Automatic Fault Detection and Diagnosis for Sensor Based on KPCA

Automatic fault detection and diagnosis for sensor is necessary, which affects the performance of the control system seriously. The KPCA effectively captures the nonlinear relationship of the process variables, which computes principal component in high-dimensional feature space by means of integral operators and nonlinear kernel functions. The KPCA method was used in diagnosing for four familiar sensor faults. At first it detected fault by Q statistics, at second it identified fault by T² contribution percent variation. The experiment showed the KPCA method had good performance in fault detection and diagnosis.