Intelligent Diagnosis Method Based on Relative Ratio Symptom Parameters and Support Vector Machines for Rotating Machinery

摘要

Many intelligent diagnosis methods based on the Traditional Statistical Theory, such as Neural Networks, Genetic Algorithms, etc., have been proposed in the field of mechanical fault diagnosis. These methods depend on the assumption that the number of samples tends to infinity, and require a large amount of training samples and highly sensitive symptom parameters (SPs). However, in many cases of condition diagnosis for rotating machinery, because the demanding samples cannot be acquired easily in a real plant, calculated SPs are not highly sensitive, and the intelligent methods, namely neural networks, genetic algorithms, etc., often cannot converge when learning. In order to solve these problems, this paper proposes a new parameter called Relative Ratio Symptom Parameter (RRSP) in the low-frequency area for diagnose structural faults of rotating machinery, and we combined it with Support Vector Machines (SVMs) to automatically diagnose structural faults of rotating machinery. The practical examples of diagnosis for rotating machinery are shown to verify the efficiency of the methods.