Participation in a regular exercise program can improve health status and contribute to an increase in life expectancy. However, exercise accidents like dehydration, exertional heatstroke, syncope, and even sudden death exist. If these accidents can be analyzed or predicted before they happen, it will be beneficial to alleviate or avoid uncomfortable or unacceptable human disease. Therefore, an exercise thermophysiology comfort prediction model is needed. In this paper, coupling the thermal interactions among human body, clothing, and environment (HCE) as well as the human body physiological properties, a human thermophysiology regulatory model is designed to enhance the human thermophysiology simulation in the HCE system. Some important thermal and physiological performances can be simulated. According to the simulation results, a human exercise thermophysiology comfort prediction method based on fuzzy inference system is proposed. The experiment results show that there is the same prediction trend between the experiment result and simulation result about thermophysiology comfort. At last, a mobile application platform for human exercise comfort prediction is designed and implemented.
In the modern society, people are more and more health conscious to improve their life quality. Exercise is often the first step in lifestyle modifications for health maintenance and management [
During exercise, the human body exchanges energy with the clothing system and environmental conditions in different forms of heat transfer; when the whole human-clothing-environment (HCE) system comes to a thermal steady state, physiological thermal neutrality is reached and the human body will be in a proper thermal and hydration state [
Research on thermophysiology comfort is important for healthy exercise. Based on the thermophysiology comfort model, the human thermal and physiological status can be described and used to predict some accidents like dehydration, exertional heatstroke, syncope, and even sudden death. If these accidents can be analyzed or predicted before they happen during exercise, it will be beneficial to alleviate or avoid disease or mortality. At the Standard Chartered Hong Kong Marathon 2013, 55 runners were reported to have fallen unconscious, to have been rendered comatose, and to have suffered from collapse because of heatstroke [
Some landmark research results can be reviewed [
In this paper, a novel thermophysiology simulation and exercise comfort prediction model is reported. Based on the HCE system, a nonlinear heart rate regulation model and the 25-node thermal regulation model are integrated together to simulate the human physiological performance like temperature, sweat rate, and heart rate. The thermal performance and physiological status of the human body can be simulated in this improved model. Comparisons among different cases show that the improved model can describe the human thermophysiology behavior in the exercise very well. And there is the same prediction trend on the experiment result and simulation result about the thermophysiology comfort.
The main contributions are as follows: Integrate nonlinear heart rate regulation model into the human thermal physiological simulation model; some important thermophysiology parameters during exercise can be simulated by this integrated model. Present a novel exercise thermophysiology comfort prediction model according to the integrated model, which can be used to describe the thermophysiology phenomenon during exercise. Implement a mobile application for comfort prediction, in which people get their physiological comfort status according to the exercise information.
The rest of this paper is organized as follows. Related work is introduced in Section
Research on exercise thermophysiology comfort prediction model involves multidisciplinary knowledge; the human body, clothing, and the environment are a coupled system in the heat and moisture transfer process. The phenomenon of heat and moisture transfer in the HCE system has a significant effect on the human thermophysiology comfort sensation. Figure
Main components of heat and moisture transfer in HCE system.
From Figure
Some research results on heat and moisture transfer in HCE system can be reviewed [
Mathematical models describing the thermoregulatory system of the human body have been the subject of research for years. Reviewed by Cheng and Fu [
Considering the interactions in the HCE system, the development of a mathematical model of the coupled heat and moisture transfer processes in the external environment is accomplished by the boundary condition equations that refer to the thermal status of the external environment and body [
Computer evaluation model is widely used to predict human comfort. Li described the thermophysiological comfort as attainment of a comfortable thermal and wetness state [
Although much progress has been made, there are knowledge gaps that need to be filled in individual areas. They are as follows: There are insufficient advances on the modeling of human body physiological mechanisms during the thermoregulatory processes. Some physiological parameters cannot be simulated in the heat and moisture transfer. There is a lack of advances in mathematical modeling of thermophysiology comfort, especially in dynamic heat balance and thermoregulation of a clothed human body.
This paper, therefore, aims to improve the HCE system simulation model and obtain more human body’s physiological indicators during exercise, to design a novel exercise thermophysiology comfort prediction model. Figure
Schematic diagram of thermophysiology comfort prediction model.
