Improvement of engine performance using an optimization procedure

This paper presents a simple and effective approach to improve engine performance of a racing car with special requirements. Attention is focused onoptimal design of the intake system, using a gradient-based approximation method of mathematical programming. Since optimization relies on accurate numerical analysis of engine processes, the sub-models and parameters needed in the analysis software are carefully determined by experiment. Subsequently,the influence of different design parameters of intake and exhaust systems on engine performance is investigated numerically. The most influencing parameters are selected to be the design variables in the optimization process. In order to improve engine power at several engine speeds, two different forms of the optimal design problem are proposed, solved, and compared as a means to identify the most appropriate one. Since the analysis software is a black-box program, the optimization procedure is implemented by employing the optimization software as a master (driver) program while the analysis software acts as the slave program. The data exchange between these programs is established by XML data files and suitable wrapper programs. The results obtained confirm the usefulness of the approach presented