Artificial Neural Networks for Internal Combustion Engine Performance and Emission Analysis

Abstract

This paper presents an analytical work for better design system that contributes to the reduction of fuel consumption and emission for vehicle performance. The main technological issue on engines today is to comply with emission standards with cost-effective measures in order to keep the engine price still attractive to customer. The experimental research of engine performance are time consuming and quite expensive. The purpose of this work is to optimize engine performance using artificial neural networks (ANN). Back propagation neural network was used to optimize prediction model performance. The paper analyzed data from various experimental tests in which different engine operating parameters are measured. The paper highlights the framework and suitable model of ANN to optimize several operating parameters of the engine. The optimization includes a range of standards engine-operating conditions, with specified limits in emissions. General TermsArtificial Neural Networks, Engine Operation, ANN approaches to management of Engine operations, ANN algorithms, architecture

Application of Artificial Neural Network for Internal Combustion Engines

“…The experimental test consists of an aluminum cylinder head of a four-cylinder spark-ignition engine. Figure (1) shows four AE sensors on the cylinder head so that compressed air enters the combustion chamber through the spark plug hole so that the airflow is measured with an accuracy of 1 liter/minute Three types of valve defects were simulated by introducing compressed air from the spark plug hole and measuring the leakage that occurs due to the faulty valve. These defects were as follows:…”

“…The automotive industry is facing a difficult time, as the shift from internal combustion engines to modern electrical technologies requires huge investments, so the search for alternative fuels is a solution to improve operating standards. Therefore, IC engines will serve as a means of transportation during the coming years (1) Diagnosing faults and monitoring the condition of the mechanical system are necessary to avoid faults. An internal combustion engine is a rotating machine that is operated in different conditions for different needs.…”

A Composition-Based Model to Predict and Optimize Biodiesel-Fuelled Engine Characteristics Using Artificial Neural Networks and Genetic Algorithms

“…The available research works in the existing literature − concerning the utility of ANN models to predict biodiesel properties and engine characteristics could be listed under two categories as follows: one that predicts the engine characteristics based on the biodiesel blend percent and the other that predicts the properties of biodiesels based on their composition. There have been no ventures so far to predict the engine characteristics based on biodiesel composition which is of utmost importance owing to larger variations in biodiesel composition.…”

“…For this purpose, quasi-Newton back-propagation, an LM learning algorithm, and a scaled conjugate gradient learning algorithm are used in the network structure. In the study conducted by Garg et al 9 the engine operating parameters in terms of speed, load, and injection pressure are used as inputs for ANN models. The algorithms used in this study include gradient descent and Levenberg−Marquardt with feed-forward network and radialbasis function network architecture.…”

The Application of Neural Networks for Prediction of Concentration of Harmful Components in the Exhaust Gases of Diesel Engines

“…The compression ratio is common for three studies [14][15][16]. The fuel injection pressure is associated with more research [17][18][19][20], as well as the injection timing [14,16,17,19,21]. In order to reduce harmful exhaust emission products, a mixture of diesel and biodiesel is used as fuel [10,15,16,[21][22][23][24].…”