As shown in Figure
According to the schematic diagram of thermophysiology comfort prediction model shown in Figure
During exercise, the human body activates effective thermoregulatory mechanisms to make the body in a proper thermal status. When the temperature of the human body increases, several physiological reactions are activated automatically to speed up body heat dissipation such as sweating and automatically adjusting the cardiovascular system. During cardiovascular adjustment, the blood is redistributed from the core organs to the skin to facilitate heat dissipation, and the active muscles require blood supply to deliver oxygen for maintenance of activity. Meanwhile, the heart rate (HR) increases to sustain cardiac output and blood supply to the working muscles and the skin [
Human thermoregulatory model can be referenced from previous researches [
core layer:
muscle layer:
fat layer:
skin layer:
central blood:
For the control system of model, we have
skin blood:
sweat rate:
Some important indicators are also presented:
Just as mentioned above, the cardiovascular system plays a key role in the thermal regulation process. The heart rate is directly affected by the thermoregulatory mechanism. In heart rate regulation, the metabolic rate and the core temperature are two important factors. In this paper, considering the heart rate regulation mechanism and its fluctuating rules, we propose a new heart rate simulation model. This model includes a quadratic function concerning core temperature and a nonlinear term concerning metabolic rate. The nonlinear term is used to simulate the great fluctuation caused by neuroregulation. The equation of the new heart rate regulation model is shown as follows:
where
The functions
Clothing plays an important role in providing thermal protection for the human body and creating a portable thermal microclimate between clothing and the human body. The heat and moisture transfer process in clothing is responsible for the temperature and humidity distributions and it directly affects the thermal performance of clothing. Heat conduction, heat convention, heat radiation, moisture absorption/desorption, and so forth are basic heat and moisture transfer ways. In this paper, heat and moisture transfer model of clothing used in the HCE system is referenced by some research reports [
Heat and moisture transfer equations of clothing.
Vapor moisture | |
Liquid moisture | |
Heat | |
Liquid diffusion | |
Condensation | |
Moisture sorption | |
Radiation | |
In Table
Considering the heat and moisture interactions in the HCE system, the human body, clothing, and the environment are a coupled system in heat and moisture transfer. The boundary condition equations of the clothing heat and moisture models are accomplished by reference to the thermal status of the external environment and the body.
In practice, the interactive communications between clothing and the human body and clothing and the environment frequently happen by two boundaries. One is the boundary between the body skin and the inner layer of the clothing close to skin; the other is the boundary between the outer layers of the clothing exposed to the environment [
For the inner side of the clothing close to the skin,
For the outer side of the clothing exposed to the environment,
Human comfort can be used to describe the overall state of the body physiologically, which is an important index of body wellbeing. Current researches on human comfort mainly focus on the unilateral prediction of thermal comfort. But, in reality, human thermal senses directly affect the physiological changes. For example, as the temperature of the human body rises, the heart rate, blood pressure, and other physiological signs will change as well. Therefore, the thermal comfort and physiological comfort should be integrated and taken into account. In this paper, an exercise thermophysiology comfort prediction model is designed. The fuzzy inference system [
For the various simulated indicators in our thermal physiological model, we select mean skin temperature, mean core temperature, and change rate of mean skin temperature as input variables, and the prediction results of thermal comfort will be got after the reasoning process. Correspondingly, we select the mean core temperature, sweat accumulation (it is approximately equal to the amount of dehydration), and heart rate as input variables and evaluate the physiological comfort. The comfort variables and the related fuzzy sets are listed in Table
The list of comfort variables and the related fuzzy sets.
Name | Variables | Fuzzy sets |
---|---|---|
Thermal comfort | | Very low, low, neutral, high, very high |
| Very low, low, neutral, high, very high | |
| Fast decrease, decrease, neutral, increase, fast increase | |
| ||
Physiological comfort | | Very low, low, neutral, high, very high |
| Severe dehydration, moderate dehydration, slight dehydration, normal | |
| Low, normal, high | |
| ||
Overall comfort | | Cold, cool, neutral, warm, hot |
| High risk, low risk, normal | |
| Uncomfortable, acceptable, comfortable |
Equation (
Figures
The membership function of thermal comfort indicators.
Mean skin temperature
Mean core temperature
Change rate of mean skin temperature
The membership function of physiological comfort indicators.
