中央制冷空调冷冻水系统模糊RBF控制研究

针对中央空调冷冻水系统回水温度快速准确调节问题,提出基于模糊径向基函数(radial basis function,RBF)网络的变流量回水温度智能控制方法。首先,对冷冻水系统旁通阀门的水量开度、泵组转速等输入量,按照模糊控制理论,进行模糊化与反模糊化处理,获得归一化的输入信息向量;然后,利用能够全局寻优的RBF网络进行温度预测,不断迭代预测产生理想的预测温度;最后,当期望温度与预测迭代的温度残差小于门限值时,停止迭代,输出并记录温度,完成冷冻水系统的非线性温度控制。仿真实验表明,相比于传统反向神经(back propagation,BP)网络控制,RBF控制方法迭代次数更少且精度更高,能够提高系统的整体性能。

Fuzzy radial basis function control for central air conditioning system

Aiming at the problem of rapid and accurate adjustment of chilled water system on central airconditioning backwater temperature,an intelligent control method based on fuzzy RBF network was proposed.Firstly,the fuzzy control theory was used to deal with the amount of water and the rotational speed of the bypass valve of the chilled water system.Then,RBF network was used to complete the global optimization of temperature prediction,and then the ideal iterative prediction temperature was produced by proposed method;finally,the iteration was stopped when the desired temperature and residual temperature prediction iterative was less than the threshold value,output and record the temperature,nonlinear temperature control of chilled water system was been completed.The simulation experiments show that the RBF control method is better than the traditional BP neural network control method,the iteration number is less and precision is higher,and improved proposed control method can improve system performance.