A Method of Intelligent Diagnosis of Equipment Fault Based on Support Vector Machine

Abstract

An intelligent diagnosis method for equipment faults based on support vector machines, including preprocessing operations for equipment data; constructing a fault diagnosis case knowledge base; performing fault diagnosis on support vector machines; obtaining fault information and providing maintenance guidance, based on support vector machines The intelligent diagnosis method of equipment faults highlights the fault characteristics of the equipment to the greatest extent, reduces the situation of incomplete and inaccurate equipment data, provides the possibility for building an accurate and reliable fault diagnosis model, solves the problem of the aging of the diagnosis model with the operation time of the equipment, and reduces the The misdiagnosis rate of the fault diagnosis model greatly improves the accuracy and speed of equipment fault diagnosis.

Description

Technical field [0001] The present invention mainly relates to the technical field of equipment fault diagnosis, and more specifically to a method in the field of diagnosis research based on support vector machines. Background technique [0002] In some important occasions such as power plants, iron smelting plants, satellite launch sites, etc., in order to ensure the safe operation of their key equipment, factories often invest a large number of maintenance personnel to ensure the safe operation of the equipment. However, due to problems such as their own diagnostic technology or attention, the staff will inevitably ignore some signs of equipment abnormality. Once the equipment abnormality develops into a production failure, it will bring huge economic losses to the enterprise. Therefore, researchers in related fields have invested a lot of energy in establishing equipment intelligent diagnosis systems for manufacturing enterprises to ensure the safe operation of important equip...

AI Technical Summary

Problems solved by technology

The present invention can highlight the fault characteristics of the equipment to the greatest extent through the regression filtering method, reduce the incomplete and inaccurate situation of the equipment data, and provide the possibility for building an accurate and reliable fault diagnosis model; the present invention is based on the use of support vector machines , use a simple and convenient way to realize the function of model incremental learning, to solve the problem of diagnostic model aging with equipment running time, and reduce the misdiagnosis rate of fault diagnosis model; the present invention uses support vector machines on the basis of identifying new faults , to realize the continuous improvement of the fault knowledge base; the fault diagnosis model realized by the present invention can provide more detailed fault diagnosis information, not only the simple fault type, but also includes the reliability, fault location and origin of each fault that this fault belongs to The fault maintenance guidance and suggestions extracted from the expert knowledge base greatly improved the accuracy and speed of equipment fault diagnosis

Images