Mean core temperature
Sweat accumulation
Heart rate
According to the human mean skin temperature, mean core temperature, and change rate of mean skin temperature, we define a thermal comfort function to predict the human thermal status during exercise. In (
At the same time, physiological comfort of the human body should be considered. In accordance with the human mean core temperature, sweat accumulation, and heart rate, we define a physiological sensation function to predict the human health status during exercise. Equation (
Under a series of comfort inference rules, the overall comfort in (
If thermal sensation is neutral and physiological sensation is normal, then overall comfort is comfortable.
If thermal sensation is cool and physiological sensation is normal, then overall comfort is acceptable.
If thermal sensation is hot and physiological sensation is high risk, then overall comfort is uncomfortable.
Three exercise cases are designed to evaluate the exercise thermophysiology comfort prediction model.
Different types of exercises are used for thermal physiological simulation and comfort prediction. Table
Cases definition.
Case | Settings | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
Subject | Clothing | Environment | Exercise | |||||||
Gender | Age | Weight | Body | Composition | Coverage | Temperature | Relative | Exercise | Duration | |
(years) | (kg) | area (m2) | | rate | (°C) | humidity | type | (min) | ||
Case 1 | Male | 25 | 70 | 1.8 | 98% cotton, 2% lycra | 70% | 32 | 50% | Running | 30 |
Case 2 | Jogging | 30 | ||||||||
Case 3 | Walking | 30 |
During exercise, the human body’s thermal status and physiological status dynamically change, mainly reflected in the following phenomenon: temperature rising, heart rate accelerating, sweating increasing, and so forth. Figure
The simulation results of our improved thermal physiological model in different scenes; they should be listed as (a) mean core temperature tendency, (b) mean skin temperature tendency, (c) sweat accumulation tendency, and (d) heart rate tendency.
Mean core temperature
Mean skin temperature
Sweat accumulation
Heart rate
Figure
Table
Results of thermophysiology comfort prediction for three cases.
Cases | Time (seconds) | Thermal values | Physiological values | Overall comfort |
---|---|---|---|---|
| | |||
(°C/°C/°C/—) | (°C/g/—/—) | |||
Case 1 | | 32.44/36.99/0.44/neutral | 36.99/0.00/78.49/normal | Comfortable |
| 34.89/37.44/2.89/neutral | 37.44/3.22/140.95/normal | Comfortable | |
| 34.91/37.45/2.91/warm | 37.45/3.27/140.97/normal | Acceptable | |
| 36.36/38.29/4.36/warm | 38.29/40.22/153.93/normal | Acceptable | |
| 36.36/38.30/4.36/hot | 38.30/40.37/153.97/low risk | Uncomfortable | |
| 35.66/39.67/3.66/hot | 39.67/338.35/171.69/high risk | Uncomfortable | |
| ||||
Case 2 | | 32.44/36.99/0.44/neutral | 36.99/0.00/78.49/normal | Comfortable |
| 34.88/37.38/2.88/neutral | 37.38/3.52/119.90/normal | Comfortable | |
| 34.90/37.39/2.90/warm | 37.39/3.57/119.90/normal | Acceptable | |
| 36.38/38.29/4.38/warm | 38.29/67.36/136.49/normal | Acceptable | |
| 36.38/38.30/4.38/hot | 38.30/67.51/136.52/low risk | Uncomfortable | |
| 35.27/38.73/3.27/hot | 38.73/243.23/145.43/low risk | Uncomfortable | |
| ||||
Case 3 | | 32.44/36.99/0.44/neutral | 36.99/0.00/78.49/normal | Comfortable |
| 34.89/37.31/2.89/neutral | 37.31/4.16/99.28/normal | Comfortable | |
| 34.90/37.31/2.90/warm | 37.31/4.20/99.29/normal | Acceptable | |
| 36.41/38.19/4.41/warm | 38.19/155.83/127.12/normal | Acceptable |
In the first 166 s of case 1, people feel comfortable, since all thermal sensation and physiological sensation keep normal. From 167 s to 513 s, people feel acceptable since thermal sensation changes to warm, but physiological sensation still keeps normal. After that, from 514 s, both the thermal values and the physiological values are changed; people feel uncomfortable with the thermal sensation getting into hot and physiological sensation getting into low risk and even high risk. In case 2, from 0 s to 213 s, people feel comfortable since both thermal sensation and physiological sensation are normal. From 214 s to 815 s, the human comfort is acceptable. From 816 s, people feel uncomfortable with the thermal sensation getting into hot and physiological sensation getting into low risk. In case 3, from 0 s to 320 s, people feel comfortable. After that, people feel acceptable until the end of the exercise.
