Research On Indoor Environment Quality Evaluation And Thermal Comfort Prediction Model For Office Buildings

With the high-performance development of green buildings,Indoor Environment Quality(IEQ)evaluation during the operation of green buildings has become an important research work,among which indoor thermal environment affects both users’ comfort and building energy consumption,and has become a hot research topic.In recent years,the evaluation of users’ thermal comfort mostly requires users to wear sensors(thermocouples,smart bracelets,etc.),which will cause certain interference and influence to users.In order to avoid the interference to users due to indoor thermal environment evaluation,this paper studies a method to predict thermal comfort by users’ cold/hot actions.First,four green office buildings in hot-summer,cold-winter and cold regions are used as objects,and subjective and objective testing methods such as questionnaires and environmental monitoring are used to determine the weights of four subscales in IEQ,such as thermal environment,air quality,light environment and sound environment,through multiple linear regression;then,an office in Guangzhou University is used as an example,and the experiments record and test the user’s cold/heat action,indoor temperature and humidity,outdoor temperature and humidity,indoor black-bulb temperature,clothing thermal resistance,user information for six groups of parameters,based on the six groups of experimentally tested parameters,the machine learning algorithm is used to train the user’s thermal preference prediction model and obtain the user’s thermal comfort prediction method based on user’s cold/hot action;finally,the causes of thermal discomfort are discussed through the analysis of typical data of individual user’s thermal discomfort.The main findings of this paper include:(1)For office buildings such as hot summer and cold winter,cold regions,etc.,the weight of building thermal environment is 0.30,which is the largest weight among the four subcomponents of IEQ.(2)The effects of six groups of parameters,such as user’s cold/hot action,indoor temperature and humidity,outdoor temperature and humidity,indoor black-bulb temperature,clothing thermal resistance,and user information,on user’s thermal comfort were obtained,and it was found that user’s cold/hot action had the greatest effect on the accuracy of the thermal preference prediction model,and its accuracy was 0.81 when predicting thermal comfort with this parameter;The accuracy can be improved to 0.85 when predicting users’ thermal preferences by using three sets of input parameters such as cold/heat action,personal information,and black-bulb temperature.(3)Through the analysis of typical data of individual user’s thermal discomfort,it is found that local heat sources(server cooling,etc.)and solar radiation through external windows are the main causes of users’ discomfort.This paper investigates a method to predict thermal comfort by user’s cooling/heating action,which can provide reference for user’s thermal comfort evaluation and building indoor thermal environment regulation.