturing and Industry 4.0, smart factories have become the mainstream model of modern industrial production. In smart factories, a large number of electrical equipment are widely used in various links of the production line, from basic power supply systems to complex automation equipment. The stable operation of electrical equipment is a key factor in ensuring production efficiency and product quality. However, due to the diversity and complexity of electrical equipment, as well as long-term, high-load operating environments, the failure rate of equipment is often high, which not only affects the normal operation of the production line, but may also cause serious safety accidents.

[0003] Traditionally, the health management of electrical equipment mainly relies on manual inspections and regular maintenance. This approach has many shortcomings, such as long inspection cycles, high maintenance costs, and untimely fault detection. Especially in smart factories, the number of electrical equipment is large and widely distributed, and manual inspections are difficult to achieve comprehensive coverage and real-time response. In addition, electrical equipment failures are often sudden and uncertain, and traditional regular maintenance methods are difficult to effectively prevent failures.

[0004] In recent years, although some sensor-based monitoring systems have been applied to the health management of electrical equipment, most of these systems can only achieve simple data collection and display, and lack intelligent fault diagnosis and handling capabilities. Therefore, when equipment fails, manual fault analysis and processing are still required, which undoubtedly increases the complexity and cost of maintenance work. Summary of the invention

[0005] In view of this, the embodiments of the present application provide an equipment health management system, method and computer equipment based on artificial intelligence, which provide effective solutions to the problems of inaccurate diagnosis, untimely maintenance, excessive manual intervention, etc. in traditional equipment health management. By introducing artificial intelligence technology and big data analysis, the system can realize intelligent and precise management of electrical equipment and improve the overall operational efficiency of smart factories.

[0006] According to one aspect of the present application, an artificial intelligence-based equipment health management system is provided for performing health management on electrical equipment in a smart factory, including: The equipment data acquisition module is used to obtain equipment environment data transmitted by sensors deployed in the influence area corresponding to each electrical equipment, and to obtain equipment status data corresponding to each electrical equipment; A data preprocessing module, used for performing standardized preprocessing on the device environment data and the device status data; A fault diagnosis module, used to perform fault diagnosis on the electrical equipment based on the pre-processed equipment environment data and equipment status data, so as to identify the faulty electrical equipment and generate fault information of the faulty electrical equipment; A judgment and disposal module is used to construct a fault analysis prompt word according to the fault information, input the fault analysis prompt word into the judgment big model, and generate fault disposal information for the faulty electrical equipment through the judgment big model; A fault handling module is used to handle the fault of the faulty electrical equipment based on the fault handling information.

[0007] Optionally, the device status data includes at least device current, device voltage, device power factor, and device temperature; the device environment data includes at least ambient temperature, ambient humidity, and near-end crosstalk; The fault diagnosis module is also used to input the preprocessed equipment status data and equipment environment data into the fault diagnosis model, and the fault diagnosis model includes a vector conversion unit, a data fusion unit, a data enhancement unit, and a fault diagnosis unit; the vector conversion unit is used to perform vector conversion on each item of input equipment status data and each item of input equipment environment data, the data fusion unit is used to perform data fusion on the data vector, and the data enhancement unit is used to perform data enhancement processing on the fused data vector to obtain a data feature vector, and the fault diagnosis unit is used to perform fault diagnosis on the electrical equipment based on the data feature vector.

[0008] Optionally, it also includes: A fault diagnosis model training module is used to obtain historical operation data of various operation states of each electrical device in a smart factory, wherein the historical operation data includes historical device state data, historical device environment data, a fault label indicating whether it is a fault state, and a type label indicating the device type of the electrical device. The various operation states include a fault operation state and various normal operation states. The various normal operation states include a stable operation state, a stress test operation state, and a pre-fault operation state. The stable operation state means that the electrical device maintains normal operation during the collection period corresponding to the historical normal operation data, and the preset periods before and after the collection period. The stress test operation state means that the electrical device maintains stable operation during the stress test process. The pre-fault operation state means that the electrical device maintains normal operation during the collection period corresponding to the historical normal operation data, and a fault occurs during the preset period after the collection period. Constructing an initial fault diagnosis model, the initial fault diagnosis model comprising an initial vector conversion unit, an initial data fusion unit, an initial data enhancement unit, a label discriminator, and initial fault diagnosis units having the same number as the sample groups; The initial fault diagnosis model is trained based on the training samples to obtain the fault diagnosis model; wherein, for any training sample, the training process includes: sequentially processing the historical equipment status data and the historical equipment environment data in the training sample by the initial vector conversion unit, the initial data fusion unit, and the initial data enhancement unit, discriminating the type labels in the training sample by the label discriminator to determine the initial fault diagnosis unit used in this round, inputting the data processed by the initial data enhancement unit into the initial fault diagnosis unit used in this round for fault label prediction, and optimizing the parameters of the initial vector conversion unit, the initial data fusion unit, the initial data enhancement unit, and the initial fault diagnosis unit used in this round in the initial fault diagnosis model based on the predicted fault label and the actual fault label corresponding to the training sample.

[0009] Optionally, the device data acquisition module is further used to obtain the device identification of each electrical device; The data preprocessing module is further used to determine the device model based on the device identifier, and determine the device type label based on the device type corresponding to the device model; The fault diagnosis module is also used to input the device type label, pre-processed device status data, and device environment data into the fault diagnosis model, perform vector conversion on each item of input device status data and each item of input device environment data respectively through the vector conversion unit, perform data fusion on the data vector through the data fusion unit, perform data enhancement processing on the fused data vector through the data enhancement unit to obtain a data feature vector, discriminate the input device type label through the label discriminator to determine the target fault diagnosis unit, and input the data feature vector into the target fault diagnosis unit for fault diagnosis.

[0010] Optionally, it also includes: A large model training module is used to obtain electrical equipment operation and maintenance records, wherein the electrical equipment operation and maintenance records include the model of the operated and maintained electrical equipment, the equipment environment data and equipment status data of the operated and maintained electrical equipment, the fault detection information corresponding to the operated and maintained electrical equipment, the fault detection results, and the operation and maintenance records. The fault detection information indicates the information detected by the operation and maintenance personnel when troubleshooting electrical equipment faults, the fault detection results indicate the detection results of the fault detection information, and the operation and maintenance records indicate the disposal methods taken by the operation and maintenance personnel for the electrical equipment after obtaining the detection results; based on the model of the operated and maintained electrical equipment, the equipment environment data and equipment status data of the operated and maintained electrical equipment, and the fault detection information corresponding to the operated and maintained electrical equipment in the electrical equipment operation and maintenance records, the initial analysis and judgment large model is trained in one stage to train the ability of the initial analysis and judgment large model to predict fault detection information; and based on the model of the operated and maintained electrical equipment, the equipment environment data and equipment status data of the operated and maintained electrical equipment, the fault detection information corresponding to the operated and maintained electrical equipment, the fault detection results, and the operation and maintenance records, the initial analysis and judgment large model after the one-stage training is trained in two stages to train the ability of the initial analysis and judgment large model to generate operation and maintenance information, thereby obtaining the analysis and judgment large model.

