The multi-objective optimization of combustion chamber of DI diesel engine by NLPQL algorithm

The optimization procedure used in the present investigation is based on NLPQL algorithms and focused on the reduction of soot and NOx simultaneously while maintaining a reasonable power performance. The parameters to be optimized are related to the injection and geometric features of the combustion chambers, which greatly influences the process of air/fuel mixing, combustion and emissions formation. lizing ESE Module of AVL FIRE software, a model of combustion system which can simulate the compression and expansion process of engine working cycle is established. From the perspective of reducing soot emissions, using orthogonal design method, we can achieve a set of optimizing parameters of combustion chamber quickly. Basing on the best design of the DOE, optimization with NLPQL method was done applying different variable range and objective definitions (weighing factors Soot/NOx/IMEP = 5/3/1). ing the transient flow field contours before and after optimization, the formation of emissions can be described accurately and comprehensively. Furthermore, the numerical results from simulations are compared with testing carried out on an 8.9 L Cummins Diesel engine. For most investigated operating conditions and Europe Steady Cycle, simulated results of emissions show a good consistency with testing data.