Fault Tolerance Engine Air-Ratio Control using Wavelet Least-squares Support Vector Machine and Fuzzy Optimization By Iong Tong Meng Thesis Supervisor: Professor Wong Pak King Department of Electromechanical Engineering There are many variables in modern automotive engine control. Among them, air-ratio is viewed as the most important one. It relates closely to control emissions and fuel consumption. Effective air-ratio control is desirable for engine systems to maintain its best performance. However, traditional air-ratio control relies strongly on the air-ratio measurement from the lambda sensor located at the tail pipe; traditional control only assumes the lambda sensor works properly. When the sensor is warming up during cold start, or under failure, the control no longer works effectively. To address this issue, this thesis utilizes an advanced modelling technique, wavelet least-squares support vector machine (WLM), to build a supplementary air-ratio model. With the prediction from the supplementary model, a limited air-ratio control performance can at least be maintained even the lambda sensor is not working. Such strategy is realized as fault tolerance control in this thesis. A fuzzy optimization is employed in this thesis to select the optimal control signal based on the predicted air-ratio from the supplementary air-ratio model. In order to verify the effectiveness of the proposed fault tolerance air-ratio control strategy, a model predictive controller, as a well-known technique for air-ratio control, is constructed based on the WLM supplementary air-ratio model and implemented on a real engine. Experimental results show that the proposed controller can regulate the air-ratio to specific target values within a satisfactory tolerance under external disturbance and the absence of air-ratio feedback signal from the lambda sensor. This implies that the proposed fault tolerance air-ratio control is a promising scheme to maintain air-ratio performance when the lambda sensor is under failure or warming up.