Fault diagnosis of inertia sensor in missile control systems

Abstract

Because the magnitude of each order of oscillation signals in the yaw and roll channels of missile control systems varies with various influences, it is impossible to diagnose faults in the sensors of these channels using modelbased fault diagnosis. Based on the characteristics of the systems, a method of fault diagnosis with adjacent wavelet coefficient algorithm is given. Further, based on the difference between the characteristics of the Mallat wavelet coefficients and that of Coilflets wavelet, different algorithms are given. Then, thresholds can be properly selected and abrupt faults diagnosed. A method to diagnose disconnection or short circuit faults is also proposed according to the zone of the current and the zone of voltage when the fault occurs gradually. The combination of these two methods effectively solves fault diagnosis problems of sensors in missile control systems.