Piecewise switched Kalman filtering for sensor fault diagnosis in a DC/DC power converter

With the extensive use of power electronics in modern HEVs comes the need for efficient methods of condition monitoring and fault diagnosis to ensure the reliability of the electrical power system. This paper addresses sensor faults in a DC/DC power converter system interfacing the main energy storage unit and the AC drive in a hybrid electric vehicle. A residual-based fault diagnosis system that detects and localizes sensor faults in the examined system is designed. The proposed scheme extends the classic Kalman-based detection filter commonly used in linear time-invariant systems to a piecewise switched filter suitable for switched linear systems as those encountered in power electronics. Residuals of measured signals are generated by employing a bank of switched Kalman filters on a stochastic model of the converter. The generalized likelihood ratio test is then used as a statistical change detection method to evaluate the residuals and generate a detection function which is compared with a decision threshold to detect the occurrence of a sensor fault.