基于生理参数的关节局部热感觉预测方法
Local thermal sensation prediction method for joints based on physiological parameters
摘要：
随着生活水平的提高,越来越多的人注重局部热舒适,尤其是关节部位的热感觉。生理参数和环境参数分别是影响人体热感觉投票(TSV)的直接内在因素和外在因素。现有基于生理参数和环境参数建立的局部热感觉模型可以实现对局部位置热感觉的预测,但对关节局部热感觉的预测精度不高。因此,本文首先基于局部皮肤温度建立了适用于关节部位的局部热感觉模型,并引入皮肤电反应(GSR)和决策树模型,以提高局部热感觉模型的预测精度。结果表明:基于局部皮肤温度、GSR构建的局部热感觉模型可以预测关节等局部敏感位置的TSV,采用的决策树方法可用于判定预测值和实际值之间偏差量的修正方向。当模型预测值和TSV实际值之间偏差在±0.5以内时,平均预测准确率在80%以上;引入GSR修正项模型的预测结果比仅采用单一局部皮肤温度构建的模型预测准确率提高了9.1%。该模型可准确预测关节局部热感觉,从而提高人体局部的热舒适性。
Abstract：
With the improvement of living standards, more and more people pay attention to local thermal comfort, especially the thermal sensation of joints. Physiological parameters and environmental parameters are the direct internal and extrinsic factors influencing human thermal sensation vote (TSV), respectively. The existing local thermal sensation models based on physiological parameters and environmental parameters can predict the local thermal sensation, but the prediction accuracy of joint local thermal sensation is not high. Therefore, in this paper, a local thermal sensation model suitable for joints is firstly established based on the local skin temperature, and the galvanic skin response (GSR) and the decision tree model are presented to improve the prediction accuracy of the local thermal sensation model. The results show that the local thermal sensation model based on local skin temperature and GSR can predict the TSV of joints and other locally sensitive locations, and the decision tree method can be used to determine the correction direction of the deviation between the predicted value and the actual value. When the deviation between the predicted value of the model and the actual value of TSV is within ±0.5, the average prediction accuracy is more than 80%, and the prediction accuracy of the model with GSR correction is 9.1% higher than that of the model constructed with only a single local skin temperature. The model can accurately predict the local thermal sensation of the joints, thereby improving the local thermal comfort of the human body.
Keywords:thermal sensation vote(TSV); local thermal sensation model; physiological parameter; galvanic skin response(GSR); decision tree model; joint