An adaptive threshold based on support vector machine for fault diagnosis

Considering the drawback of the big error when using fixed threshold in fault diagnosis for hydraulic servo system, many factors that may affect the fault threshold are analyzed. By integrating the key factors in threshold model, such as modeling error, random disturbance, input instructions, system current status and etc, an adaptive threshold scheme for fault diagnosis is proposed in this paper, which is based on a pattern recognition algorithm called support vector machine (SVM). It is very effective to adaptively adjust the fault threshold according to a variety of influencing factors. And the robustness is improved by the proposed method, which is verified by experimental results.