[0011] Optionally, the fault information includes the faulty electrical equipment model, equipment environment data, and equipment status data of the faulty electrical equipment; The analysis and judgment disposal module is further used to construct a first fault analysis prompt word according to the fault information, input the first fault analysis prompt word into the analysis and judgment large model, and generate target fault detection information for the faulty electrical equipment through the analysis and judgment large model, wherein the target fault detection information includes information required for detection when troubleshooting the faulty electrical equipment; The fault handling module is further used to perform an interface call based on the target fault detection information to obtain a fault detection result corresponding to the faulty electrical equipment; The analysis and disposal module is further used to construct a second fault analysis prompt word according to the fault information and the fault detection result, input the second fault analysis prompt word into the analysis and judgment large model, and generate operation and maintenance information of the faulty electrical equipment through the analysis and judgment large model; The fault handling module is further used to send the operation and maintenance information to an administrator terminal to prompt the administrator to handle the fault of the faulty electrical equipment based on the operation and maintenance information.

[0012] Optionally, the fault handling module is further used to receive feedback information from the administrator terminal on the operation and maintenance information, and perform fault handling on the faulty electrical equipment based on the feedback information.

[0013] Optionally, the second fault analysis prompt word includes content guiding the reasoning process of the large model output to generate the operation and maintenance information; The analysis and judgment disposal module is further used to input the second fault analysis prompt word into the analysis and judgment large model, and generate operation and maintenance information and reasoning process information of the faulty electrical equipment through the analysis and judgment large model; The fault handling module is further used to send the operation and maintenance information and the reasoning process information to the administrator terminal to prompt the administrator to confirm or modify the operation and maintenance information based on the reasoning information, and to provide feedback.

[0014] According to another aspect of the present application, there is provided an artificial intelligence-based equipment health management method, characterized in that it is applied to the above-mentioned artificial intelligence-based equipment health management system, including: The equipment data acquisition module acquires equipment environment data transmitted by sensors deployed in the influence area corresponding to each electrical equipment, and acquires equipment status data corresponding to each electrical equipment; The data preprocessing module performs standardized preprocessing on the device environment data and the device status data; The fault diagnosis module performs fault diagnosis on the electrical equipment based on the preprocessed equipment environment data and equipment status data to identify the faulty electrical equipment and generate fault information of the faulty electrical equipment; The analysis and disposal module constructs a fault analysis prompt word according to the fault information, inputs the fault analysis prompt word into the analysis and judgment large model, and generates fault disposal information for the faulty electrical equipment through the analysis and judgment large model; The fault handling module performs fault handling on the faulty electrical equipment based on the fault handling information.

[0015] Optionally, the device status data includes at least device current, device voltage, device power factor, and device temperature; the device environment data includes at least ambient temperature, ambient humidity, and near-end crosstalk; The fault diagnosis module performs fault diagnosis on the electrical equipment based on the preprocessed equipment environment data and equipment status data to identify the faulty electrical equipment and generate fault information of the faulty electrical equipment, including: The fault diagnosis module inputs the preprocessed equipment status data and equipment environment data into the fault diagnosis model, and the fault diagnosis model includes a vector conversion unit, a data fusion unit, a data enhancement unit, and a fault diagnosis unit; the vector conversion unit performs vector conversion on each item of input equipment status data and each item of input equipment environment data, the data fusion unit performs data fusion on the data vector, and the data enhancement unit performs data enhancement processing on the fused data vector to obtain a data feature vector, and the fault diagnosis unit performs fault diagnosis on the electrical equipment based on the data feature vector.

[0016] Optionally, it also includes: The fault diagnosis model training module obtains historical operation data of various operation states of each electrical device in the smart factory, wherein the historical operation data includes historical device state data, historical device environment data, a fault label indicating whether it is a fault state, and a type label indicating the device type of the electrical device. The various operation states include a fault operation state and various normal operation states. The various normal operation states include a stable operation state, a stress test operation state, and a pre-fault operation state. The stable operation state means that the electrical device maintains normal operation during the collection period corresponding to the historical normal operation data, and the preset periods before and after the collection period. The stress test operation state means that the electrical device maintains stable operation during the stress test process. The pre-fault operation state means that the electrical device maintains normal operation during the collection period corresponding to the historical normal operation data, and a fault occurs during the preset period after the collection period. Constructing an initial fault diagnosis model, the initial fault diagnosis model comprising an initial vector conversion unit, an initial data fusion unit, an initial data enhancement unit, a label discriminator, and initial fault diagnosis units having the same number as the sample groups; The initial fault diagnosis model is trained based on the training samples to obtain the fault diagnosis model; wherein, for any training sample, the training process includes: sequentially processing the historical equipment status data and the historical equipment environment data in the training sample by the initial vector conversion unit, the initial data fusion unit, and the initial data enhancement unit, discriminating the type labels in the training sample by the label discriminator to determine the initial fault diagnosis unit used in this round, inputting the data processed by the initial data enhancement unit into the initial fault diagnosis unit used in this round for fault label prediction, and optimizing the parameters of the initial vector conversion unit, the initial data fusion unit, the initial data enhancement unit, and the initial fault diagnosis unit used in this round in the initial fault diagnosis model based on the predicted fault label and the actual fault label corresponding to the training sample.

[0017] Optionally, it also includes: The device data acquisition module obtains the device identification of each electrical device; The data preprocessing module determines the device model based on the device identifier, and determines the device type label based on the device type corresponding to the device model; The fault diagnosis module inputs the pre-processed device status data and device environment data into the fault diagnosis model, including: The fault diagnosis module inputs the device type label, pre-processed device status data, and device environment data into the fault diagnosis model, performs vector conversion on each item of input device status data and each item of input device environment data respectively through the vector conversion unit, performs data fusion on the data vector through the data fusion unit, performs data enhancement processing on the fused data vector through the data enhancement unit to obtain a data feature vector, discriminates the input device type label through the label discriminator to determine a target fault diagnosis unit, and inputs the data feature vector into the target fault diagnosis unit for fault diagnosis.

[0018] Optionally, it also includes: The large model training module obtains the operation and maintenance records of the electrical equipment, which include the model of the operated and maintained electrical equipment, the equipment environment data and equipment status data of the operated and maintained electrical equipment, the fault detection information corresponding to the operated and maintained electrical equipment, the fault detection results, and the operation and maintenance records. The fault detection information indicates the information detected by the operation and maintenance personnel when troubleshooting the electrical equipment fault, the fault detection result indicates the detection result of the fault detection information, and the operation and maintenance record indicates the disposal method taken by the operation and maintenance personnel for the electrical equipment after obtaining the detection result; based on the model of the operated and maintained electrical equipment, the equipment environment data and equipment status data of the operated and maintained electrical equipment, and the fault detection information corresponding to the operated and maintained electrical equipment in the operation and maintenance records of the electrical equipment, the initial analysis and judgment large model is trained in one stage to train the ability of the initial analysis and judgment large model to predict fault detection information; and based on the model of the operated and maintained electrical equipment, the equipment environment data and equipment status data of the operated and maintained electrical equipment, the fault detection information corresponding to the operated and maintained electrical equipment, the fault detection results, and the operation and maintenance records, the initial analysis and judgment large model after the one-stage training is trained in two stages to train the ability of the initial analysis and judgment large model to generate operation and maintenance information, thereby obtaining the analysis and judgment large model.