The experiment results show some important and valuable suggestions: for example, walking is a comfortable and acceptable exercise in daily life; our simulated results also tell us that walking within 30 minutes is acceptable and cannot cause any discomfort. Jogging for a relatively long period of time also makes people feel comfortable, while it will make people feel uncomfortable when the exercise time exceeds 13.5 minutes. Therefore, we should pay attention to drinking water and cooling while jogging for a long time. The results also show that running will easily cause body discomfort. When people run at 32°C in 8 minutes, it is easy for them to feel uncomfortable. Running makes people feel uncomfortable by changing body temperature into a hot state and putting the human body’s physiological state into a high risk state. So, we recommend not to run for a long time in a hot environment.
The aim of the case study is to evaluate human comfort under different exercise intensities. Therefore, we take exercise intensity as variable, and other factors (subject, clothing, environment, etc.) as invariants in the setting of the case study. It is worth noting that this does not mean that our model cannot simulate the thermal physiological changes and predict human comfort caused by other factors. To support this conclusion, some extra cases are discussed as follows.
The tendency curves of mean skin temperature and sweat accumulation with different subjects.
Mean skin temperature
Sweat accumulation
The tendency curves of mean core temperature and sweat accumulation with different clothing.
Mean core temperature
Sweat accumulation
Results of thermophysiology comfort prediction in different environments.
Cases | Time | Thermal values | Physiological values | Overall comfort |
---|---|---|---|---|
| | |||
(°C/°C/°C/—) | (°C/g/—/—) | |||
25°C, 50% RH | | 31.67/37.01/ | 37.01/0.00/79.19/normal | Comfortable |
| 34.90/37.44/2.90/neural | 37.44/5.88/117.47/normal | Comfortable | |
| 34.90/37.44/2.91/warm | 37.44/5.93/117.48/normal | Acceptable | |
| 35.92/38.30/3.92/warm | 38.30/87.40/133.78/normal | Acceptable | |
| 35.92/38.30/3.92/hot | 38.30/87.55/133.80/low risk | Uncomfortable | |
| 35.96/38.77/3.96/hot | 38.77/210.52/146.46/low risk | Uncomfortable | |
| ||||
35°C, 70% RH | | 32.88/37.04/0.88/neutral | 37.04/0.00/80.31/normal | Comfortable |
| 34.89/37.36/2.89/neutral | 37.36/2.66/121.68/normal | Comfortable | |
| 34.91/37.36/2.91/warm | 37.36/2.70/121.68/normal | Acceptable | |
| 36.60/38.30/4.60/warm | 38.30/61.56/138.11/normal | Acceptable | |
| 36.60/38.30/4.60/hot | 38.30/61.72/138.14/low risk | Uncomfortable | |
| 36.60/39.30/4.60/hot | 39.30/299.33/160.91/low risk | Uncomfortable |
The tendency curves of mean core temperature and sweat accumulation with different environments.
Mean core temperature
Sweat accumulation
Although we have not elaborated the effects of subject, clothing, and environment on human thermal physiological simulation as well as comfort prediction, our thermal physiological model and comfort prediction model are capable of simulating and analyzing the effects caused by these factors in HCE systems.
With the development of mobile communication technology and increasing popularization of the Internet, mobile multimedia services are more and more favored by users. Various mobile devices and applications are designed to aid people to improve their life quality. Exercise thermophysiology comfort is regarded as one of the most important and significative research areas, which has been focused upon in recent years. According to the previous description and the mobile application requirements of human comfort, a user-friendly smart application with low computational requirements has been developed to evaluate human exercise comfort in daily life, which allows easily changing the simulation scenes and simulating the human physiological status as well as carrying out comfort evaluation and prediction. The basic architecture for the prototype is shown in Figure
The basic architecture for the comfort prediction prototype in mobile scenario.
Various parameters in the scene of case study directly affect simulation results, and the different combination of scene parameters will produce different physiological state and comfort sensation. Four main types of scene parameters are defined, which are personal parameter, clothing parameter, activity parameter, and boundary parameter. Figure
The scene definition views of the app.
Scene definition
Personal parameter
Activity parameter
Server side is used to handle the time-consuming and computing resource-intensive simulation task in HCE system. It takes scene parameters as input and outputs the body comfort sensation.
