摘要
介绍了基于BP神经网络PID控制器的算法,在此基础上提出了多模态神经网络-PID方法。将神经网络PID与普通PID控制方法进行比较,结果表明,该方法具有自适应能力强、过渡时间短、鲁棒性好等特点,弥补了常规PID在聚合釜控制中参数难以整定等缺点。
The algorithm of BP neural network PID controller is introduced. The multi-model neural network-PID method is put forward. The comparison of neural network PID with conventional PID control method is performed as well. The results show that this method has the characteristics of strong adaptability, short transition time and good robustness. It has made up for the shortcoming of the difficulty to set in the conventional PID eontrollment of the polymerizer.
出处
《现代化工》
CAS
CSCD
北大核心
2016年第2期165-167,169,共4页
Modern Chemical Industry
