基于改进神经网络集成算法的软测量建模  

Soft sensing modeling based on improved neural network ensemble algorithm

作　　者：陈如清[1] 俞金寿[2] 

机构地区：[1]嘉兴学院机电工程学院,嘉兴314001 [2]华东理工大学自动化研究所,上海200237

出　　处：《仪器仪表学报》2008年第6期1240-1244,共5页Chinese Journal of Scientific Instrument

基　　金：浙江省教育厅高校科研计划(20070616)项目资助

摘　　要：为提高神经网络集成中成员网络的精度、增加成员网络间的差异度,提出一种改进的Bagging神经网络集成算法。通过分析初始样本集中样本间的欧式距离提取各子训练集,子训练集的元素在样本空间具有良好的遍历性和代表性;集成策略采用加权平均法,用粒子群优化算法求解成员网络的集成权重。几个典型回归分析型数据集的测试表明,本算法有效提高了训练样本质量,增强了集成泛化能力。最后将改进算法用于工业乙烯收率神经网络软测量建模,应用结果表明该软测量模型泛化性能好,测量精度高。To increase the diversity and accuracy of the component artificial neural networks （ANNs） in neural network ensemble, an improved Bagging neural network ensemble algorithm is proposed. The Euclidean distances between two arbitrary samples in the original training set are analyzed, then the training subsets of component ANNs are distilled from this set. The subset elements have good properties of ergodicity and representativeness in sample space. The outputs of component ANNs are combined via weighted averaging and the optimal weights are determined by particle swarm optimization. Experimental studies on four typical regression datasets show that this approach has improved the quality of training subsets. Thus, the ensemble generalization ability is improved. Finally the improved algorithm is applied to construct an ANN-based soft sensing model for real-time measurement of ethylene yield. Application results show that this model has high measurement precision as well as good generalization ability.

关 键 词：神经网络集成 BAGGING 欧式距离 粒子群优化算法 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]