The server receives input parameters from clients, and then numerical solutions are taken to solve the human physiological model, heat and moisture transfer model, and the interactive equations between body and clothing. After that, the server uses the simplified neurofuzzy inference system to carry out comfort evaluation and prediction. The simulated results such as human skin temperature, heart rate, sweat rate, and comfort sensation are generated. At last, all these results are transferred back to the smart devices. The data transmission between the client and the server is encapsulated as a customized file in XML format.
A variety of graphical representations are used in our application to help users visualize the change of physiological state and comfort sensation. Figure
The views of simulation results.
Physiological data list
Comfort result
During exercise, the heat balance of the human body is maintained by the processes of heat production and heat loss via radiation, conduction, convection, and evaporation. These processes would have effects on the physiological responses and influence the thermal status and comfort perception. In order to predict human thermophysiology comfort, in this paper, a heart rate regulation model is added to HCE system to simulate the human body thermal physiological behavior; according to this improved model, some important physiological parameters can be obtained. Further, in this paper, a novel thermophysiology comfort prediction model and a user-friendly mobile application for human comfort prediction are designed. The experiment results show that there is the same prediction trend on the experiment result and simulation result about thermophysiology comfort. The proposed exercise thermophysiology comfort prediction model still has some limitation. The thermal physiological mechanism needs to be researched to simulate the human physiological sensation further. We need to achieve and analyze more exercises and even investigate how to apply this simulation model and comfort model in the health services.
Radiation absorption constant of the fiber
Porosity of the fabric
Volume fraction of water vapor
Volume fraction of fibers
Volume fraction of liquid phase
Dynamic viscosity of liquid (kg/ms)
Surface tension of fiber (J/m)
Effective sorption rate of the moisture
Evaporation/condensation rate of the liquid/vapor
Heat of sorption or desorption of liquid by fibers (kJ/kg)
Heat of sorption or desorption of vapor by fibers (kJ/kg)
Density of the liquid water (kg/m3)
Stefan-Boltzmann constant (W/m2 K)
Effective tortuosity of the fabric for water vapor diffusion
Contact angle of the liquid water on the fiber surface
Heat transfer by blood flow in node
Blood flow rate of node
Basal blood flow rate of node
Thermal capacity of the blood (Wh/°C)
Thermal capacity in node
Saturated water vapor concentration at
Water vapor concentration in the air filling the interfiber void space (kg/m3)
Shivering metabolic heat generation in node
Volumetric heat capacity of the fabric (kJ/m3 K)
Weighting and distribution coefficient of shivering muscles
Cold signal of node
Integrated cold signal of the whole skin surface
Heat transfer by thermal conduction in node
Diffusion coefficient of water vapor in the air of the fabric (m2/s)
Diffusion coefficient of water vapor in the fibers of the fabric (m2/s)
Control signal of vasodilation
Diffusion coefficient of liquid water in the fibers of the fabric (m2/s)
Heat loss by evaporation through the skin surface in node
Error signal of node
Elementary total thermal radiation incident inside the clothing travelling to the left/right (W/m2)
Convection heat transfer coefficient (W/m2 K)
Radiation heat transfer coefficient (W/m2·°C)
Integrated heat transfer coefficient (W/m2·°C)
Evaporation heat of water (J/kg)
Mass transfer coefficient for evaporation and condensation (m/s)
Thermal conductivity of the air (mmW/m2·°C)
Effective thermal conductivity of the fabric (W/m/K)
Regional influence factor
Sweating accumulation on the skin surface in the
Regulatory sweating in the
Water vapor pressure of ambient temperature in the
Saturation water vapor pressure on the skin temperature in the
Water vapor pressure on the skin surface in the
Metabolic heat generation in node
Basal metabolic heat generation in node
Heat loss by convection and thermal radiation in node
Radius (mm)
Evaporation heat resistance on the skin surface in the
Evaporation resistance of the skin in the
Latent respiration heat loss in node
Width of temperature
Control signal of vasoconstriction
Surface-to-volume ratio of the fiber (
Weighting and distribution coefficient of vasoconstriction in the
Integrated weight coefficient
Weighting and distribution coefficient of sweating in the
Weighting and distribution coefficient of vasodilation in the
Temperature of the fabric (K)
Temperature of the blood (°C)
Temperature of node
Thickness of the air layer (mm)
The set-point temperature of node
Work accomplished in node
Warm signal of node
Integrated warm signal of the whole skin surface.
The authors declare no competing interests.
This research is supported by the National Natural Science Foundation of China (NSFC) (nos. 61320106008, 61402185, and 61672547).