Control oriented engine modelling and engine multi-objective optimal feedback control

With increasing number of degrees of freedom for engine operations, traditional mapping based engine calibration methods are reaching their limits. To take their place, automated, multi-objective engine optimisation approaches are desirable. In this thesis, the author presents a model-based multi-objective engine optimisation algorithm using the Strength Pareto Evolutionary Algorithm (SPEA2). The author demonstrates the performance of his approach using a model of a Homogenous Charge Compression Ignition (HCCI) and Spark Ignition (SI) engine. The HCCI combustion model has been developed based on the Apparent Fuel Burning Rate formulations (AFBR) method. The model has been validated from 1500 rpm to 2250 rpm with different engine loads. For the experiments reported here, the calibration objective is the optimisation of valve timing (IVO, EVC) and λ with respect to indicated specific fuel consumption (ISFC) and indicated specific hydrocarbon (ISHC) for HCCI cases, and optimise throttle position, spark timing, injection timing, IVO and EVC for ISHC, ISPMM and ISPMN for SI cases, while achieving a predefined power output at a given engine speed. The method is able to find optimal engine parameters with good accuracy and calculation speed. The initial model calibration requires 90 and 108 engine test bed experiments for SI and HCCI cases respectively, after that the run time for any set of operating conditions is around 3hours and 20mins