Research On Control Strategy Of Human Thermal Comfort State Based On PPG Signal

Thermal comfort is a "psychological state",which is a subjective evaluation of the satisfaction of the thermal environment.The human body's feedback to the thermal environment is mainly optimized by the nervous system,and the hypothalamus plays the role of "controller".Hypothalamus has different thermal regulation mechanisms under different thermal comfort states.It is very important to study the physiological parameters of the human body in different thermal environments for understanding the thermal comfort state of the human body,the thermal comfort control strategy develops in line with the real needs of the human body.It can not only improve the satisfaction of the human body to the thermal environment,but also optimize the consumption of building energy and reduce carbon emissions,which has extremely important practical significance.The existing control strategy for thermal comfort environment is mainly on the basis of the external environment parameters such as temperature and humidity environment,it's common that the thermal comfort is consistent with the environmental temperature,but this method cannot meet the demand of the real human body,and the physiological parameters of thermal comfort study can reflect the thermal comfort.The existing studies have proved that bioelectrical signals can be used for thermal comfort analysis of the human body,such as EEG,ECG,skin conductance,etc.However,the acquisition of bioelectrical signals often requires certain experimental conditions,so it is difficult to be applied in practice.PPG has been widely embedded in wearable devices,so this work takes PPG as the main research object.In this work,a multi-channel PPG acquisition system including high-precision constant-current source,phase-locked amplifier,analog-to-digital conversion circuit design,etc.is designed,Secondly,the features in time domain,frequency domain,time frequency domain and nonlinear characteristics are feed into the thermal comfort model of human body for machine learning.Experimental result shows that the recognition rate of the thermal comfort state of human body thermal comfort model is 97.1%.As for building energy experiment,ABM model simulates the human agent model,voting system,HVAC system model.Python mesa library is used to build ABM model and Energyplus platform to realize dynamic simulation of building energy.This work research fully considers the influence of human factors,the number of hot zones and the control strategy of thermal comfort on building energy consumption.The results show that human thermal comfort preference is the most critical factor for building energy efficiency.In mainly experiment,the control model of HVAC system based on thermal comfort drive has the potential of saving energy compared with the benchmark temperature of 21.5?.Of course,under some extreme experimental conditions,buildings actually consume more energy.