应用改进RBF神经网络的室内环境舒适度评价  

Indoor Environment Comfort Evaluation Based on Improved RBF Neural Network

作　　者：杨亮[1] 王谊 YANG Liang;WANG Yi(Department of Aeronautical Engineering,Shaanxi Polytechnic Institute,Xianyang 712000,China)

机构地区：[1]陕西工业职业技术学院航空工程学院,陕西咸阳712000

出　　处：《微型电脑应用》2021年第7期86-89,共4页Microcomputer Applications

基　　金：陕西工业职业技术学院项目(2020YKYB-043)。

摘　　要：室内环境舒适度评价研究具有重要意义,当前室内环境舒适度评价方法不能准确描述室内环境舒适度的变化规律,导致室内环境舒适度评价精度低,为了更加准确对室内环境舒适度进行评价,提出了改进RBF神经网络的室内环境舒适度评价模型。首先对室内环境舒适度评价研究进展进行分析,并设计了环境舒适度评价指标体系,然后采集环境舒适度评价指标的数据,并确定环境舒适度等级,得到环境舒适度评价数据集,最后采用改进RBF对环境舒适度评价数据集进行学习,建立环境舒适度评价模型,与其它环境舒适度评价模型的对比结果表明,改进RBF神经网络的环境舒适度评价精度要高于对比模型,能够更好描述环境舒适度变化规律,具有十分广泛的应用前景。The research of indoor environment comfort evaluation is of great significance.The current indoor environment comfort evaluation method cannot accurately describe the change law of indoor environment comfort,which may lead to the low accuracy of indoor environment comfort evaluation.In order to evaluate indoor environment comfort more accurately,an improved RBF neural network model is proposed for indoor environment comfort evaluation.Firstly,the research progress of indoor environment comfort evaluation is analyzed,and an index system of environment comfort evaluation is designed.Then,the data of environment comfort evaluation index are collected,the level of environment comfort is determined,and the data set of environment comfort evaluation is obtained.Finally,the improved RBF is used to study the data set of environment comfort evaluation,and an environment comfort evaluation model is established.Compared with other environmental comfort evaluation models,the results show that the accuracy of the improved RBF neural network is higher than that of those comparison models,it can better describe the change law of environmental comfort and has a very wide application prospect.

关 键 词：室内环境 数据集合 评价效果 神经网络 对比实验 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]