Product equipment fault maintenance knowledge search recommendation method based on fusion recommendation algorithm

Abstract

A product equipment fault maintenance knowledge search recommendation method based on a fusion recommendation algorithm comprises the following steps: acquiring an equipment model identified by a maintenance personnel through photographing a nameplate or manually input, and acquiring information related to an equipment fault described through voice or manually input; processing the data acquired in the information acquisition step, identifying nameplate information of a product or equipment through a character identification technology to extract an equipment model, and identifying and extracting fault core keywords in voice or characters; based on the keyword extraction step, generating a maintenance scheme recommendation result through a QDT algorithm, a feature type mixed recommendation algorithm and a training model, and outputting the maintenance scheme recommendation result; and feedback information of the maintenance personnel on the recommendation result is received, and the feature type mixed recommendation algorithm model is optimized. According to the method, the precision and efficiency of maintenance knowledge recommendation can be remarkably improved, particularly, maintenance personnel with less experience can be helped to quickly complete equipment fault diagnosis and maintenance, the maintenance cost is reduced, and the production efficiency is improved.

Description

Technical Field

[0001] The present invention belongs to the technical fields of artificial intelligence and data mining, especially a knowledge recommendation system for the maintenance of industrial products and equipment. In particular, it relates to a method for searching and recommending product equipment fault maintenance knowledge based on a fusion recommendation algorithm, which is particularly applicable to the maintenance of industrial equipment that requires a large amount of experience and professional knowledge support. Background Art

[0002] In the field of industrial manufacturing, the maintenance problems of manufactured products and production equipment failures have become increasingly complex. Traditional maintenance methods mainly rely on the rich experience of maintenance personnel to quickly complete maintenance tasks. However, in the manufacturing industry, most maintenance personnel have insufficient experience. When faced with complex equipment failures, they often cannot quickly find the best way to solve the problem. Existing maintenance knowledge bases mostly rely on simple keyword searches, but this method has problems such as inaccurate, incorrect recommendation results and low recommendation efficiency. Therefore, a system that can effectively improve the accuracy and efficiency of maintenance knowledge recommendation is needed in the maintenance knowledge base.

[0003] In the prior art, although there are some systems based on data mining and recommendation algorithms, most recommendation systems are relatively single in algorithm application and fail to effectively integrate multiple algorithms, resulting in low recommendation accuracy. Moreover, when dealing with complex data, intelligent dynamic optimization cannot be achieved. Summary of the Invention

[0004] In view of the technical problems in the prior art, the technical solution of the method for searching and recommending product equipment fault maintenance knowledge based on a fusion recommendation algorithm of the present invention is proposed. The aim is to improve the accurate recommendation ability of maintenance knowledge by integrating different recommendation algorithms, help maintenance personnel quickly locate faults, and provide solutions.

[0005] The method for searching and recommending product equipment fault maintenance knowledge based on a fusion recommendation algorithm provided by the present invention includes:

[0006] Information collection step: Obtain the equipment model recognized by the maintenance personnel through photographing the nameplate or manually inputting, and obtain the information related to the equipment failure described by voice or manually input.

[0007] Keyword extraction step: Process the data collected in the information collection step, identify the nameplate information of the product or equipment through optical character recognition technology to extract the equipment model, and identify and extract the core fault keywords in the voice or text.

[0008] Matching and recommendation step: Based on the steps extracted in the keyword extraction step, generate and output the recommended results of the maintenance plan through the QDT algorithm, the feature-based hybrid recommendation algorithm, and the training model.

[0009] Feedback step: used to receive the feedback information of the maintenance personnel on the recommendation results and optimize the feature-based hybrid recommendation algorithm model for subsequent use.

[0010] Preferably, in the product equipment fault maintenance knowledge search and recommendation method based on the fusion recommendation algorithm of the present invention, the matching recommendation step specifically includes the following steps:

[0011] Use the QDT algorithm to extract and match the features of the processed fault-related information, and determine the type of product or equipment fault and possible maintenance solutions;

[0012] According to the feature-based hybrid recommendation algorithm, combine historical maintenance data with the fault type of the product or equipment to generate multiple potential maintenance solutions.

[0013] Preferably, in the product equipment fault maintenance knowledge search and recommendation method based on the fusion recommendation algorithm of the present invention, the information related to the equipment fault includes the fault type and the fault phenomenon.

[0014] Preferably, in the product equipment fault maintenance knowledge search and recommendation method based on the fusion recommendation algorithm of the present invention, the recommended results of the maintenance plan include the recommended maintenance methods, operation steps, precautions, and tools and jigs required for maintenance, and help the maintenance personnel quickly repair the product or equipment fault through video or graphic methods.

[0015] Preferably, in the product equipment fault maintenance knowledge search and recommendation method based on the fusion recommendation algorithm of the present invention, the text recognition technology includes OCR text recognition technology.

[0016] Preferably, in the product equipment fault maintenance knowledge search and recommendation method based on the fusion recommendation algorithm of the present invention, the identification and extraction of the fault core keywords in the voice or text are realized through NLP technology.

