基于改进ABC算法的模糊Elman变频空调温度控制  

Temperature control for Inverter Air-Conditioner Based on Fuzzy Elman Network Using Improved ABC Algorithms

作　　者：陈以[1] 莫迪 陈睿星 孙俊佚雄 CHEN Yi;Mo Di;CHEN Rui-xing;SUN Jun-yi-xiong(College of electronic engineering and automation,Guilin University of Electronic Technology,Guilin 541004,China)

机构地区：[1]桂林电子科技大学电子工程与自动化学院,广西桂林541004

出　　处：《计算机仿真》2021年第7期203-208,共6页Computer Simulation

基　　金：桂林电子科技大学研究生教育创新计划资助项目(2019YCXS097);桂林电子科技大学研究生优秀学位论文培育项目(17YJPYSS30);广西区自动化虚拟仿真实验教学中心项目(C77JWS01BX01)。

摘　　要：为解决变频空调温度控制中存在超调而造成用电浪费的问题,给出了基于改进人工蜂群算法(Improved Artificial Bee Colony,IABC)的模糊Elman网络控制(Fuzzy Elman Network Control,FENC)方法。方法将应用于变频空调控制的模糊规则针对网络训练过程中易陷入局部收敛的缺陷,将模拟退火思想引入人工蜂群算法的局部搜索过程,给出IABC算法。与Elman神经网络相结合,增强对动态信息的处理能力。仿真结果表明,FENC方法明显改善了训练后期易陷入局部极小点的问题,有效提高了训练精度和控制精度,限制了控制过程中的温度超调,表明FENC方法是一种精确可行的控制方法,可以有效降低空调能耗。In order to solve the problem of power waste caused by overshoot in the temperature control of variable frequency air conditioning,a fuzzy Elman network control(fenc)method based on improved artificial bee colony(IABC)algorithm is proposed.The method combines the fuzzy control rules of inverter air-conditioner with Elman neural network to enhance the dynamic information processing ability.Aiming at the shortcoming of local convergence in network training,simulated annealing was introduced into the local search process of artificial bee colony algorithm,and the IABC algorithm was proposed.The simulation results show that the FENC can clearly improve the problem of easily falling into local minimum in the later training period,effectively improve the training accuracy and control accuracy,and limit the temperature overshoot in the control process.It shows that FENC method is accurate and feasible,which can effectively reduce the energy consumption of air conditioner.

关 键 词：人工蜂群 神经网络 变频空调 温度控制 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]