Artificial Intelligence-Machine Learning Algorithms for the Simulation of Combustion Thermal Analysis

Abstract

With the advances in big-data science and the availability of emerging high-end computing powers such as the GPUs and cloud computing infrastructures, Artificial intelligence (AI)-Machine learning (ML)/Deep learning (DL) methods are increasingly being used for predictive analysis and design analysis of various complex engineering problems. ML/DL models are developed for various classes of problems involving large nonlinear parameters from energy usage patterns in power distributions including renewable power sources; building energy consumption; and management to improve energy consumption and efficiency to optimized design and predictive analysis of thermal heat management systems in electronics and battery storage systems. The objective of this article is to explore the development of an AI/ML model for studying predictive and design analysis of an internal combustion engine as an alternative to high fidelity computation fluid dynamic model. ML-artificial neural network (ANN) algorithm is considered for characterizing and predicting the combustion process through the use of historical performance data produced by a computational model involving complex multiphysics flow dynamics, heat transfer, and chemical kinetics. A number of key ANN parameters such as the number of hidden layers, number of neurons, and activation functions are numerically experimented to achieve an acceptable level of prediction accuracy through minimizing the training and validation losses.