[0017] The product equipment fault maintenance knowledge search and recommendation method based on the fusion recommendation algorithm provided by the present invention can significantly improve the accuracy and efficiency of the maintenance knowledge recommendation through the integration of multiple technologies. In particular, it can help the maintenance personnel with less experience quickly complete the diagnosis and maintenance of equipment faults, reduce the maintenance cost, and improve the production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The present invention will be further described below in conjunction with the drawings and embodiments. In the drawings:

[0019] Figure 1 is a flowchart of an embodiment of the product equipment fault maintenance knowledge search and recommendation method based on the fusion recommendation algorithm of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0020] For a clearer understanding of the technical features, objectives, and effects of the present invention, the specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0021] Reference Figure 1 , the method for searching and recommending product equipment fault repair knowledge based on a fusion recommendation algorithm provided by the present invention includes:

[0022] Information collection step: Obtain the equipment model identified by the maintenance personnel through photographing the nameplate or manual input, and obtain the information related to the equipment fault (including the fault type and fault phenomenon) described by voice or manual input.

[0023] Keyword extraction step: Process the data collected in the information collection step, identify the nameplate information of the product or equipment through OCR character recognition technology to extract the equipment model, and identify and extract the core fault keywords in the voice or text through NLP technology.

[0024] Matching and recommendation step: Based on the steps extracted in the keyword extraction step, generate a recommended result of the repair plan through the QDT algorithm, the feature-based hybrid recommendation algorithm, and the training model and output it; specifically, it includes the following steps:

[0025] (1) Use the QDT algorithm to perform feature extraction and matching on the processed information related to the fault, determine the type of product or equipment fault and the possible repair plans, including the recommended repair methods, operation steps, precautions, and tools and fixtures required for the repair, and help the maintenance personnel quickly repair the product or equipment fault through video or graphic means

[0026] (2) According to the feature-based hybrid recommendation algorithm, combine the historical repair data with the fault type of the product or equipment to generate multiple potential repair solutions.

[0027] (3) Use the training model to optimize the recommended result in real time, and the system continuously improves the accuracy of the recommendation according to the historical feedback.

[0028] Feedback step: Used to receive the feedback information of the maintenance personnel on the recommended result and optimize the feature-based hybrid recommendation algorithm model for subsequent use.

[0029] Key code implementation (simplified example code):

[0030] # Example: Fault and repair matching based on the QDT algorithm

[0031] import numpy as np

[0032] from sklearn.tree import DecisionTreeClassifier

[0033] # Training Data (Fault Types and Repair Solutions)

[0034] X_train = np.array([[0, 1], [1, 0], [1, 1], [0, 0]]) # Example features

[0035] y_train = np.array([0, 1, 0, 1]) # Corresponding repair solutions

[0036] # Build a decision tree classifier

[0037] clf = DecisionTreeClassifier(random_state = 42)

[0038] clf.fit(X_train, y_train)

[0039] # Input a new fault feature for prediction

[0040] new_fault = np.array([[1, 0]]) # Features of the new fault

[0041] recommended_solution = clf.predict(new_fault) # Output the recommended repair solution

[0042] print(f"Recommended repair solution: {recommended_solution}")

[0043] This code implements a simple fault diagnosis and repair solution recommendation based on the QDT algorithm. By training the model to classify the input fault features, the appropriate repair method is recommended.

[0044] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms without departing from the spirit and scope protected by the present invention and the claims. These all fall within the protection scope of the present invention.

Claims

1. A product equipment fault maintenance knowledge search and recommendation method based on a fusion recommendation algorithm, characterized in that: Include: Information collection steps: obtain the equipment model identified by taking a photo of the nameplate or manually input by the maintenance personnel, and obtain the information related to the equipment failure through voice description or manual input; Keyword extraction step: Process the data collected in the information collection step, identify the nameplate information of the product or equipment through text recognition technology to extract the equipment model, and identify and extract the core keywords of the fault in the voice or text; Matching recommendation step: Based on the steps extracted in the keyword extraction step, the maintenance plan recommendation results are generated and output through the QDT algorithm, the feature-based hybrid recommendation algorithm and the training model; Feedback step: used to receive feedback from maintenance personnel on the recommendation results and optimize the feature-based hybrid recommendation algorithm model.

2. The product equipment fault maintenance knowledge search and recommendation method based on the fusion recommendation algorithm according to claim 1 is characterized in that: The matching recommendation step specifically includes the following steps: Use the QDT algorithm to extract and match features of the processed fault-related information to determine the type of product or equipment fault and possible repair solutions; Based on the feature-based hybrid recommendation algorithm, historical maintenance data and the fault type of the product or equipment are combined to generate multiple potential maintenance solutions; The training model is used to optimize the recommendation results in real time, and the system continuously improves the accuracy of recommendations based on historical feedback.

3. The product equipment fault maintenance knowledge search and recommendation method based on the fusion recommendation algorithm according to claim 1 is characterized in that: The information related to the equipment failure includes the failure type and the failure phenomenon.

4. The product equipment fault maintenance knowledge search and recommendation method based on the fusion recommendation algorithm according to claim 1 is characterized in that: The maintenance plan recommendation results include recommended maintenance methods, operating steps, precautions, and tools and fixtures required for maintenance, which help maintenance personnel quickly repair product or equipment failures through videos or graphics.

5. The product equipment fault maintenance knowledge search and recommendation method based on the fusion recommendation algorithm according to claim 1 is characterized in that: The text recognition technology includes OCR text recognition technology.

6. The product equipment fault maintenance knowledge search and recommendation method based on the fusion recommendation algorithm according to claim 1 is characterized in that: The identification and extraction of core fault keywords in speech or text is achieved through NLP technology.

Images