K-nearest neighbour technique for the effective prediction of refrigeration parameter compatible for automobile

Abstract

Manufacturing simulation is an encouraging research area in resent decade. Creation or development of better simulation tool or technique is one of the ma-jor intension in manufacturing simulation. In resent research most of the manu-facturing processes are simulated successfully. But some processes are not yet simulated effectively, especially automatic air conditioning (AC) system or re-frigeration system. The automatic AC system for the passenger vehicle are not yet effectively simulated. Hence in this paper a machine learning technique is adopt-ed for the effective prediction of parameter of automatic AC system. The pro-posed system uses k-nearest neighbour technique for the prediction of parameter will less error and high accuracy. The proposed system is implemented using MATLAB and its performance is compared with the support vector machine and ANN in terms of mean square error and accuracy. The proposed technique out-performs the conventional technique and suggest that the k-nearest neighbour become the most suitable technique for the modelling and performance analysis of automatic AC system.