一种基于数据的夏季居住环境热舒适度控制方法

摘要

A hyper-ball CMAC(HCMAC) neural network based model is built to express the relationship among indoor and outdoor environmental parameters (temperature and humidity) to predict indoor relative humidity. A novel computing experimental method is explored to determine temperature set points of a domestic air-conditioner. The thermal comfort control method can adaptively meet the inhabitants' comfort needs by HCMAC learning of environmental sample data and simple and convenient computing experiment. Simulations demonstrate that the presented HCMAC model can accurately predict the indoor air humidity, and the indoor PMV can be controlled at desired level by appropriately adjusting temperature set points, facilitating home energy efficiency and comfort.%采用超闭球CMAC(HCMAC)神经网络,建立家用空调器制冷环境下室内外温、湿度关联模型,预测室内空气的相对湿度.提出了基于PMV指标的家用空调温度设定值的计算实验方法.该舒适度控制方法只利用室内外环境参数历史数据,通过HCMAC样本学习和简便的计算实验,自适应地满足住户热舒适偏好需求.仿真结果表明,HCMAC模型能够较准确地预测室内空气的相对湿度;通过调整温度设定值,可将PMV值调整到任意给定的范围,实现室内建筑环境的节能和舒适.