A Study on the Identification and Speed Control of a Diesel Engine Using Levenberg-Marquardt Backpropagation Algorithm Neural Networks

Abstract

Diesel engine is known as nonlinear system because of its dead time due to injection delay and ignition delay. So, it is very difficult and complex to model this nonlinear system because it varies widely according to number of cylinder and RPM.

In this paper, in order to design the speed control system of a diesel engine, neural network architecture is introduced and the optimal structure of neuro emulator is determined based on the modelling of a diesel engine, trained with various backpropagation algorithms and the performance of each trained networks is compared . Also, neuro controller, the inversely trained neural network of neuro emulator, is designed for the speed control system of a diesel engine. The selective neuro controller is proposed for the sake of improvement of the neuro controller performance and by combining a PI controller with the proposed controller, the efficiency of this combination speed control system of a diesel engine is ascertained.