最小二乘支持向量机在瞬变热环境实验研究中的应用  

The Application of Experimental Study for LS-SVM in Transient Thermal Environment

作　　者：许红波[1] 金权[1] 李敏[1] 端木琳[1] 

机构地区：[1]大连理工大学建设工程学部建筑环境与设备工程研究所,大连116024

出　　处：《人类工效学》2013年第2期42-46,共5页Chinese Journal of Ergonomics

基　　金：国家自然科学基金资助项目(50678030);国家自然科学基金资助项目(51078052)

摘　　要：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。Thermal response and environmental parameters are closely linked in transient thermal environment.The paper attempts to quantify the relationship in order to predict the overall thermal sensation under the transition based on LS-SVM theory and the experimental data of subjects under transient thermal environment.The validation results of prediction model by testing samples show that the percentage of the sample error of less than 0.3 has reached 90%,which shows that the results are well predicted,and the prediction accuracy is better than the BP model.Moreover,given the impact of ambiguity of thermal sensation and individual differences,we also give simultaneous 95% confidence intervals for the test sample sets,which can make more accurate judgments for the change interval of test samples.

关 键 词：瞬变热环境 热感觉 最小二乘支持向量机 BP神经网络 

分 类 号：TU111.1[建筑科学—建筑理论]