神经PID在地源热泵空调系统中的应用

针对地源热泵空调系统被控对象具有大滞后、慢时变、非线性特点及不确定干扰因素多的实际情况,结合常规PID和神经网络的优点,提出了一种基于BP神经网络的PID控制器设计方法,并应用于该空调系统中。仿真结果表明,该方法可以显著改善系统的动态性能和控制精度,实现了PID控制参数的在线动态调整,取得了良好的控制效果。

Aiming at reality that the controlled object in groundsource heat pump air conditioning system has large inertia,pure lag and nonlinear characteristic,and that the number of the uncertain disturbance factor is large,combining the advantage of the traditional PID cont roller and Neural Network,an improved PID controller designing approach based on BP （Back Propagation） Neural Network are proposed and is used in the system.The simulation results demonstrate that the method significantly improves system＇s dynamic performance and control accuracy,ensures the dynamic adjustment of control pa rameters,and achieve good control effect.