Prediction of the performance and exhaust emissions of ethanol-diesel engine using different neural network

The object of this paper is to use back-propagation (BP), Elman network, radial basis network (RBF) and generalized regression neural network (GRNN) to predict the performance and exhaust emissions of an ethanol-diesel engine. The four neural network models used three input parameters and eight output parameters. The three input parameters are percentage of ethanol, power and engine speed. The eight output parameters are brake specific fuel consumption (BSFC), effective brake specific fuel consumption (EBSFC), effective brake thermal efficiency (EBTE), exhaust gas temperature (EGT), CO, HC, NOX, and Soot. In this work, the simulation compared the capabilities of the four networks in predicting the exhaust emissions and performance of the target engine. The results show that all four networks have obtained satisfactory prediction results. These four networks can be used to predict engine exhaust and performance, reducing manpower, time and effort. In addition, according to the results of the four network assessments, RBF has the best prediction effect. The RBF is very precise and useful way to perform the prediction and model nonlinear phenomena of internal combustion engine.

Automated summary

The objective of this study is to predict the performance and exhaust emissions of an ethanol-diesel engine using multiple neural network models. The results indicate that all four networks perform well in predicting the target engine's exhaust emissions and performance, with radial basis network demonstrating the best prediction effect.