Sensor Multifault Diagnosis With Improved Support Vector Machines

In this paper, two multifault diagnosis methods based on improved support vector machine (SVM) are proposed for sensor fault detection and identification respectively. First, online sparse least squares support vector machine (OS-LSSVM) is utilized to detect and predict sensor faults. Then, a method which combines the SVM and error-correcting output codes (ECOC) called ECOC-SVM is proposed to solve the sensor fault feature extraction and online identification problem. We regard nonlinear transformation as the input of classifiers to enhance the separability of initial characteristics. ECOC-SVM is utilized to classify the fault states. Some typical faults are investigated and the experimental results indicate that ECOC-SVM has high identification accuracy and can be implemented in real-time to meet the requirements of online fault identification. This method can also be extended to solve other related problems. Note to Practitioners-This paper was motivated by the problem of sensor fault detection and recognition which, to a great extent, can reflect the overall performance of the complex information system. The identification part can be applied to other multiclassification problems. Existing approaches to detect sensor fault with least squares support vector machine (LSSVM) suffer from nonsparse problem, which blocks the prediction speed, especially in industrial operations where high efficiency is needed. In this paper, by changing the threshold, we can adjust the number of the vector groups, decrease the number of support vectors, reduce the dimension size and improve the operation speed. Meanwhile, we proposed a novel method to sensor fault identification which bridges multiclass problems and binary-class classifiers with Error Correcting Output Codes (ECOC) and Support Vector Machine (SVM). Compared with other methods, ECOC-SVM can lead to higher accuracy, while it needs slightly more time than one-versus-one method. In future research, we will further improve the efficiency of ECOC-SVM to achieve better real-time performance.