摘要
为了改善火电厂直接空冷机组汽轮机背压控制超调量大,有迟延,控制不稳定,传统的单级闭环回路PID以及改进后的串级PID系统都难以获得满意控制效果的问题。采用模糊控制和神经网络相结合的控制理论,设计出一种基于模糊神经网络的PID控制器来控制直接空冷机组的背压。模糊神经网络PID控制器既具有自学习、自适应的优点,同时又可以很好地利用已有的经验知识在控制过程中。通过控制仿真试验,比较常规PID、模糊PID和模糊神经PID算法的控制效果,验证得到了模糊神经网络PID算法的可行性和在响应速度上的优越性。
The back-pressure control of direct air-cooled system in thermal power plants often has the problem of large overshoot, delay, and unstable control. The traditional PID systems are difficult to obtain satisfactory control effects. According to, fuzzy control and neural network control theory, designed fuzzy neural network PID controller, used to control the direct air-cooled unit of the back pressure. The fuzzy neural network PID controller not only has the advantages of self -learning and self-adaptation, but also can make good use of the existing experience knowledge. Through the simulation, comparing the control effects of conventional PID, fuzzy PID and fuzzy neural PID algorithm, the control curve shows the feasibility of the fuzzy neural network PID algorithm and its superiority in response speed.
作者
刘韶军
王琦
王岚
白建云
景鑫
LIU Shaojun;WANG Qi;WANG Lan;BAI Jianyun;JING Xin(Guodian Yuci Thermal Power Co.,Ltd.,Jinzhong Shanxi 030600, China;Department of Automation, Shanxi University, Taiyuan 03006, China)
出处
《自动化与仪器仪表》
2019年第4期122-125,共4页
Automation & Instrumentation
基金
山西省科技重大专项项目(MD2016-02)
山西省研究生联合培养基地人才培养项目(2018JD08)
