摘要
为了解决菌群优化(BFO)算法易陷入局部最优,趋化操作中翻转方向不确定的问题,利用模拟退火(SA)算法在获得局部最优解的情况下能够以极大可能趋向于全局最优解的特点,提出模拟退火-菌群优化(SA-BFO)算法。同时,将改进后的算法用于优化RBF神经网络,建立基于甲醇净化CO2含量的软测量模型。仿真结果表明该模型具有更高的精度和准确性,对甲醇生产量的提高具有一定的贡献价值。
In order to solve the problems of the bacterial foraging optimization(BFO)algorithm easily falls into the local optimum,and the reverse direction during chemotaxis operation is uncertain,the characteristic of simulated annealing(SA)al?gorithm is used to propose the simulated annealing and bacterial foraging optimization (SA?BFO) algorithm. SA algorithm can reach the global optimal solution to the maximum extent while obtaining the local optimal solution. The improved algorithm is ap?plied to optimizing the RBF neural network,and establishment the soft measurement model for CO2 content purified by metha?nol. The simulation results show that the model has high accuracy and precision,and has a certain contribution value for greatly improving the methanol production.
出处
《现代电子技术》
北大核心
2016年第13期93-98,共6页
Modern Electronics Technique
基金
中央高校基本科研业务专项资金:上海市重点学科项目(B504)
关键词
菌群算法
模拟退火
RBF神经网络
甲醇净化
bacterial foraging optimization
simulated annealing
RBF neural network
methanol purification