[0019] Optionally, the fault information includes the faulty electrical equipment model, equipment environment data, and equipment status data of the faulty electrical equipment; The analysis and disposal module constructs a fault analysis prompt word according to the fault information, inputs the fault analysis prompt word into the analysis and judgment large model, and generates fault disposal information for the faulty electrical equipment through the analysis and judgment large model, including: The analysis and disposal module constructs a first fault analysis prompt word according to the fault information, inputs the first fault analysis prompt word into the analysis and judgment large model, and generates target fault detection information for the faulty electrical equipment through the analysis and judgment large model, wherein the target fault detection information includes information required for detection when troubleshooting the faulty electrical equipment; The method further comprises: The fault handling module performs an interface call based on the target fault detection information to obtain a fault detection result corresponding to the faulty electrical equipment; The analysis and disposal module constructs a second fault analysis prompt word according to the fault information and the fault detection result, inputs the second fault analysis prompt word into the analysis and judgment large model, and generates operation and maintenance information of the faulty electrical equipment through the analysis and judgment large model; The fault handling module performs fault handling on the faulty electrical equipment based on the fault handling information, including: The fault handling module sends the operation and maintenance information to the administrator terminal to prompt the administrator to handle the fault of the faulty electrical equipment based on the operation and maintenance information.

[0020] Optionally, it also includes: the fault handling module receives feedback information from the administrator terminal on the operation and maintenance information, and performs fault handling on the faulty electrical equipment based on the feedback information.

[0021] Optionally, the second fault analysis prompt includes content that guides the reasoning process of the large model output to generate the operation and maintenance information; and further includes: The analysis and disposal module inputs the second fault analysis prompt word into the analysis and judgment large model, and generates operation and maintenance information and reasoning process information of the faulty electrical equipment through the analysis and judgment large model; The fault handling module sends the operation and maintenance information and the reasoning process information to the administrator terminal to prompt the administrator to confirm or modify the operation and maintenance information based on the reasoning information, and to provide feedback.

[0022] According to another aspect of the present application, a storage medium is provided, on which a computer program is stored, and when the program is executed by a processor, the above-mentioned artificial intelligence-based device health management method is implemented.

[0023] According to another aspect of the present application, a computer device is provided, including a storage medium, a processor, and a computer program stored on the storage medium and executable on the processor, wherein the processor implements the above-mentioned artificial intelligence-based device health management method when executing the program.

[0024] Through the above technical solutions, the embodiments of the present application provide an artificial intelligence-based equipment health management system, method and computer equipment. By introducing fault diagnosis models and large-scale research and judgment models, the system can more accurately identify faults of electrical equipment and reduce false alarms and missed alarms. Based on real-time data and fault prediction, the system can formulate maintenance plans in advance to avoid equipment shutdown due to sudden failures, reduce maintenance costs, and improve equipment reliability and availability. The system can automatically generate fault handling information and guide maintenance personnel to handle faults, reducing the time and errors of manual judgment and improving maintenance efficiency. The embodiments of the present application provide effective solutions to the problems of inaccurate diagnosis, untimely maintenance, and excessive manual intervention in traditional equipment health management. By introducing artificial intelligence technology and big data analysis, the system can realize intelligent and precise management of electrical equipment and improve the overall operational efficiency of smart factories.

[0025] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings: Figure 1 A schematic diagram of the structure of an artificial intelligence-based equipment health management system provided in an embodiment of the present application is shown; Figure 2 A flow chart of an artificial intelligence-based device health management method provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0027] The present application will be described in detail below with reference to the accompanying drawings and in combination with embodiments. It should be noted that the embodiments and features in the embodiments of the present application can be combined with each other without conflict.

[0028] In this embodiment, an artificial intelligence-based equipment health management system is provided to manage the health of electrical equipment in a smart factory. Figure 1 As shown, including: The equipment data acquisition module is used to obtain equipment environment data transmitted by sensors deployed in the influence area corresponding to each electrical equipment, and to obtain equipment status data corresponding to each electrical equipment; A data preprocessing module, used for performing standardized preprocessing on the device environment data and the device status data; A fault diagnosis module, used to perform fault diagnosis on the electrical equipment based on the pre-processed equipment environment data and equipment status data, so as to identify the faulty electrical equipment and generate fault information of the faulty electrical equipment; A judgment and disposal module is used to construct a fault analysis prompt word according to the fault information, input the fault analysis prompt word into the judgment big model, and generate fault disposal information for the faulty electrical equipment through the judgment big model; A fault handling module is used to handle the fault of the faulty electrical equipment based on the fault handling information.

[0029] The embodiment of the present application provides an artificial intelligence-based equipment health management system, which is designed for electrical equipment in smart factories. The core goal of the system is to achieve comprehensive health management of electrical equipment, thereby improving the operating efficiency of the equipment, reducing downtime due to failures, and optimizing maintenance strategies. The system mainly consists of the following modules: Equipment data acquisition module: responsible for collecting environmental data (such as temperature, humidity, vibration, etc.) transmitted by sensors deployed around electrical equipment, as well as the status data of the electrical equipment itself (such as current, voltage, power factor, etc.), which are the basis for subsequent analysis and diagnosis. Data preprocessing module: standardize and preprocess the collected equipment environment data and equipment status data to ensure the consistency and comparability of the data, including data cleaning, denoising, normalization and other operations. Fault diagnosis module: based on the preprocessed data, use fault diagnosis models (such as machine learning, deep learning, etc.) to diagnose faults in electrical equipment, identify faulty electrical equipment, and generate fault information. Research and judgment disposal module: construct fault analysis prompt words based on the fault information generated by the fault diagnosis module, and input these prompt words into the research and judgment large model, which generates fault disposal information for faulty electrical equipment based on historical data and expert knowledge. The disposal information may include maintenance suggestions, replacement of parts, adjustment of operating parameters, etc. Fault disposal module: According to the fault disposal information generated by the analysis and disposal module, actual fault disposal is performed on the faulty electrical equipment, such as manual intervention or automated maintenance system.

[0030] By applying the technical solution of this embodiment, by introducing fault diagnosis models and large-scale research and judgment models, the system can more accurately identify faults in electrical equipment and reduce false alarms and missed alarms. Based on real-time data and fault prediction, the system can formulate maintenance plans in advance to avoid equipment shutdowns due to sudden failures, reduce maintenance costs, and improve equipment reliability and availability. The system can automatically generate fault handling information and guide maintenance personnel to handle faults, reducing the time and errors of manual judgment and improving maintenance efficiency. The embodiment of the present application provides an effective solution to the problems of inaccurate diagnosis, untimely maintenance, and excessive manual intervention in the health management of traditional equipment. By introducing artificial intelligence technology and big data analysis, the system can realize intelligent and precise management of electrical equipment and improve the overall operational efficiency of smart factories.

[0031] In an embodiment of the present application, optionally, the device status data includes at least device current, device voltage, device power factor, and device temperature; the device environment data includes at least ambient temperature, ambient humidity, and near-end crosstalk; the fault diagnosis module is also used to input the preprocessed device status data and device environment data into a fault diagnosis model, and the fault diagnosis model includes a vector conversion unit, a data fusion unit, a data enhancement unit, and a fault diagnosis unit; the vector conversion unit is used to respectively perform vector conversion on each item of input device status data and each item of input device environment data, the data fusion unit is used to perform data fusion on the data vector, and the data enhancement unit is used to perform data enhancement processing on the fused data vector to obtain a data feature vector, and the fault diagnosis unit is used to perform fault diagnosis on the electrical equipment based on the data feature vector.

