Research On AIS Based Fault Diagnosis Method And System

The technology of machinery and equipment monitoring and diagnosis assures us that the important machinery can run in a safety, reliable, long-term and efficient way. It plays a key role both in eliminating hidden fault and avoiding the appearance of accident. Based on Independent Component Analysis"(National Nature Science Fund Project, NO: 50205025), this paper takes "Artificial Immune System"(next for short as "AIS") as its theory, studies the intelligent fault diagnosis system of large rotating machinery.This paper firstly studies the principium, common models and some related arithmetic of AIS, then we build the AIS response model for fault diagnosis with the traits of the machinery fault diagnosis; design and develop the common-used fault diagnosis system for machinery based AIS, then, a test for the system performance has been taken; finally, we use Web Services to finish the integration of the fault diagnosis, upgrade our system as a service published in the internet, realize the remote distribute corporation diagnosis.Chapter one discusses the significance of monitoring and diagnosis of the machinery, especially of large rotating machinery, describes its development history and current status, point out its future development orientation. As a newly developing intelligent method, AIS is introduced briefly, including its current research and application status. At the end of chapter one, the research background, main contents, general structure scheme are presented.Chapter two introduces the biology immune principles and the principium, common model and some related arithmetic of AIS, which is developing from the metaphor of biology immune principles. Finally, the similarities and differences among AIS, ANN and GA are discussed.Chapter three describes the classic diagnosis method for rotating machinery, studies the diagnosis problem of rotating machinery based on AIS, build the response model of fault diagnosis with the help of negative selection arithmetic and common immune arithmetic. We research and discuss deeply on the related arithmetic and parameters, finish the local improvement.Chapter four presents the structure of a common machine fault diagnosis system based on the AIS response model, then designs and develops the fault diagnosis system and finishes the testing work.In chapter five, we analyses the application situation and develop aspects of distributed remote fault detection and diagnosis, uses Web Services technology to finish the integration of the fault diagnosis system based on AIS.Chapter six sums up the work of the dissertation and presents conclusion and prospect.