Application of Information-Geometric Support Vector Machine on Fault Diagnosis of Hydraulic Pump

The growing demand for the safety and reliability in industries triggers the development of condition monitoring and fault diagnosis technologies. Hydraulic pump is the critical part of a hydraulic system. The diagnosis of hydraulic pump is very crucial for reliability. This paper presents a method based on information-geometric support vector machine (IG-SVM), which is employed for fault diagnosis of hydraulic pump. The IG-SVM, which uses information geometry to modify SVM, improves the performance in a data dependent way. To diagnose faults of hydraulic pump, a residual error generator is designed based on the IG-SVM. This residual error generator is firstly trained using data from normal state. Then, it can be used for fault clustering by analysis of the residual error. Its feasibility and efficiency has also been validated via a plunger pump test-bed.