[0032] In the above embodiment, the device status data at least includes device current, device voltage, device power factor, and device temperature. These data reflect the basic state of the electrical equipment during operation and are important bases for diagnosing whether the equipment has a fault. The device environment data at least includes ambient temperature, ambient humidity, and near-end crosstalk. These data describe the external environment in which the electrical equipment is located and have an important impact on the operating state and life of the equipment. The fault diagnosis model includes a vector conversion unit, a data fusion unit, a data enhancement unit, and a fault diagnosis unit. Among them, the vector conversion unit converts the input device status data and device environment data into vector form for subsequent data processing and analysis. The data fusion unit performs data fusion on the converted data vectors, and integrates the device status data and the device environment data into a unified data representation to capture the potential association between the two. The data enhancement unit performs data enhancement processing on the fused data vectors, and improves the robustness and accuracy of the fault diagnosis model by increasing the diversity and richness of the data. The fault diagnosis unit uses machine learning or deep learning algorithms to perform fault diagnosis on the electrical equipment based on the enhanced data feature vectors.

[0033] In an embodiment of the present application, optionally, it also includes: a fault diagnosis model training module, which is used to obtain historical operation data of various operation states of each electrical equipment in the smart factory, the historical operation data includes historical equipment status data, historical equipment environment data, a fault label indicating whether it is a fault state, and a type label indicating the equipment type of the electrical equipment. The various operation states include a fault operation state and a plurality of normal operation states. The various normal operation states include a stable operation state, a stress test operation state, and a pre-fault operation state. The stable operation state means that the electrical equipment maintains normal operation during the collection period corresponding to the historical normal operation data and the preset time periods before and after the collection period. The stress test operation state means that the electrical equipment maintains stable operation during the stress test process. The pre-fault operation state means that the electrical equipment maintains normal operation during the collection period corresponding to the historical normal operation data, and a fault occurs during the preset time period after the collection period. An initial fault diagnosis model is constructed, and the initial fault diagnosis model includes an initial vector conversion unit. element, an initial data fusion unit, an initial data enhancement unit, a label discriminator, and initial fault diagnosis units the same in number as the sample groups; the initial fault diagnosis model is trained based on the training samples to obtain the fault diagnosis model; wherein, for any training sample, the training process comprises: sequentially processing the historical equipment status data and the historical equipment environment data in the training sample by the initial vector conversion unit, the initial data fusion unit, and the initial data enhancement unit, discriminating the type labels in the training sample by the label discriminator to determine the initial fault diagnosis unit used in this round, inputting the data processed by the initial data enhancement unit into the initial fault diagnosis unit used in this round for fault label prediction, and optimizing the parameters of the initial vector conversion unit, the initial data fusion unit, the initial data enhancement unit, and the initial fault diagnosis unit used in this round in the initial fault diagnosis model based on the predicted fault label and the actual fault label corresponding to the training sample.

[0034] In the above embodiment, the fault diagnosis model training module is used to build and optimize the fault diagnosis model. This module trains the initial fault diagnosis model by using the historical operation data of various operating states of each electrical device in the smart factory to improve the fault diagnosis ability of the model. Specifically, the historical operation data of various operating states of each electrical device in the smart factory are obtained, including historical equipment state data, historical equipment environment data, fault labels and type labels. These data cover various operating states of electrical equipment, such as fault operating state, stable operating state, stress test operating state and pre-fault operating state. Model construction: Construct an initial fault diagnosis model, which includes an initial vector conversion unit, an initial data fusion unit, an initial data enhancement unit, a label discriminator and an initial fault diagnosis unit with the same number as the sample group. This design allows the model to perform customized fault diagnosis for different types of electrical equipment. For any training sample, firstly, the historical equipment status data and the historical equipment environment data are processed in sequence by the initial vector conversion unit, the initial data fusion unit and the initial data enhancement unit; the type labels in the training samples are discriminated by the label discriminator to determine the initial fault diagnosis unit used in this round of training; the enhanced data is input into the determined initial fault diagnosis unit for fault label prediction; based on the predicted fault label and the real fault label corresponding to the training sample, the parameters of each unit in the initial fault diagnosis model are optimized to improve the fault diagnosis accuracy of the model. The embodiment of the present application uses the historical operation data of multiple operating states for training, and the model can learn the characteristics of electrical equipment under different operating states, thereby improving the generalization ability and adaptability of the model. By introducing the label discriminator and the initial fault diagnosis unit with the same number as the sample group, the model can perform customized fault diagnosis for different types of electrical equipment, improving the accuracy and pertinence of the diagnosis. Through accurate fault diagnosis and prediction, the system can detect potential fault hazards in advance and take corresponding maintenance measures, thereby reducing the downtime and maintenance costs caused by equipment failure.

[0035] In the embodiment of the present application, optionally, the device data acquisition module is further used to obtain the device identification of each electrical device; The data preprocessing module is further used to determine the device model based on the device identifier, and determine the device type label based on the device type corresponding to the device model; The fault diagnosis module is also used to input the device type label, pre-processed device status data, and device environment data into the fault diagnosis model, perform vector conversion on each item of input device status data and each item of input device environment data respectively through the vector conversion unit, perform data fusion on the data vector through the data fusion unit, perform data enhancement processing on the fused data vector through the data enhancement unit to obtain a data feature vector, discriminate the input device type label through the label discriminator to determine the target fault diagnosis unit, and input the data feature vector into the target fault diagnosis unit for fault diagnosis.

[0036] In the above embodiment, in addition to collecting device status data and device environment data, the device data acquisition module also adds the function of obtaining the device identification of each electrical device. The device identification is the unique identification code of the electrical device, which is used to distinguish different devices. The data preprocessing module can also determine the device model based on the device identification, and further determine the device type label based on the device type corresponding to the device model. The fault diagnosis module can also use the device type label as one of the input functions. The label discriminator in the fault diagnosis model will determine the target fault diagnosis unit that is most suitable for fault diagnosis based on the input device type label. In this way, each device type can have a customized fault diagnosis unit for its characteristics, thereby improving the accuracy and efficiency of fault diagnosis. By introducing the device identification and device type label, the embodiment of the present application can accurately identify the model and type of each electrical device, thereby selecting the fault diagnosis unit that best suits its characteristics for diagnosis, thereby improving the accuracy and pertinence of fault diagnosis. Since each device type has a corresponding customized fault diagnosis unit, the system can identify the fault type, location and severity more quickly, thereby shortening the time of fault diagnosis and improving efficiency. The system can handle electrical equipment of different models and types, making it more flexible and adaptable. Whether it is old equipment or new equipment, the system can provide accurate fault diagnosis services.

