Skip to Main Content
A model-free hybrid fault diagnosis technique is proposed to improve the performance of fault detection and isolation. This is a model-free hybrid method, which combines a parity equation approach with a multi-resolution signal decomposition by using discrete wavelet transform. There exists a required minimum number of measurements to isolate the faulty sensor in a given sensor configuration. The proposed technique can isolate the faulty sensor even when the above condition is not satisfied. To verify the effectiveness of the proposed fault diagnosis method, numerical simulations are performed for a skew-configured inertial sensor system. Experiments using the hardware in a loop simulation system are also performed.