Sensory Recognition Of Blowing Air Based On Objective Physiological Parameters

Thermal sensation is the human body’s subjective description of the hot and cold environment,and there is a lot of subjectivity.Moreover,researchers have been working on reflecting thermal sensation changes through some kind of objective index,which is essentially to realize the evaluation and prediction of subjective feeling through some kind of objective index.The blowing sensation is a pressure-type tactile sensation caused by the flow of air,and is one of the important factors affecting the evaluation of human thermal sensation.The airflow environment will affect the human body’s perception of blowing,and the thermal sensation of the human body will also be affected by the blowing.Therefore,the blowing factor cannot be ignored in the evaluation of the thermal sensation of the indoor environment.In addition,the successful identification of pain and smell directions by electroencephalogram(EEG)technology provides a reference for the identification of wind sensation.In order to evaluate the effect of blowing on the human body and realize the effective identification of the feeling of blowing,after excluding the interference of other factors,an isothermal blowing experiment was designed.In the artificial climate laboratory,the subjects were stimulated by head-on blowing at a fixed wind speed and frequency,and the thermal sensation questionnaires of the subjects were collected.the effect of parameters.It was found that(1)isothermal blowing can significantly reduce the thermal sensation and skin temperature of indoor personnel.(2)The logarithmic average power values of the four bands of δ,θ,α and β bands in EEG under isothermal blowing are higher than those without blowing.(3)Isothermal blowing was significantly correlated with skin temperature and logarithmic mean power of θ,α and βbands in EEG.At the same time,using the EEGLAB toolbox to separate the EEG artifact through Independent Component Analysis(ICA)preprocessing,and the preprocessed EEG data is separated from the EEG signal by Fourier transform in the delta band(1-3 Hz),theta band(4-7 Hz),alpha band(8-13 Hz)and beta band(14-30 Hz)for energy feature extraction.The extracted feature vectors are then classified and identified by two machine learning algorithms,Support Vector Machines(SVM)and K-Nearest Neighbor(KNN),respectively.It is found that the SVM and KNN algorithms pass the EEG energy The recognition accuracy rate of the feature on the feeling of blowing has reached more than 88%.This paper starts from the influence of isothermal blowing on the evaluation of human thermal sensation and objective physiological parameters,and uses the energy characteristics of the four bands of EEG to realize the effective identification of the feeling of blowing through the machine learning method.the research basis.