文献类型：期刊文献

中文题名：基于Alopex的粒子群算法及其在软测量上的应用

英文题名：Particle swarm optimization based on Alopex and its application in soft sensor

作者：奚玮君[1];李绍军[1];钱锋[1]

机构：[1]华东理工大学自动化研究所,上海200237

年份：2006

卷号：23

期号：11

起止页码：1045

中文期刊名：计算机与应用化学

外文期刊名：Computers and Applied Chemistry

收录：CSTPCD;;北大核心:【北大核心2004】；CSCD:【CSCD2011_2012】；

基金：国家973计划(2002CB3122000)上海科委科技攻关项目(04DZ11010)"十五"国家高技术研究发展(863)计划项目(2003AA412010);上海市优秀学科带头人计划的资助

语种：中文

中文关键词：粒子群优化算法;Alopex算法;软测量;神经网络

外文关键词：particle swarm optimization, Alopex, soft sensor, neural network

摘要：针对粒子群优化算法(PSO)容易陷入局部最优值的缺点,提出一种利用Alopex算法(algorithms of pattern extraction)和PSO算法结合的新算法,该算法将Alopex的步长取法加以改变,并加入了随机噪声,具有很强的全局搜索能力和很高的搜索效率。最终将算法应用于BP网络的权重和偏置量的优化计算,完成软测量的建模。结果表明改进型粒子群算法搜索效率明显提高,应用于软测量建模能提高模型的精确度,减少预测误差。
This paper proposes a new PSO（ particle swarm optimization） which aims at the disadvantage of the original PSO that is easily trapped in the local optimization. It combines PSO with Alopex （algorithms of pattern extraction）. The new algorithm changesthe way how step length is decided and a stochastic noise is added in its iteration. These make this algorithm more effective and stronger global searching ability. Finally it is used in optimizing BP＇s weights and biases to achieve a soft sensor model to predict the acetylene concentration of the acetylene hydrogenation reactor. Results show that the soft sensor model＇s output precison is increased and predicting error is decreased greatly.