摘要
把混沌寻优思想引入菌群优化算法中,利用Logistic映射的遍历性、随机性及对初值的敏感性等,对当前菌群群体中的最优细菌进行混沌寻优,以预防算法"早熟"。同时,用混沌菌群算法优化BP神经网络过程,建立延迟焦化生焦高度斜率预测模型。仿真结果表明:该模型具有较高的精度和较好的泛化能力,能够实现生焦高度的实时监测。
Embedding the chaotic search into original bacterial foraging optimization was implemented,including making use of Logistic mapping’s ergodicity,stochastic property and sensitivity to the initial value to do chaotic optimization of the optimal bacteria in the current population so as to prevent algorithm’s pre-mature;meanwhile,adopting the bacterial colony chemotaxis algorithm to optimize BP neural network was carried out to establish a prediction model for the slope of coke height. The simulation result shows that,this model has higher precision and better generalization ability,and it can realize real-time monitoring of the coke height.
作者
刘剑
张凌波
王潇凌
LIU Jian;ZHANG Ling-bo;WANG Xiao-ling(College of Electrical and Information Engineering,Hunan University of Technology;MOE Key Laboratory of Advanced Control of Chemical Process and Optimization Technique,East China University of Science and Technology)
出处
《化工自动化及仪表》
CAS
2019年第10期811-815,833,共6页
Control and Instruments in Chemical Industry
关键词
软测量
焦炭塔
生焦高度
菌群算法
混沌BP神经网络
soft measurement
coke tower
coke height
bacterial colony chemotaxis algorithm
chaos BP neural network
