The Quality Prediction Model Based on Neural Network Ensembles

摘要：

　　To establish the ralationship between products” design parameters and overall quality characteristics,realize quality prediction and control,as well as to help new products” development and designing,PSO algorithm is firstly applied to optimize the weight and threshold of BP neural network.On base of ensuring better generalization ability and function fitting ability of single BP network,a qulity prediction model based on FCM- Heuristic idea neural network ensemble is established.Prediction for relationship between antenna”s 7 design parameters and its 3 quality characteristics is taken as an example to realize the proposed algorithm.The simulation result shows that the model constructed in this research is of hign prediction precision,which can well reflect the relationship between the designing variables and the optimizing object.

关键词： PSO algrithm grey incidence degree BP neural network neural network ensembles quality prediction

作者: XU Lan FANG Zhi-Geng LIU Si-feng

作者单位: School of Economics and Management,Jiangsu University of Science and Technology,Zhengjiang,212003;Sc School of Economics and Management,Nanjing University of Aeronautics and Astronautics,Nanjing,210016

会议类型: 国内会议

会议名称: 第28届全国灰色系统学术会议

会议地点: 武汉

会议语种:英文

页码: 107-113

在线出版日期: 2016-04-22（万方平台首次上网日期，不代表论文的发表时间）