[0037] In an embodiment of the present application, optionally, it also includes: a large model training module for obtaining electrical equipment operation and maintenance records, wherein the electrical equipment operation and maintenance records include the model of the operated and maintained electrical equipment, the equipment environment data and equipment status data of the operated and maintained electrical equipment, the fault detection information corresponding to the operated and maintained electrical equipment, the fault detection result, and the operation and maintenance record, wherein the fault detection information indicates the information detected by the operation and maintenance personnel when troubleshooting electrical equipment faults, the fault detection result indicates the detection result of the fault detection information, and the operation and maintenance record indicates the disposal method of the electrical equipment taken by the operation and maintenance personnel after obtaining the detection result; based on the operation and maintenance records in the electrical equipment operation and maintenance records The initial analysis and judgment big model is trained in one stage based on the model of the operation and maintenance electrical equipment, the equipment environment data and equipment status data of the operation and maintenance electrical equipment, and the fault detection information corresponding to the operation and maintenance electrical equipment, so as to train the ability of the initial analysis and judgment big model to predict the fault detection information; and based on the model of the operation and maintenance electrical equipment, the equipment environment data and equipment status data of the operation and maintenance electrical equipment, the fault detection information corresponding to the operation and maintenance electrical equipment, the fault detection results, and the operation and maintenance records in the electrical equipment operation and maintenance records, the initial analysis and judgment big model after the one-stage training is trained in a second stage so as to train the ability of the initial analysis and judgment big model to generate operation and maintenance information, so as to obtain the analysis and judgment big model.

[0038] In the above embodiment, the large model training module is intended to train a large model for analysis and judgment through the operation and maintenance records of electrical equipment to improve the intelligent level of fault detection and operation and maintenance of electrical equipment. The training process of the large model training module includes: first, obtaining the operation and maintenance records of electrical equipment from the electrical equipment operation and maintenance system. These records contain the model of the electrical equipment under operation and maintenance, the equipment environment data and the equipment status data, the information detected by the operation and maintenance personnel when troubleshooting the electrical equipment fault (fault detection information), the detection results of the fault detection information (fault detection results) and the disposal methods taken by the operation and maintenance personnel after obtaining the detection results (operation and maintenance records). Phase 1 training: Based on the model of the electrical equipment under operation and maintenance, the equipment environment data and the equipment status data, and the fault detection information in the operation and maintenance records, the initial analysis and judgment large model is trained in the first phase. The main goal of this phase is to train the model's ability to predict fault detection information, that is, to predict the information that the operation and maintenance personnel may detect when troubleshooting the fault based on the model, environment data and status data of the equipment. Phase 2 training: On the basis of the first phase training, the fault detection results and operation and maintenance records are further introduced to train the initial analysis and judgment large model in the second phase. The main goal of this stage is to train the model to generate operation and maintenance information, that is, to predict the possible disposal methods that the operation and maintenance personnel may take based on the model, environmental data and status data of the equipment, fault detection information and fault detection results, and generate corresponding operation and maintenance suggestions. In the embodiment of the present application, through the first stage of training, the research and judgment large model can predict the fault detection information based on the model, environmental data and status data of the equipment, thereby assisting the operation and maintenance personnel to locate the fault point more accurately and improve the intelligent level of fault detection. Through the second stage of training, the research and judgment large model can predict the possible disposal methods that the operation and maintenance personnel may take based on the model, environmental data and status data of the equipment, fault detection information and fault detection results, and generate corresponding operation and maintenance suggestions. This helps the operation and maintenance personnel to make more reasonable operation and maintenance decisions and improve the efficiency and quality of operation and maintenance. The research and judgment large model can learn the fault characteristics and operation and maintenance rules of electrical equipment of different models and types through training, thereby enhancing the adaptability and flexibility of the system. Whether it is old equipment or new equipment, the system can provide intelligent fault detection and operation and maintenance suggestions. Through intelligent fault detection and operation and maintenance suggestions, the system can reduce the workload of operation and maintenance personnel and reduce operation and maintenance costs. At the same time, since fault detection and operation and maintenance decisions are more accurate and efficient, it also helps to reduce downtime and maintenance costs caused by equipment failure.

[0039] In the embodiment of the present application, optionally, the fault information includes the faulty electrical equipment model, equipment environment data, and equipment status data of the faulty electrical equipment; The analysis and judgment disposal module is further used to construct a first fault analysis prompt word according to the fault information, input the first fault analysis prompt word into the analysis and judgment large model, and generate target fault detection information for the faulty electrical equipment through the analysis and judgment large model, wherein the target fault detection information includes information required for detection when troubleshooting the faulty electrical equipment; The fault handling module is further used to perform an interface call based on the target fault detection information to obtain a fault detection result corresponding to the faulty electrical equipment; The analysis and disposal module is further used to construct a second fault analysis prompt word according to the fault information and the fault detection result, input the second fault analysis prompt word into the analysis and judgment large model, and generate operation and maintenance information of the faulty electrical equipment through the analysis and judgment large model; The fault handling module is further used to send the operation and maintenance information to an administrator terminal to prompt the administrator to handle the fault of the faulty electrical equipment based on the operation and maintenance information.

[0040] In the above embodiment, when fault information (including the model of the faulty electrical equipment, equipment environment data, and equipment status data) is received, the analysis and disposal module will construct a first fault analysis prompt word based on this information. This prompt word is a general description of the current state of the faulty electrical equipment, which is used to guide the analysis and judgment model to perform detection information analysis. The first fault analysis prompt word is input into the analysis and judgment model, and the model will generate the information required for troubleshooting the faulty electrical equipment based on its internally learned knowledge and experience, that is, the target fault detection information. This information is the content that the operation and maintenance personnel need to focus on and detect when troubleshooting. After obtaining the fault detection result, the analysis and disposal module will construct a second fault analysis prompt word based on the fault information and the fault detection result. This prompt word is a detailed description of the fault condition of the faulty electrical equipment, including the fault type, location, severity, etc., which is used to further guide the analysis and judgment model to perform operation and maintenance analysis. The second fault analysis prompt word is input into the analysis and judgment model, and the model will generate operation and maintenance information for the faulty electrical equipment based on its internally learned operation and maintenance knowledge and experience. This information includes operation and maintenance suggestions, disposal methods, required tools or spare parts, etc., which are used to guide administrators to handle faults. Based on the target fault detection information generated by the analysis and disposal module, the fault disposal module will call the corresponding interface or tool to perform fault detection on the faulty electrical equipment and obtain the detection results. After obtaining the operation and maintenance information generated by the analysis and disposal module, the fault disposal module will send this information to the administrator terminal. The administrator can perform fault disposal on the faulty electrical equipment based on this information, including repair, replacement of parts, adjustment of parameters, etc.

[0041] The embodiment of the present application can generate more accurate target fault detection information by constructing a first fault analysis prompt word and inputting it into the research and judgment large model through the research and judgment disposal module, and guide the operation and maintenance personnel to perform targeted troubleshooting, thereby improving the accuracy and efficiency of fault detection. The research and judgment disposal module constructs a second fault analysis prompt word based on the fault information and fault detection results, and inputs it into the research and judgment large model to generate operation and maintenance information. This information can provide administrators with more comprehensive and accurate operation and maintenance suggestions, helping administrators make more reasonable operation and maintenance decisions. Through intelligent fault detection and operation and maintenance suggestions, the system can reduce the workload of operation and maintenance personnel and reduce operation and maintenance costs. At the same time, since fault detection and operation and maintenance decisions are more accurate and efficient, it also helps to reduce downtime and maintenance costs caused by equipment failures.

[0042] In the embodiment of the present application, optionally, the fault handling module is also used to receive feedback information from the administrator terminal on the operation and maintenance information, and perform fault handling on the faulty electrical equipment based on the feedback information.

