Knowledge-based fault diagnosis method

Abstract

The invention discloses a knowledge-based fault diagnosis method. The method comprises the following steps: establishing a typical fault case database; extracting fault feature information and selecting main data variables by adopting a signal processing technology and a main component analysis method; Learning the fault sample by using an improved neural network algorithm; the comprehensive representation of the diagnosis knowledge is realized by adopting a mode of combining a generation type rule and a framework; constructing a system fault propagation model through data analysis and the relationship among the devices in the system, and defining a system fault propagation path; and based on the obtained fault knowledge, constructing a knowledge graph, and giving an inference rule by utilizing a road algebra theory to realize fault positioning based on knowledge inference. Through an improved neural network algorithm and a knowledge reasoning theory, automatic acquisition of knowledgeand accurate positioning of faults are realized, the diagnosis precision is improved, and a basis is provided for fault diagnosis of an industrial control system.

Description

technical field [0001] The invention relates to a fault diagnosis method of an industrial control system, in particular to a knowledge-based fault diagnosis method. Background technique [0002] With the integration of industrialization and industrialization, industry 4.0, and the implementation of the concepts of Made in China 2025, traditional factories and industrial control systems are bound to become interconnected and integrated, and the maintenance costs of equipment are getting higher and higher. How to improve the reliability of equipment To ensure the safe, stable and long-term operation of equipment has become an urgent problem in the field of fault diagnosis of intelligent production equipment. Fault diagnosis technology runs through the entire life process of industrial equipment design, development, manufacturing, testing, and maintenance. Fault diagnosis software has become an important means of industrial equipment design and development, a quality assurance ...

AI Technical Summary

Problems solved by technology

[0005] Although most of the modern fault diagnosis and prediction systems can automatically diagnose the faults of the measured objects at a very fast speed, with the improvement of equipment intelligence, networking and complexity, the existing systems are likely to cause ambiguity in fault diagnosis. and uncertainty, and the automaticity of fault knowledge acquisition is poor

Images