To reveal the principles of human thermal responses and find out the effects of body parts on whole-body thermal sensation,through a subjective survey,experimental investigations on human responses are carried out when a single body part is thermally stimulated.Cooling airflow is sent to seven body parts,respectively.Totally 94 samples are tested.To eliminate the obvious multicollinearity of thermal sensation among different body parts,the principal component regression approach is adopted to obtain the principal components for the body parts under different experimental conditions.Through regression and analysis of principal components,the weighting factors of the seven body parts are obtained.A predictive model on whole-body thermal sensation is obtained based on the weighting factors.The results show that the different characteristics of trunk and limbs are clearly seen.The weighting factors of local thermal sensation are integrated values,and there is little difference among values of different body parts.
瞬变热环境下,热反应与环境参数是紧密联系的。本文基于最小二乘支持向量机LS-SVM(LeastSquares Support Vector Machine)理论,结合瞬变热环境下受试者的投票实验数据,试图将这种关系量化,以达到对瞬变热环境下整体热感觉预测的目的。通过样本测试对预测模型的验证结果表明,向冷环境过渡和向热环境过渡中误差﹤0.3的样本比例都达到了90%以上,预测结果较理想,并且预测精度优于BP神经网络所建立的模型。另外,考虑到热感觉的模糊性以及个体化差异造成的影响,还给出了测试样本集在置信水平为95%时的置信区间,能对测试样本的变化区间作出较为准确的判断。