[0043] In the above embodiment, the fault handling module can receive feedback information from the administrator terminal on the operation and maintenance information. These feedback information may include the administrator's confirmation of the operation and maintenance information, modification suggestions, additional disposal requirements, etc. After receiving the feedback information, the fault handling module will handle the fault of the faulty electrical equipment according to this information. This may include executing the operation and maintenance suggestions according to the administrator's confirmation, adjusting the operation and maintenance plan according to the modification suggestions, adding or modifying the disposal steps according to the additional disposal requirements, etc. In this way, the fault handling module can respond to various fault conditions more flexibly and ensure that the faulty electrical equipment is handled in a timely and effective manner. In the embodiment of the present application, the system can receive and process the administrator's feedback information, thereby enhancing the interactivity of the system. The administrator can confirm, modify or put forward additional disposal requirements according to the system's operation and maintenance suggestions, and the system can make corresponding adjustments based on these feedbacks. By receiving and processing feedback information, the fault handling module can respond to various fault conditions more flexibly. Whether the administrator needs to confirm the operation and maintenance suggestions, modify the disposal plan or put forward additional disposal requirements, the system can make corresponding adjustments based on these feedbacks to ensure that the faulty electrical equipment is handled most effectively. For the administrator, the feedback information processing function provided by the system enables it to participate more directly in the fault handling process, thereby improving its participation and satisfaction. At the same time, the system can also make real-time adjustments based on the administrator's feedback to ensure that the fault handling process is more in line with the administrator's expectations. By receiving and processing feedback information, the system can more accurately understand the actual situation and handling requirements of the faulty electrical equipment, thereby optimizing the use of resources. For example, the system can adjust the scheduling of operation and maintenance personnel, the use of spare parts, etc. based on the administrator's feedback to ensure the most effective use of resources.

[0044] In the embodiment of the present application, optionally, the second fault analysis prompt word includes content that guides the reasoning process of the large model output to generate the operation and maintenance information; The analysis and judgment disposal module is further used to input the second fault analysis prompt word into the analysis and judgment large model, and generate operation and maintenance information and reasoning process information of the faulty electrical equipment through the analysis and judgment large model; The fault handling module is further used to send the operation and maintenance information and the reasoning process information to the administrator terminal to prompt the administrator to confirm or modify the operation and maintenance information based on the reasoning information, and to provide feedback.

[0045] In the above embodiment, the second fault analysis prompt word is a specially designed prompt information used to guide the reasoning process of the large model output to generate operation and maintenance information. These prompt words may include problem definition, analysis steps, assumptions, data requirements, etc., to ensure that the large model can be reasoned according to the predetermined logical framework. The judgment and disposal module inputs the second fault analysis prompt word into the large model. According to these prompt words, the large model not only generates operation and maintenance information for the faulty electrical equipment, but also generates detailed reasoning process information. The reasoning process information may include how the model extracts features from the input data, how to apply the algorithm for reasoning, how to draw the final conclusion, etc. The fault disposal module sends the generated operation and maintenance information and reasoning process information to the administrator terminal together. In this way, the administrator can not only see the operation and maintenance suggestions, but also understand how these suggestions are derived, so as to be more trusting and willing to adopt these suggestions. After receiving the operation and maintenance information and reasoning process information, the administrator can confirm or modify the operation and maintenance information based on this information. At the same time, the administrator can also evaluate the accuracy and reliability of the large model according to the transparency of the reasoning process information, and give corresponding feedback. The embodiment of the present application introduces a second fault analysis prompt word and combines it with the research and judgment model. The second fault analysis prompt word guides the research and judgment model to reason according to a predetermined logical framework, which helps to ensure the accuracy and consistency of the research and judgment process. This helps to improve the accuracy and reliability of operation and maintenance information and reduce the possibility of misjudgment and missed judgment. After understanding the reasoning process of operation and maintenance information, the administrator can more accurately evaluate the rationality and feasibility of operation and maintenance suggestions. This helps administrators make more informed decisions and improve the efficiency and effectiveness of fault handling. By providing detailed reasoning process information, the system enhances the transparency of operation and maintenance information, and administrators can understand how operation and maintenance suggestions are derived, so that they can trust these suggestions more, which helps to establish administrators' trust in the system and improve the system's acceptance and usage rate. Feedback from administrators is an important source of continuous optimization and improvement of the system. By receiving administrators' evaluation and feedback on the reasoning process information, the system can identify potential problems and improvement points of the research and judgment model, and provide strong data support for future system upgrades and optimizations. In summary, by introducing the second fault analysis prompt word and combining it with the large model for analysis and judgment, the equipment health management system of the embodiment of the present application has achieved significant results and advantages in enhancing transparency and credibility, improving the accuracy of analysis and judgment, optimizing administrator decision-making, and promoting continuous optimization of the system. These improvements enable the system to better support the handling process of faulty electrical equipment, improve the efficiency and quality of operation and maintenance, and provide strong guarantees for the safe production and sustainable development of enterprises.

[0046] Further, as a refinement and extension of the specific implementation of the above embodiment, in order to fully illustrate the specific implementation process of this embodiment, an artificial intelligence-based device health management method is provided, such as Figure 2 As shown, the method includes: Step 201: The device data acquisition module acquires device environment data transmitted by sensors deployed in the influence area corresponding to each electrical device, and acquires device status data corresponding to each electrical device; Step 202: The data preprocessing module performs standardized preprocessing on the device environment data and the device status data; Step 203: The fault diagnosis module performs fault diagnosis on the electrical equipment based on the preprocessed equipment environment data and equipment status data to identify the faulty electrical equipment and generate fault information of the faulty electrical equipment; Step 204: the analysis and disposal module constructs a fault analysis prompt word according to the fault information, inputs the fault analysis prompt word into the analysis and determination large model, and generates fault disposal information for the faulty electrical equipment through the analysis and determination large model; Step 205: The fault handling module performs fault handling on the faulty electrical equipment based on the fault handling information.

[0047] Optionally, the device status data includes at least device current, device voltage, device power factor, and device temperature; the device environment data includes at least ambient temperature, ambient humidity, and near-end crosstalk; The fault diagnosis module performs fault diagnosis on the electrical equipment based on the preprocessed equipment environment data and equipment status data to identify the faulty electrical equipment and generate fault information of the faulty electrical equipment, including: The fault diagnosis module inputs the preprocessed equipment status data and equipment environment data into the fault diagnosis model, and the fault diagnosis model includes a vector conversion unit, a data fusion unit, a data enhancement unit, and a fault diagnosis unit; the vector conversion unit performs vector conversion on each item of input equipment status data and each item of input equipment environment data, the data fusion unit performs data fusion on the data vector, and the data enhancement unit performs data enhancement processing on the fused data vector to obtain a data feature vector, and the fault diagnosis unit performs fault diagnosis on the electrical equipment based on the data feature vector.

