|
Article Excerpt
INTRODUCTION
One of the most important research targets in the field of heating, ventilating, and air conditioning (HVAC) is thermal sensation. In 1997, Murakami et al. proposed a numerical analytical tool--defined as a computational thermal manikin on the basis of coupled simulation of computational fluid dynamics (CFD), radiation, moisture transport, and heat transfer inside the human body--as a new resolution method with regard to thermal sensation. Since then, it has been accepted as a more economical, effective, and attractive tool for examinations of thermal sensation when compared with the experimental method using an experimental manikin or human subjects. However, the human thermal physiological models in the studies of Fanger (1973) and Gagge et al. (1971), which dealt with whole-body heat balance rather than local balance, were adopted to calculate heat transfer inside the human body. Additionally, a simply shaped human body, with feet and arms together or close to the body, was used in CFD and radiation simulations. Because thermal sensation is highly dependent on local heat transfer characteristics, which greatly depend on the properties of the local microclimate, Murakami et al.'s (1997) computational thermal manikin is inapposite for investigating thermal sensation of a human body located in non-uniform environmental conditions.
Therefore, in a previous paper (Zhu et al. 2007), in order to develop a computational thermal manikin that can successfully predict thermal sensation of a person in typical multifarious non-uniform environments, we adopted the multi-element human thermal physiological model developed by Smith (1991), which has a three-dimensional (3D) shape and can produce 3D temperature distributions, to calculate the heat transfer inside the body; in addition, a human body with a complex shape that included feet, arms, jaw, and breasts was used in the CFD and radiant simulations. The coupled simulation method was applied to calculate the heat transfer over the body surface of a person in uniform and front-back asymmetric radiant environments. According to the results, although the simulation results closely reflected the effect of the warm/cold radiation on the local heat transfer, the local skin temperatures differed greatly from those obtained in the corresponding subject experiments at the limbs, even in a uniform environment. It can be concluded that the improper modeling of the Arteriovenous Anastomose (AVA) phenomenon in Smith's model is primarily responsible for the large discrepancy generated at the limbs.
In view of the above results, a new human thermal physiological model, Sakoi's model, was developed with a 3D body configuration similar to Smith's model (Sakoi et al. 2005a, 2006a). In Sakoi's model, in order to generate the AVA blood flow at the limbs, the blood circulatory system is remodeled as in Stolwijk's (1971) model, with the blood perfusion rate for skin tissue changed numerically under a fixed volume of vessels. In this paper, Sakoi's model is adopted in the coupled simulation. Moreover, moisture transport is combined into the simulation of CFD, radiation, and Sakoi's model to evaluate the latent heat transfer from a human body. The coupled simulation is applied to a human body located in uniform and front-back asymmetric radiant environments with an ambient air temperature of 28[degrees]C, and its prediction accuracy is verified by comparison with the corresponding results of the human subject experiments and the coupled simulation using Smith's model, as introduced in the previous paper, in terms of skin temperatures. Based on the comparison, the deficiencies existing in the present simulation method are examined.
OUTLINE OF SAKOI'S HUMAN THERMAL PHYSIOLOGICAL MODEL
Sakoi's model (Sakoi et al. 2005a, 2006a) is proposed to predict local skin temperatures and heat losses under non-uniform thermal environments and various clothing conditions. Like Smith's (1991) model, it can numerically analyze the 3D temperature distribution, the redistribution of body heat due to blood circulation, the promotion of heat conduction in skin tissue by conduction of blood flow, the respiratory heat loss, and the total (sensible and latent) heat loss at the skin surface using a finite element method. However, in contrast to Smith's model, Sakoi's model adopts a thermo-regulatory mechanism proposed by Yokoyama et al. (2000) to deal with the segmental characteristics of physiological responses.
As shown in Figure 1a, in Sakoi's model, the 3D shape of Smith's model is adopted and the whole body is also approximated geometrically by 15 cylindrical body parts (head, neck, torso, upper arms, thighs, forearms, calves, hands, and feet). However, in Sakoi's model each body part is divided into more detail to obtain a more accurate 3D temperature distribution, and the whole body is made up of 2744 3D (triangular or rectangular) tissue elements (compared to 936 in Smith's model), of which 536 elements (compared to 338 elements in Smith's model) have surfaces exposed to the external environment. Each tissue element is designated as a specific type, such as brain, bone, lung, viscera, muscle, fat, or skin, with a corresponding physical property following that proposed by Smith. Although the model also has two blood circulations, i.e., a main circulation and a pulmonary circulation, its blood circulatory system is wholly different from Smith's model. Figure 1b illustrates the vascular networks. The arterial networks and venous networks are symmetrical, and a 3D tissue, arterial blood pools, and venous blood pools share the same 3D space at a fixed volume. The blood flow is not controlled by the dilation and constriction of vessels of the model; rather, its perfusion rate for skin tissue changes numerically under the fixed volume of the vessel. Based on a review of past studies of mean values for the whole body, the most suitable values are selected for thermal conductivities, densities, specific heats, volumes of arterial and venous blood, basal blood perfusion rates, basal metabolic heat production, and diameters for the one-dimensional arteries and veins (Ganong 1973; Sasaki 1981; Kodansha 1989; Ochi 1990; Smith 1991; Yokoyama 1993). In terms of results, in Sakoi's model, the body weight is 75.94 kg, the surface area is 1.837 [m.sup.2], the total blood volume is 6682 mL, the total heart output rate in a thermally neutral state is 5782 mL/min, and the basic metabolic rate is 47.8 W/[m.sup.2].
[FIGURE 1 OMITTED]
In conclusion, Sakoi's model (Sakoi et al. 2005a, 2006a) is the only human...
|
