Effect of kernel function in support vector machine for the fault diagnosis of pump

Pumps are widely used in a variety of applications. defects and breakdown of these pumps will result in significant economic loss. therefore,these must be under continuous observation. in various applications,the role of pump is decisive and condition monitoring is crucial. a completely automated on-line pump condition monitoring system which can automatically inform the operator of any faults,promising reduction in maintenance cost with a greater productivity saving both time and money.this paper presents the application of support vector machine for classification using statistical features extracted from vibration signals under good and faulty conditions of a pump. effectiveness of various kernel functions of c-svc and γ-svc models are compared. the study gives some empirical guidelines for selecting an appropriate kernel in a classification problem. © school of engineering,taylor’s university.