Zonotopic Kalman filtering based fault diagnosis algorithm for linear system with state constraints

A novel fault diagnosis method based on zonotopic Kalman filter for linear system with state constraints is proposed. The augmented system is constructed by extending the state constraints to the system output vector and the fault to the state vector, respectively. Then, a zonotopic Kalman filter is designed to obtain the interval estimation. When a fault occurs, the fault estimate can be directly found, vice versa. The effectiveness and accuracy of this algorithm are demonstrated through the sensor fault diagnosis case of the electrothermal coupled model of Li-ion battery.