Review of Artificial Neural Nertwork Applications in Engine Research Lukman Shalahuddin, Rizqon Fajar, Sahid Bismantoko
Badan Pengkajian dan Penerapan Teknologi (BPPT)
Abstract
Considering the environmental and energy concerns, the need for finding an alternative fuel is increasing greatly. Emission legislation has become progressively tighter, making the development of new or modified internal combustion engines very challenging. Multiple test combinations in engine experiments required for obtaining optimum results are usually time-consuming and costly.
The potential of artificial intelligence application for prediction of internal combustion engine performance has been studied by various researchers. Many have utilized the advance of artificial neural networks (ANN), in order to predict either engine performance, exhaust emissions, optimum control, or optimum blend of alternative fuels.
The methodology used were initially to obtain data from experimental engine tests. The fuel compositions or the engine operating conditions were then varied. An optimum algorithm and architecture of the network were developed, trained, and tested to compare the predicted values with the experimental values.
This paper reviews and discussed major works by various authors in this regard. The purpose is to give an overview of the various applications of ANN for internal combustion engines. Applications include engine performance, exhaust emissions, optimum control, or optimum blend of alternative fuels are discussed. Thus, the benefits of ANN, and future possibilities are assessed.
