基于改进蚁群免疫算法的PIDNN加热炉温度控制  

Temperature control of PIDNN heating furnace based on improved ant colony immune algorithm

作　　者：周建新 黄剑雄 ZHOU Jianxin;HUANG Jianxiong(College of Electrical Engineering,North China University of Science and Technology,Tangshan 063210,Hebei,China)

机构地区：[1]华北理工大学电气工程学院,河北唐山063210

出　　处：《四川冶金》2020年第3期22-26,共5页Sichuan Metallurgy

基　　金：河北省教育厅重点项目(ZD2015059)。

摘　　要：针对蚁群算法搜索时间长、易于出现早熟、停滞等问题,引入免疫信息处理机制,该方法通过抗体之间的亲和力反映种群的多样性,同时对蚁群的状态转移规则和信息素更新规则进行改进。针对BP神经网络收敛速度慢的问题,采用改进蚁群免疫算法对PID神经网络控制器的权值进行整定。仿真结果表明,改进蚁群免疫算法在收敛路径长度和收敛速度上均比传统蚁群算法效果更佳,并且与传统PIDNN(PID Neural Network)控制器相比,改进蚁群免疫算法的PID神经网络加热炉控制具有较快的收敛速度和较小的超调量,其暂态性能和稳态性能均得到有效改善。Aiming at the problems of long search time,premature and stagnation,the immune information processing mechanism was introduced.The affinity between the antibodies reflected the diversity of the population.At the same time,the state transfer rules and pheromone update rules of ant colony were improved.To solve the problem of slow convergence of BP neural network,the weight of PID neural network controller was adjusted by improved ant colony immune algorithm.The simulation results showed that the improved ant colony immune algorithm had better convergence path length and convergence speed than the traditional ant colony algorithm,and compared with the traditional PID neural network heating furnace controller,the improved ant colony immune algorithm PID neural network controller had faster convergence speed and smaller overshoot,and the transient performance and steady-state performance of the controller were effectively improved.

关 键 词：蚁群算法 免疫信息处理机制 神经网络 加热炉 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]