Optimization of the parameters of an internal combustion engine using a neural network

Abstract

© 2018, Institute of Advanced Scientific Research, Inc. All rights reserved. The article describes the possibilities of an artificial neural network for optimizing the design and adjusting parameters of a diesel engine in order to reduce the amount of harmful emissions with exhaust gases, describes the methodology for constructing artificial neural networks and their main types. In this paper, we propose an optimization technique based on the teaching of ANN with the use of both experimental and calculated data obtained using the mathematical model of the working process on the basis of a system of differential equations for the energy and mass balances of the working fluid, which significantly reduces the time and material resources development of ICE. Also, a complex mathematical model allows calculations to be made to substantiate the changes introduced to the design of engines for the purpose of their refinement and modernization in order to increase power, economic characteristics and reduce emissions of emissions of explosives (EE) with exhaust gases (EG). The mathematical model of the working process is based on the system of differential equations of the energy balance of the working fluid, which includes dynamic models of the state of the working fluid in the combustion chamber (1), in the intake and exhaust manifolds, the necessary parameters of the turbocharger, cylinder piston group, gas distribution mechanism and fuel equipment.