闵喜瑞1，周建新2
（华北理工大学 电气工程学院，河北 唐山063000）
关键词：IPSO算法；RBF神经网络；带钢厚度控制预测
中图分类号：TP183 文献标志码：A
摘要：提出了一种基于IPSO算法优化RBF神经网络的带钢厚度控制预测的新方法，该方法先将PSO算法中的权重和学习因子进行优化，再将改进后的新粒子群算法应用于RBF神经网络的参数确定中，实现了RBF神经网络隐含层高斯函数的中心值和宽度向量以及隐含层与输出层之间权值的优化，改善了RBF神经网络的预测精度。仿真结果表明，优化的RBF网络用于带钢厚度控制预测中具有可靠的精度和较好的收敛速度，具有广阔的应用推广前景。
A new method of the steel strip thickness controlforecasting based on optimized RBF neural network using IPSO algorithm
MIN Xi-rui1，ZHOU Jian-xin2
（1.College of ElectricalEngineering,North China University Of Science and Technology,Tangshan063000,China；2.Collegeof Electrical Engineering,North China University Of Science andTechnology,Tangshan 063000,China）
Key words:IPSO algorithm，RBF neural network，steel strip thickness control forecasting
Abstract:A new method of the thickness controlforecasting of steel strip was put forward,which was based on the optimized RBFneural network using IPSO algorithm.First,the weight functions and the learningfactors of PSO algorithm were optimized.And then, the improved IPSO algorithmwas applied to optimize the positions of data centers,width vectors and weightfunctions of RBF neural network.The prediction precision of RBF neural networkwas improved. The simulation results show that the improved RBF neural networksapplied inthe thickness controlforecasting of the steel strip are qualified with reliable accuracy and betterconvergence rate. This method possesses great practical value and futureprospects.