A work ticket content recommendation method and system for intelligent operation and maintenance

Abstract

The application relates to a work ticket content recommendation method for intelligent operation and maintenance, which comprises the following steps: S1: uniformly accessing maintenance related data to an unstructured component of a national grid data middle platform, and completing quality evaluation of related files by a quality detection tool; S2: after uniform access, the unstructured component carries out text extraction on document type data uploaded to the unstructured component to form to-be-labeled data through an xxl-job distributed task scheduling system; S3: adopting a bart labeling technology of a Transformer to carry out text labeling on the to-be-labeled data and positioning core content of a work ticket; S4: constructing a work ticket content automatic recommendation model based on a time sequence prediction LightGBM neural network, and carrying out API encapsulation and containerization deployment; and S5: integrating the work ticket content automatic recommendation model with a business system in a micro-service mode, and recommending work ticket preparation content in real time according to problem description uploaded by an operation and maintenance personnel. The application realizes real-time recommendation of high-quality work ticket content.

Description

Technical Field

[0001] This invention relates to the field of power system technology, and in particular to a method and system for recommending work order content for intelligent operation and maintenance. Background Technology

[0002] With the advancement of the national smart grid informatization construction, diverse computer technologies have been applied to power grid construction. Provincial grid companies have cumulatively invested heavily in information equipment and systems to support power operations. These various devices and systems have integrated a vast and heterogeneous power internal network IT environment. Traditional manpower-based operation and maintenance methods are becoming increasingly inadequate, posing significant challenges to power IT operation and maintenance personnel. In particular, large-scale systems uniformly deployed by multiple provincial units, such as the State Grid unstructured data management platform and collaborative office system, are experiencing problems with current operation and maintenance methods, including errors in work order content and high personnel costs, severely impacting on-site operation and maintenance efficiency. Summary of the Invention

[0003] To address the aforementioned problems, the present invention aims to provide a method and system for recommending work order content for intelligent operation and maintenance. This method autonomously learns from unstructured historical work order maintenance records on the platform, automatically generating work order maintenance content. This assists in the manual preparation of work order maintenance plans, thereby improving the work efficiency of platform operation and maintenance personnel. To achieve the above objectives, the present invention adopts the following technical solution: A method for recommending work order content for intelligent operation and maintenance includes the following steps: S1: All maintenance-related data is uniformly integrated into the unstructured components of the State Grid data platform, and the quality assessment of related documents is completed by quality inspection tools; S2: After unified access, unstructured components use the xxl-job distributed task scheduling system to extract text from document data uploaded to the unstructured components to form data to be labeled. S3: Use Transformer's bart annotation technology to annotate the data to be annotated and locate the core content of the work order; S4: Build an automatic recommendation model for work ticket content based on the LightGBM neural network for time series prediction, and encapsulate the API and deploy it in a containerized manner; S5: The automatic work order content recommendation model is integrated with the business system in a microservice manner. It recommends the content of the work order in real time based on the maintenance plan, abnormal log materials and work order problem description uploaded by the operation and maintenance personnel.

[0004] Furthermore, maintenance-related data, including power maintenance plans, anomaly logs, and work orders, are uniformly connected to the unstructured components of the State Grid data platform via the State Grid webservice / HTTP path.

[0005] Furthermore, the document quality inspection tool is the TextIn document quality inspection tool, used to complete the quality assessment of work order forms and scanned images.

[0006] Furthermore, the automatic recommendation model for work ticket content is constructed as follows: the training and validation sets required for modeling are read from Elasticsearch; Pandas is used to preprocess the data before training to optimize the data; and the LightGBM neural network for time series prediction, using the Tensorflow machine learning framework, is used to complete the construction of the automatic recommendation model for work ticket content.

