Indoor Thermal Response Of Human Body In Uniform Environment(Warm Conditions)

With the development of the national economy and society, higher quality of indoor environment is required. How to build healthy and comfort indoor thermal environments has become a key question in the research of building, which is based on the research of the thermal response of human body. Thermal response of human body includes physiological response and thermal sensation; according to the heat transfer theory it can be distinguished by thermal response under steady state and non- steady state. By literature review, there are few studies on thermal response especailly under non-steady state. In this context, theoretical and experimental research on thermal response of human body in uniform environments was carried out under warm conditions.A large amount of human experiments under steady or non-steady state were carried out. There are five series of experiments including 21 typical thermal conditions, 400 subjects and 760 hours of exposure. By physiological test and questionnaire in the experiments, the characteristic of subjects’ physiological and physiological response to typical warm environment were acquired. Mathematical models(Human Thermal Physiological Model and Thermal Sensation Model) are used to predict the physiological response and thermal sensation of occupants in thermal environments.Considering the situation that few studies on the evaluation method for thermal response models, this thesis summarizes the existing evaluation methods and critically analyses the flaws. A method for the evaluating the accuracy of the population based models is proposed, which is used to evaluate thermal response models. Based on the theory of statistical inference, agreement analysis and error analysis, two key questions in model evaluation study namely: i) ‘How to validate the prediction accuracy of models?’ ii) ‘How to compare the prediction accuracy of models’ can be properly solved by using this new method. A framework of the evaluation process for thermal response models has been proposed, which validates a model’s accuracy both from the statistical and empirical aspects. Five steps are proposed in the framework in the new method as: i) Define the null hypothesis and alternative hypothesis; ii) Define the confidence intervals for population means; iii) Validate the model’s accuracy by comparing model predictions with the confidence intervals of the population means; iv) Validate the model’s accuracy by analyzing the agreement between the model prediction and the sample mean; v) Compare the accuracy of models based on accuracy level and RMSE. For validation of thermal response models, three levels of accuracy are proposed as: Ⅰ—statistically accurate; Ⅱ—empirically accurate; Ⅲ—inaccurate. The new method provides a scientific way to validate and compare the accuracy of population based models.Based on the theory of heat transfer and physiology, a theoretical framework for human thermal physiological model is proposed; Then, empirical formula and parameters of the model are optimized by subjects’ physiological test results. A human thermal physiological model based on Chinese population is developed. The new model is simplified but with guaranteed accuracy, which was validated by experiments with 400 samples. The principle of physiological response and thermoregulation of human body is revealed in the model, which can be used to simulate subjects’ transient physiological response and predict the transient skin temperature in dynamic thermal process. This model is consisted by three part: i) the physical model, which is the abstract of the real body; ii) the controlled system, which is used to simulate the heat transfer of the body; and iii) the controlling system, which describes the thermoregulatory control mechanisms of human body. Comparing with the past models, the proposed model is optimized in three aspects involving i) physical modeling, ii) evaporation heat loss by sweating and iii) gender difference.According to physiological and psychological test results, a semi-empirical model of thermal sensation(TSm) based on skin temperature is proposed to predict occupants’ transient thermal sensation in dynamic thermal process. This model reveals the dynamic characteristic of thermal sensation, that is, in the dynamic process people’s thermal sensation is not only the function of the absolute value of parameters(eg. Temperature) but also related to the change of parameters. TSm model is consisted by two parts: static term and dynamic term. The static term describes the relationship between thermal sensation and parameters when body under steady state; while the dynamic term describes the contribution of parameters changing on thermal sensation when body under non-steady state. TSm and PMV models are evaluated by the experimental data in this research. The results show that: TSm model’s prediction is statistically accurate when body under steady state and empirically accurate(the limit of agreement is less than 0.5 ASHRAE scale) when body under non-steady state; while PMV model’s is evaluated as inaccurate since there is unacceptable bias between predictions and actual sensation. Occupant’s transient thermal sensation can well evaluated by TSm model, which can be used to evaluate the dynamic thermal process.Based on the human thermal physiological model and TSm model, a simulation software for thermal response of human body is compiling by Matlab. The input of the software includes: i) the basic information of subject(gender, age, height, weight, body fat percentage); ii) the time-dependent parameter of thermal environment and subject in the thermal process; iii) initial conditions and the variables during the simulation period. The output of the software includes the distribution of body temperature and thermal sensation of the subject at any moment. Therefore, given with the conditions of thermal environment and subject, the software can simulate subjects’ transient thermal response in dynamic thermal process. The software is practical in application, which can be directly used in teaching, research and engineering.This thesis proposes an evaluation method for population based models, a human thermal physiological model and a thermal sensation model. These models are significant for the national research on dynamic thermal comfort, which can contribute to the improvement for national evaluation standard for thermal comfort.