[0048] Optionally, it also includes: The fault diagnosis model training module obtains historical operation data of various operation states of each electrical device in the smart factory, wherein the historical operation data includes historical device state data, historical device environment data, a fault label indicating whether it is a fault state, and a type label indicating the device type of the electrical device. The various operation states include a fault operation state and various normal operation states. The various normal operation states include a stable operation state, a stress test operation state, and a pre-fault operation state. The stable operation state means that the electrical device maintains normal operation during the collection period corresponding to the historical normal operation data, and the preset periods before and after the collection period. The stress test operation state means that the electrical device maintains stable operation during the stress test process. The pre-fault operation state means that the electrical device maintains normal operation during the collection period corresponding to the historical normal operation data, and a fault occurs during the preset period after the collection period. Constructing an initial fault diagnosis model, the initial fault diagnosis model comprising an initial vector conversion unit, an initial data fusion unit, an initial data enhancement unit, a label discriminator, and initial fault diagnosis units having the same number as the sample groups; The initial fault diagnosis model is trained based on the training samples to obtain the fault diagnosis model; wherein, for any training sample, the training process includes: sequentially processing the historical equipment status data and the historical equipment environment data in the training sample by the initial vector conversion unit, the initial data fusion unit, and the initial data enhancement unit, discriminating the type labels in the training sample by the label discriminator to determine the initial fault diagnosis unit used in this round, inputting the data processed by the initial data enhancement unit into the initial fault diagnosis unit used in this round for fault label prediction, and optimizing the parameters of the initial vector conversion unit, the initial data fusion unit, the initial data enhancement unit, and the initial fault diagnosis unit used in this round in the initial fault diagnosis model based on the predicted fault label and the actual fault label corresponding to the training sample.

[0049] Optionally, it also includes: The device data acquisition module obtains the device identification of each electrical device; The data preprocessing module determines the device model based on the device identifier, and determines the device type label based on the device type corresponding to the device model; The fault diagnosis module inputs the pre-processed device status data and device environment data into the fault diagnosis model, including: The fault diagnosis module inputs the device type label, pre-processed device status data, and device environment data into the fault diagnosis model, performs vector conversion on each item of input device status data and each item of input device environment data respectively through the vector conversion unit, performs data fusion on the data vector through the data fusion unit, performs data enhancement processing on the fused data vector through the data enhancement unit to obtain a data feature vector, discriminates the input device type label through the label discriminator to determine a target fault diagnosis unit, and inputs the data feature vector into the target fault diagnosis unit for fault diagnosis.

[0050] Optionally, it also includes: The large model training module obtains the operation and maintenance records of the electrical equipment, which include the model of the operated and maintained electrical equipment, the equipment environment data and equipment status data of the operated and maintained electrical equipment, the fault detection information corresponding to the operated and maintained electrical equipment, the fault detection results, and the operation and maintenance records. The fault detection information indicates the information detected by the operation and maintenance personnel when troubleshooting the electrical equipment fault, the fault detection result indicates the detection result of the fault detection information, and the operation and maintenance record indicates the disposal method taken by the operation and maintenance personnel for the electrical equipment after obtaining the detection result; based on the model of the operated and maintained electrical equipment, the equipment environment data and equipment status data of the operated and maintained electrical equipment, and the fault detection information corresponding to the operated and maintained electrical equipment in the operation and maintenance records of the electrical equipment, the initial analysis and judgment large model is trained in one stage to train the ability of the initial analysis and judgment large model to predict fault detection information; and based on the model of the operated and maintained electrical equipment, the equipment environment data and equipment status data of the operated and maintained electrical equipment, the fault detection information corresponding to the operated and maintained electrical equipment, the fault detection results, and the operation and maintenance records, the initial analysis and judgment large model after the one-stage training is trained in two stages to train the ability of the initial analysis and judgment large model to generate operation and maintenance information, thereby obtaining the analysis and judgment large model.

[0051] Optionally, the fault information includes the faulty electrical equipment model, equipment environment data, and equipment status data of the faulty electrical equipment; The analysis and disposal module constructs a fault analysis prompt word according to the fault information, inputs the fault analysis prompt word into the analysis and judgment large model, and generates fault disposal information for the faulty electrical equipment through the analysis and judgment large model, including: The analysis and disposal module constructs a first fault analysis prompt word according to the fault information, inputs the first fault analysis prompt word into the analysis and judgment large model, and generates target fault detection information for the faulty electrical equipment through the analysis and judgment large model, wherein the target fault detection information includes information required for detection when troubleshooting the faulty electrical equipment; The method further comprises: The fault handling module performs an interface call based on the target fault detection information to obtain a fault detection result corresponding to the faulty electrical equipment; The analysis and disposal module constructs a second fault analysis prompt word according to the fault information and the fault detection result, inputs the second fault analysis prompt word into the analysis and judgment large model, and generates operation and maintenance information of the faulty electrical equipment through the analysis and judgment large model; The fault handling module performs fault handling on the faulty electrical equipment based on the fault handling information, including: The fault handling module sends the operation and maintenance information to the administrator terminal to prompt the administrator to handle the fault of the faulty electrical equipment based on the operation and maintenance information.

[0052] Optionally, it also includes: the fault handling module receives feedback information from the administrator terminal on the operation and maintenance information, and performs fault handling on the faulty electrical equipment based on the feedback information.

[0053] Optionally, the second fault analysis prompt includes content that guides the reasoning process of the large model output to generate the operation and maintenance information; and further includes: The analysis and disposal module inputs the second fault analysis prompt word into the analysis and judgment large model, and generates operation and maintenance information and reasoning process information of the faulty electrical equipment through the analysis and judgment large model; The fault handling module sends the operation and maintenance information and the reasoning process information to the administrator terminal to prompt the administrator to confirm or modify the operation and maintenance information based on the reasoning information, and to provide feedback.

[0054] For the corresponding description of the artificial intelligence-based device health management method provided in the embodiment of the present application, please refer to the corresponding description in the above system and will not be repeated here.

[0055] The embodiment of the present application also provides a computer device, which can be a personal computer, a server, a network device, etc. The computer device includes a bus, a processor, a memory and a communication interface, and can also include an input and output interface and a display device. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store location information. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the steps in each method embodiment are implemented.

[0056] Those skilled in the art will appreciate that the structure of the above-mentioned computer device is only a partial structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components, or combine certain components, or have a different arrangement of components.

[0057] In one embodiment, a computer-readable storage medium is provided. The computer-readable storage medium may be non-volatile or volatile, and stores a computer program thereon. When the computer program is executed by a processor, the steps in the above-mentioned method embodiments are implemented.

[0058] In one embodiment, a computer program product is provided, including a computer program, which implements the steps in the above method embodiments when executed by a processor.

[0059] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, stored data, displayed data, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties.

[0060] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0061] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0062] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. An artificial intelligence-based equipment health management system, characterized in that: Used to manage the health of electrical equipment in smart factories, including: The equipment data acquisition module is used to obtain equipment environment data transmitted by sensors deployed in the influence area corresponding to each electrical equipment, and to obtain equipment status data corresponding to each electrical equipment; A data preprocessing module, used for performing standardized preprocessing on the device environment data and the device status data; A fault diagnosis module, used to perform fault diagnosis on the electrical equipment based on the pre-processed equipment environment data and equipment status data, so as to identify the faulty electrical equipment and generate fault information of the faulty electrical equipment; A judgment and disposal module is used to construct a fault analysis prompt word according to the fault information, input the fault analysis prompt word into the judgment big model, and generate fault disposal information for the faulty electrical equipment through the judgment big model; A fault handling module is used to handle the fault of the faulty electrical equipment based on the fault handling information.

