Comparison of Modeling Techniques to Be Used As An Objective Function in Optimization Techniques

Abstract

In recent years, the stringent environmental regulations from the government have made the automotive industry to find alternative ways to decrease the emission level in the internal combustion engines. Compressed natural gas (CNG) is a promising alternative fuel, which decreases emissions, in dual fuel mode; the thermal efficiency is slightly less than that of a diesel engine. Optimizing the input parameters like load, CNG flow rate, and compression ratio, such that the output parameters like efficiency are high and emissions are lower, is very important.Optimization techniques are being widely used in this efforts. In this paper, an attempt has been made to compare different modeling techniques like Adaptive neuro fuzzy inference system (ANFIS), radial basis function extreme learning machine (RBF-ELM), and response surface methodology (RSM), to evaluate them and to identify the best possible alternative as an objective function for use in optimization techniques.