[0007] Furthermore, the API is encapsulated and containerized for deployment, specifically as follows: The FastAPI framework is used to package the model into a RESTful web service interface, exposing standardized public protocol interfaces and supporting HTTP / HTTPS request and response mechanisms; by defining standard API interface specifications, including input parameter formats, output result structures, and exception handling mechanisms, seamless integration of the model service with various business systems is ensured; then, Docker containerization technology is used to package the encapsulated model service, creating a complete Docker image containing the model runtime environment, dependency libraries, and configuration files, achieving standardized application deployment and cross-platform compatibility; finally, the Kubernetes (K8s) container orchestration platform is used for the release and management of the model service, including service deployment, load balancing, automatic scaling, and health checks.

[0008] A work order content recommendation system for intelligent operation and maintenance includes a processor, a memory, and a computer program stored in the memory. When the processor executes the computer program, it specifically performs the steps in the work order content recommendation method for intelligent operation and maintenance described above.

[0009] A computer storage medium storing a plurality of instructions adapted for loading by a processor and executing the method steps described above.

[0010] The present invention has the following beneficial effects: This invention autonomously learns from the historical work order maintenance records of an unstructured platform, enabling the automatic generation of work order maintenance content. This assists in the manual preparation of work order maintenance plans, thereby improving the work efficiency of platform maintenance personnel. Attached Figure Description

[0011] Figure 1 This is a flowchart of the method of the present invention; Figure 2 This is a schematic diagram of model construction and publishing in one embodiment of the present invention. Detailed Implementation

[0012] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments: refer to Figure 1 In this embodiment, a method for recommending work ticket content for intelligent operation and maintenance is provided, including the following steps: S1: All maintenance-related data will be uniformly accessed into the unstructured components of the State Grid data platform, and the quality inspection tool will complete the quality assessment of the relevant documents. If there are obvious problems such as incomplete work order forms or scanned images, or blurry fonts, the documents will be returned and the relevant business systems will be notified to re-upload and replace the documents that do not meet the quality requirements. S2: After unified access, unstructured components use the xxl-job distributed task scheduling system to extract text from document data uploaded to the unstructured components to form data to be labeled. Preferably, document data extraction uses the Tika component to perform full-text extraction for common document types such as PDF, DOC, and TXT. For file types such as CEB, a document conversion service is used to convert CEB files to PDF files before full-text extraction. For scanned documents, Baidu OCR technology from the AI ​​platform is used for extraction.

[0013] S3: Use Transformer's bart annotation technology to annotate the data to be annotated and locate the core content of the work order; S4: Build an automatic recommendation model for work ticket content based on the LightGBM neural network for time series prediction, and encapsulate the API and deploy it in a containerized manner; S5: The automatic work order content recommendation model is integrated with the business system in a microservice manner. It recommends the content of the work order in real time based on the maintenance plan, abnormal log materials and work order problem description uploaded by the operation and maintenance personnel.

[0014] In this implementation, maintenance-related data, including power maintenance plans, anomaly logs, and work orders, are uniformly accessed to the unstructured components of the State Grid data platform via the State Grid webservice / HTTP path.

[0015] In this embodiment, the document quality inspection tool is the TextIn document quality inspection tool, which is used to evaluate the quality of work order forms and scanned images.

[0016] In this embodiment, reference Figure 2The automatic recommendation model for work ticket content is constructed as follows: Training and validation sets are read from Elasticsearch; Pandas is used to preprocess the data before training to optimize it; and the LightGBM neural network for time series prediction, using the Tensorflow machine learning framework, is used to complete the construction of the automatic recommendation model for work ticket content.

[0017] In this embodiment, API encapsulation and containerized deployment are performed as follows: The FastAPI framework is used to package the model into a RESTful web service interface, exposing standardized public protocol interfaces and supporting HTTP / HTTPS request and response mechanisms; by defining standard API interface specifications, including input parameter formats, output result structures, and exception handling mechanisms, it is ensured that the model service can be seamlessly integrated with various business systems; then, Docker containerization technology is used to package the encapsulated model service, creating a complete Docker image containing the model runtime environment, dependency libraries, and configuration files, achieving standardized application deployment and cross-platform compatibility; finally, the Kubernetes (K8s) container orchestration platform is used for the release management of the model service, including service deployment, load balancing, automatic scaling, and health checks.

