A genetic algorithm for determining cylinder pressure in internal combustion engines

In this paper, a genetic-algori thin-(GA-) based method has been developed to determine the instantaneous pressure in cylinders of internal combustion engines. Some parameters associated with a pressure curve in each cylinder are optimized from two validated submodels which reproduce the engine behavior using a genetic algorithm. In this process, the profile of both measured and modeled instantaneous angular speeds of the engine are compared. According to a typical design of a genetic algorithm, different operators, such as the selection process, crossover, and mutation, have been evaluated and applied over theoretical data. The best results define the adopted GA structure, and it is applied to three different engines: single-cylinder diesel engine(DE), three-cylinder spark ignition engine, and a 16-cylinder vee power plant diesel engine. This algorithm makes it possible to estimate a good quality pressure profile in cylinders for any kind of engine (operation basis and the number and disposition of cylinders), using indirect measurements and low-cost equipment.