Economic Design of Integrated SPC and EPC Using Neural Network Approach

• Main Authors: Tzu-yuan Huang, 黃資元
• Other Authors: Chau-Chen Torng
• Format: Others
• Language: zh-TW
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/89985280982419733223

碩士 === 國立雲林科技大學 === 工業工程與管理研究所碩士班 === 93 === The purposes of process control are improving quality and reducing cost, it’s an important subject to balance between these factors. In order to reduce variance for improving quality, combining the Statistical Process Control and Engineering Process Control become for expansionary topic. Many researches prove that integrating SPC and EPC is better than using alone. Some researches using neural network approach in this topic, and proving that have good performance, but less than probing into the cost due to EPC adjustment and the problem of over control. This research considers combining bounded adjustment using Taguchi loss function for EPC method with Neural Network controller. The purpose is to avoid the problem of over control to result in a load of cost due to EPC adjustment. Furthermore, in this study to verify this model is suitable for use in different levels of disturbance, different EPC controller and different adjustment cost. By the result could see that the bounded adjustment using Taguchi method in EPC control has good performance in this study, combining this model for Neural Network controller still has good performance and prove the rationality using Neural Network approach in process control.