[0018] A work order content recommendation system for intelligent operation and maintenance includes a processor, a memory, and a computer program stored in the memory. When the processor executes the computer program, it specifically performs the steps in the work order content recommendation method for intelligent operation and maintenance described above.

[0019] A computer storage medium storing a plurality of instructions adapted for loading by a processor and executing the method steps described above. Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0020] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0021] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0022] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0023] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any other way. Any person skilled in the art may make changes or modifications to the above-disclosed technical content to create equivalent embodiments. However, any simple modifications, equivalent changes, and modifications made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the protection scope of the present invention.

Claims

1. A method for recommending work order content for intelligent operation and maintenance, characterized in that, Includes the following steps: S1: All maintenance-related data is uniformly integrated into the unstructured components of the State Grid data platform, and the quality assessment of related documents is completed by quality inspection tools; S2: After unified access, unstructured components use the xxl-job distributed task scheduling system to extract text from document data uploaded to the unstructured components to form data to be labeled. S3: Use Transformer's bart annotation technology to annotate the data to be annotated and locate the core content of the work order; S4: Build an automatic recommendation model for work ticket content based on the LightGBM neural network for time series prediction, and encapsulate the API and deploy it in a containerized manner; S5: The automatic work order content recommendation model is integrated with the business system in a microservice manner. It recommends the content of the work order in real time based on the maintenance plan, abnormal log materials and work order problem description uploaded by the operation and maintenance personnel.

2. The method for recommending work order content for intelligent operation and maintenance according to claim 1, characterized in that, The maintenance-related data includes power maintenance plans, anomaly logs, and work orders, which are uniformly accessed to the unstructured components of the State Grid data platform via the State Grid webservice / HTTP path.

3. The method for recommending work order content for intelligent operation and maintenance according to claim 1, characterized in that, The document quality inspection tool is the TextIn document quality inspection tool, which is used to evaluate the quality of work order forms and scanned images.

4. The method for recommending work order content for intelligent operation and maintenance according to claim 1, characterized in that, The automatic recommendation model for work ticket content is constructed as follows: the training and validation sets required for modeling are read from Elasticsearch; Pandas is used to preprocess the data before training to optimize the data; and the LightGBM neural network for time series prediction in the Tensorflow machine learning framework is used to complete the construction of the automatic recommendation model for work ticket content.

5. The method for recommending work order content for intelligent operation and maintenance according to claim 1, characterized in that, The API encapsulation and containerized deployment are as follows: The FastAPI framework is used to package the model into a RESTful web service interface, exposing standardized public protocol interfaces and supporting HTTP / HTTPS request / response mechanisms. By defining standard API interface specifications, including input parameter formats, output result structures, and exception handling mechanisms, seamless integration of the model service with various business systems is ensured. Next, Docker containerization technology is used to package the encapsulated model service, creating a complete Docker image containing the model runtime environment, dependency libraries, and configuration files, achieving standardized application deployment and cross-platform compatibility. Finally, the Kubernetes (K8s) container orchestration platform is used for the release and management of the model service, including service deployment, load balancing, automatic scaling, and health checks.

6. A work order content recommendation system for intelligent operation and maintenance, characterized in that, It includes a processor, a memory, and a computer program stored in the memory. When the processor executes the computer program, it specifically performs the steps in the work ticket content recommendation method for intelligent operation and maintenance as described in any one of claims 1-5.

7. A computer storage medium, characterized in that, The computer storage medium stores a plurality of instructions, which are adapted to be loaded by a processor and executed as the method steps of any one of claims 1 to 5.

Images