Acoustic Performance of Exhaust Muffler Based Genetic Algorithms and Artificial Neural Network

Abstract

 The noise level was one of the important indicators as a measure of the quality and performance of the diesel engine.Exhaust noise in diesel engines machine accounted for an important proportion of installed performance exhaust muffler and it was an effective way to control exhaust noise. This article using orthogonal test program for the muffler structure parameters as input to the sound pressure level and diesel fuel each output artificial neural network (BP network) learning sample. Matlab artificial neural network toolbox to complete the training of the network, and better noise performance and fuel consumption rate performance muffler internal structure parameters combination was obtained through genetic algorithm gifted collaborative validation of artificial neural networks and genetic algorithms to optimize application exhaust muffler design is entirely feasible.