基于改进GSO-BPNN的变风量空调冷却水控制  

Optimization Control of Cooling Water System of VAV Air Conditioning Based on Improved GSO-BPNN Algorithm

作　　者：杨世忠 逄铄 李善伟 YANG Shi-zhong;PANG Shuo;LI Shan-wei(College of Information and Control Engineering,Qingdao University of Technology,Qingdao Shandong 266520,China)

机构地区：[1]青岛理工大学信息与控制工程学院,山东青岛266520

出　　处：《计算机仿真》2022年第6期319-324,共6页Computer Simulation

摘　　要：为了改善变风量空调冷却水系统控制中存在的能耗和稳定性存在缺陷的情况,提出一种基于BP神经网络和改进萤火虫算法的综合优化控制算法。首先对变风量空调的冷却水系统进行建模,然后利用改进后的萤火虫算法对BP神经网络进行优化,优化对象为初始权值和阈值,从而加快对目标函数的求值,得到最优的结果。最后利用改进后的BP神经网络优化冷却水模型中的变频PID控制器的参数。通过进行仿真对比以后可知,对比于基本萤火虫算法,改进萤火虫算法优化后的BP神经网络有着更好的权值和阈值,系统稳定性得到加强,同时具有良好的鲁棒性,在符合室内负荷的条件下,也有较为明显的节能效果。In order to improve the energy consumption and stability defects in VAV cooling water system control, a comprehensive optimal control algorithm based on BP neural network and improved Glowworm swarm optimization is proposed in this paper. Firstly, the cooling water system of VAV was modeled. Then the improved Glowworm swarm optimization was used to optimize the BP neural network, and the optimized objects were initial weights and thresholds, so as to accelerate the evaluation of the objective function and obtain the optimal result. Finally, the modified BP neural network was used to optimize the parameters of the variable flow PID controller in the cooling water model. Through the simulation comparison, it can be seen that compared with the basic Glowworm swarm optimization, the BP neural network optimized by the improved Glowworm swarm optimization has better weights and thresholds, enhanced system stability enhanced, and stronger robustness. Under the condition of meeting the indoor load, it also has a relatively obvious energy-saving effect.

关 键 词：变风量空调 冷却水系统 神经网络 改进萤火虫优化 比例积分微分控制 鲁棒性 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]