Model Based Fault Diagnosis for Multiple Input Multiple Output Systems

A state observer is a system in which we can model a real system in order to give an estimate of the internal state of the system. In the first part of the thesis, we discuss some of the design techniques and compare three different types of state observers are presented in this thesis. The considered observers include Luenberger observer, unknown input observer and sliding mode observer. The application of these observers to a Multiple Input Multiple Output (MIMO) DC servo motor model and the performance of observers is assessed. In order to evaluate the effectiveness of these schemes, the simulated results on the position of DC servomotor in terms of residuals including white noise disturbance are discussed. This thesis also presents a hybrid model based and statistical fault diagnosis system, for nonlinear three-tank model systems. The purpose of fault diagnosis is to generate and to analyze the residual to find out the fault occurrence. This fault diagnosis system includes residual generator and residual processor. The fault generator is applied with the Luenberger observer, which has its own algorithm to compute the parameters. The thesis starts with introduction and literature review, systems monitoring, fault detection, modeling of DC servo motor and then we discussed three different types of state observer systems and then shows the methods of residual generation and residual processing. The final chapter presents the simulation results, which operated by MATLAB Simulink. The fault diagnosis system successfully captured different kind of fault in three-tank model. The results prove the effectiveness of this fault diagnosis system.