2. The artificial intelligence-based equipment health management system according to claim 1 is characterized in that: The device status data at least includes device current, device voltage, device power factor, and device temperature; the device environment data at least includes ambient temperature, ambient humidity, and near-end crosstalk; The fault diagnosis module is further used to input the preprocessed device status data and device environment data into a fault diagnosis model, wherein the fault diagnosis model includes a vector conversion unit, a data fusion unit, a data enhancement unit, and a fault diagnosis unit; The vector conversion unit performs vector conversion on each input device status data and each input device environment data respectively, the data fusion unit performs data fusion on the data vector, the data enhancement unit performs data enhancement processing on the fused data vector to obtain a data feature vector, and the fault diagnosis unit performs fault diagnosis on the electrical equipment based on the data feature vector.

3. The artificial intelligence-based equipment health management system according to claim 2 is characterized in that: Also includes: A fault diagnosis model training module is used to obtain historical operation data of various operation states of each electrical device in a smart factory, wherein the historical operation data includes historical device state data, historical device environment data, a fault label indicating whether it is a fault state, and a type label indicating the device type of the electrical device. The various operation states include a fault operation state and various normal operation states. The various normal operation states include a stable operation state, a stress test operation state, and a pre-fault operation state. The stable operation state means that the electrical device maintains normal operation during the collection period corresponding to the historical normal operation data, and the preset periods before and after the collection period. The stress test operation state means that the electrical device maintains stable operation during the stress test process. The pre-fault operation state means that the electrical device maintains normal operation during the collection period corresponding to the historical normal operation data, and a fault occurs during the preset period after the collection period. Constructing an initial fault diagnosis model, the initial fault diagnosis model comprising an initial vector conversion unit, an initial data fusion unit, an initial data enhancement unit, a label discriminator, and initial fault diagnosis units having the same number as the sample groups; The initial fault diagnosis model is trained based on the training samples to obtain the fault diagnosis model; wherein, for any training sample, the training process includes: sequentially processing the historical equipment status data and the historical equipment environment data in the training sample by the initial vector conversion unit, the initial data fusion unit, and the initial data enhancement unit, discriminating the type labels in the training sample by the label discriminator to determine the initial fault diagnosis unit used in this round, inputting the data processed by the initial data enhancement unit into the initial fault diagnosis unit used in this round for fault label prediction, and optimizing the parameters of the initial vector conversion unit, the initial data fusion unit, the initial data enhancement unit, and the initial fault diagnosis unit used in this round in the initial fault diagnosis model based on the predicted fault label and the actual fault label corresponding to the training sample.

4. The artificial intelligence-based equipment health management system according to claim 3 is characterized in that: The device data acquisition module is also used to obtain the device identification of each electrical device; The data preprocessing module is further used to determine the device model based on the device identifier, and determine the device type label based on the device type corresponding to the device model; The fault diagnosis module is also used to input the device type label, pre-processed device status data, and device environment data into the fault diagnosis model, perform vector conversion on each item of input device status data and each item of input device environment data respectively through the vector conversion unit, perform data fusion on the data vector through the data fusion unit, perform data enhancement processing on the fused data vector through the data enhancement unit to obtain a data feature vector, discriminate the input device type label through the label discriminator to determine the target fault diagnosis unit, and input the data feature vector into the target fault diagnosis unit for fault diagnosis.

5. The artificial intelligence-based equipment health management system according to any one of claims 1 to 4, characterized in that: Also includes: A large model training module is used to obtain electrical equipment operation and maintenance records, wherein the electrical equipment operation and maintenance records include the model of the operated and maintained electrical equipment, the equipment environment data and equipment status data of the operated and maintained electrical equipment, the fault detection information corresponding to the operated and maintained electrical equipment, the fault detection results, and the operation and maintenance records. The fault detection information indicates the information detected by the operation and maintenance personnel when troubleshooting electrical equipment faults, the fault detection results indicate the detection results of the fault detection information, and the operation and maintenance records indicate the disposal methods taken by the operation and maintenance personnel for the electrical equipment after obtaining the detection results; based on the model of the operated and maintained electrical equipment, the equipment environment data and equipment status data of the operated and maintained electrical equipment, and the fault detection information corresponding to the operated and maintained electrical equipment in the electrical equipment operation and maintenance records, the initial analysis and judgment large model is trained in one stage to train the ability of the initial analysis and judgment large model to predict fault detection information; and based on the model of the operated and maintained electrical equipment, the equipment environment data and equipment status data of the operated and maintained electrical equipment, the fault detection information corresponding to the operated and maintained electrical equipment, the fault detection results, and the operation and maintenance records, the initial analysis and judgment large model after the one-stage training is trained in two stages to train the ability of the initial analysis and judgment large model to generate operation and maintenance information, thereby obtaining the analysis and judgment large model.

6. The artificial intelligence-based equipment health management system according to claim 5 is characterized in that: The fault information includes the faulty electrical equipment model, equipment environment data, and equipment status data of the faulty electrical equipment; The analysis and judgment disposal module is further used to construct a first fault analysis prompt word according to the fault information, input the first fault analysis prompt word into the analysis and judgment large model, and generate target fault detection information for the faulty electrical equipment through the analysis and judgment large model, wherein the target fault detection information includes information required for detection when troubleshooting the faulty electrical equipment; The fault handling module is further used to perform an interface call based on the target fault detection information to obtain a fault detection result corresponding to the faulty electrical equipment; The analysis and disposal module is further used to construct a second fault analysis prompt word according to the fault information and the fault detection result, input the second fault analysis prompt word into the analysis and judgment large model, and generate operation and maintenance information of the faulty electrical equipment through the analysis and judgment large model; The fault handling module is further used to send the operation and maintenance information to an administrator terminal to prompt the administrator to handle the fault of the faulty electrical equipment based on the operation and maintenance information.

7. The artificial intelligence-based equipment health management system according to claim 6, characterized in that: The fault handling module is further used to receive feedback information from the administrator terminal on the operation and maintenance information, and perform fault handling on the faulty electrical equipment based on the feedback information.

8. The artificial intelligence-based equipment health management system according to claim 7, characterized in that: The second fault analysis prompt word includes content that guides the reasoning process of the large model output to generate the operation and maintenance information; The analysis and judgment disposal module is further used to input the second fault analysis prompt word into the analysis and judgment large model, and generate operation and maintenance information and reasoning process information of the faulty electrical equipment through the analysis and judgment large model; The fault handling module is further used to send the operation and maintenance information and the reasoning process information to the administrator terminal to prompt the administrator to confirm or modify the operation and maintenance information based on the reasoning information, and to provide feedback.

9. An artificial intelligence-based equipment health management method, characterized in that: An artificial intelligence-based equipment health management system applied to any one of claims 1 to 8, comprising: The equipment data acquisition module acquires equipment environment data transmitted by sensors deployed in the influence area corresponding to each electrical equipment, and acquires equipment status data corresponding to each electrical equipment; The data preprocessing module performs standardized preprocessing on the device environment data and the device status data; The fault diagnosis module performs fault diagnosis on the electrical equipment based on the preprocessed equipment environment data and equipment status data to identify the faulty electrical equipment and generate fault information of the faulty electrical equipment; The analysis and disposal module constructs a fault analysis prompt word according to the fault information, inputs the fault analysis prompt word into the analysis and judgment large model, and generates fault disposal information for the faulty electrical equipment through the analysis and judgment large model; The fault handling module performs fault handling on the faulty electrical equipment based on the fault handling information.

10. A computer device comprising a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, characterized in that: When the processor executes the computer program, the method according to claim 9 is implemented.

Images