【作者】 曹晖； 司刚全； 张彦斌； 贾立新；

【Author】 Cao Hui, Si Gangquan, Zhang Yanbin, Jia Lixin(School of Electrical Engineering,Xi’an Jiao Tong University,Xi’an 710049,China)

【机构】 西安交通大学电气工程学院；

【摘要】 针对火电厂球磨机制粉系统具有的多变量、强耦合、非线性、大滞后及系统时变的特点,本文提出了一种基于自适应神经模糊推理系统(ANFIS)的自寻优算法,该算法通过ANFIS对寻优的搜索步长进行调节,在保证系统稳定运行的前提下提高了寻优算法搜索速度,实现了最佳工作点的动态寻优。在此基础上,结合模糊控制对制粉系统进行实时控制,使得制粉系统能够始终稳定、高效运行。仿真结果表明,本文提出的控制策略具有较高的控制品质。该控制策略已在现场成功应用,可有效降低制粉系统的运行能耗。更多还原

【Abstract】 Ball mill pulverizing system is a multi-variables,strong coupling,nonlinear and time-varying system and the paper proposes a self-optimizing algorithm based on adaptive network-based fuzzy inference system (ANFIS). The proposed algorithm adjusts the step of self-optimizing algorithm by using ANFIS. Therefore,the best operating point can be found automatically and the optimal speed can be improved. Moreover,the paper integrates the fuzzy control with self-optimizing algorithm to control ball mill pulverizing system real-time and keep it working stably and economically all along. Simulations results verify that the proposed algorithm has higher control quality. Moreover,the proposed algorithm has been put into practice successfully and the energy consumption of ball mill pulverizing system is reduced.更多还原