ANOMALY DETECTION IN A TWO-TANK SYSTEM USING AN INTERACTING, MULTIPLE-MODEL-BASED FAULT DETECTION METHOD

摘要

This paper presents a method of fault detection and identification (FDI) for nonlinear system that is constructed based on the interacting multiple model (IMM) algorithm. The fault diagnostic approach is formulated as a hybrid multiple model estimation scheme. The proposed approach provides an integrated framework for fault detection, diagnosis, and state estimation. The proposed method can be used with different kind of filters like linear Kalman filter, extended Kalman filter, unscented Kalman filter and particle filter with respect to the application. Its performance is illustrated for fault detection of a nonlinear two-tank system. A set of models is assumed to present the possible system behavior pattern or modes and they run in parallel as a bank of filters. Different filter is used for testing and comparing the performance fairly. Performance comparisons are made between filters using the confusion matrix and classification accuracy.