A novel hybrid ICA-ANFIS model for prediction of manufacturing processes performance

作     者：Baseri, Hamid Belali-Owsia, Moosa 

作者机构：Amirkabir Univ Technol Dept Mech Engn Tehran Iran Babol Univ Technol Dept Mech Engn Babol Sar Iran 

出 版 物：《PROCEEDINGS OF THE INSTITUTION OF MECHANICAL ENGINEERS PART E-JOURNAL OF PROCESS MECHANICAL ENGINEERING》 (机械工程师学会会报；E辑：加工机械工程杂志)

年 卷 期：2017年第231卷第2期

页      面：181-190页

核心收录：

学科分类：08[工学] 0802[工学-机械工程] 

主　　题：Combined algorithm imperialistic competitive algorithm adaptive neuro-fuzzy inference system optimization 

摘      要：In order to reduce the manufacturing process variability and improve the production yield, it is important to predict the performance of manufacturing process. The adaptive neuro-fuzzy inference system (ANFIS) is a powerful network which can predict the output parameters of manufacturing process. However, design of ANFIS needs trial and errors to select the best structure. In this study, an imperialistic competitive algorithm has been used to determine the ANFIS architecture to reach minimum values of the prediction error. In order to evaluate the performance of this combined method, two illustrative examples of manufacturing processes have been used. Results indicated that the combined method has superiority in prediction of output, rather than previous developed ANFIS models and so it can be applied for modeling of the other manufacturing processes.