摘要
针对传统的PID控制算法很难获得比较理想的控制效果的问题,提出一种基于改进型BP神经网络的PID控制算法。根据BP神经网络的结构和特点,介绍了PID控制器的结构和BP神经网络算法描述,利用最小二乘法和神经网络建立被控对象的预测数学模型,并用该模型所计算的预测输出取代预测输出的实测值,对基于BP网络的PID控制器的权值调整算法进行改进。通过实例进行仿真分析,结果表明,改进型BP神经网络PID控制器具有良好的控制效果,自适应能力和抗干扰性强,增强了系统的鲁棒性,优于常规BP神经网络P1D控制器。
Aiming at the problem that traditional PID control algorithm is difficult to get ideal control effect,a PID control algorithm based on BP neural network is proposed.According to the structure and characteristic of BP neural network,the construction of PID controller and the description of improved BP neural network algorithm were introduced at first.Then,on the basis of the least square method and neural network prediction model of controlled object,the weight adjustment algorithm of PID was improved by replacing the measured values of BP network with calculated forecast output.Through simulation examples,the results show that the improved BP neural network PID controller has good control performance,strong anti-interference and self-adjustment,,enhancing system robustness, better than conventional PID controller based on BP neural network.
出处
《控制工程》
CSCD
北大核心
2012年第S1期119-121,236,共4页
Control Engineering of China
