Prediction of the overall sound level of four-stroke diesel engine with biofuel and the help of back-propagation error algorithm

MF399 tractor, is one of the common tractors used in agriculture in Iran. Due to the reduction of fossil fuels and environmental considerations, these days, tendency to use renewable fuels such as biodiesel, bioethanol and biomethanol in compound with diesel fuel has increased. Given the importance of noise and its effect on the driver's mental and physical health at work, the sound volume resulting from the use of bioethanol fuel in a tractor engine was examined and compared with the noise of diesel fuel. In this study, artificial neural networks are used to predict the noise level of MF399 tractor. Test site was prepared exactly according to international standards and signals emitted from the vehicle were measured, then the overall sound level values were obtained in the frequency domain. The results showed that the multi-layer perceptron network with back propagation error algorithm, two hidden layer, three neurons in the first hidden layer and two neurons in the second hidden layer to predict the overall system noise of vehicle at driver's ear position is Influenced by engine speed input variables and different combinations of proper fuel.

Keywords:

Noise , Diesel Engines , Artificial Neural Networks , Multi-Layer Perceptron , Back Propagation (Error) Algorithm