SENSOR FAULT DETECTION BY SPARSITY OPTIMIZATION

Measurement faults in control systems may result in perma-nent damages to the system components. Therefore, sensor validation is essential before the measurements are used for any system reconfiguration. In this paper, a statistical ap-proach for sensor fault identification is proposed. Specifically, the potential sensor fault is assumed to be an additive bias term in the measurement model. The problem of fault iden-tification is formulated as a least-squares optimization prob-lem with an ℓ1 penalty on the bias term. An algorithm is further introduced to determine the regularization parameter automatically. Experimental results show that the proposed method can accurately detect multiple sensor failures from noisy measurements. Index Terms- sensor validation, bias detection, analytical redundancy, ℓ1 regularization selection 1