Abstract Environmental control strategy via computerized implementation is one of the most efficient approaches to integrate new advanced knowledge in research of human thermal comfort to a mechanical air-conditioning unit. Recently, a new conceptual development in designing air-conditioning systems has indicated that the indoor comfort temperature strongly depends upon changes of the outdoor air temperature rather than to be a conventional fixed temperature set-point. The explanation is due to occupants’ adaptability of thermal comfort to a dynamic environment in terms of their clothing and/or activities while the outdoor temperature can be explicitly used as an ultimate indicator of such changes to empirical function of the indoor comfort temperature. In this paper, the first prototype embedded system is developed to emulate such an adaptive algorithm to numerically determine an indoor comfort temperature for a real-time control in an air-conditioning system. From a theoretical point of view, an adaptive comfort model together with grey prediction model is presented for exploring a practical application of a comfort temperature-based control for a single air-conditioned space, so as to show the viability of the proposed methodology by simulated results. The field studies by interview survey of satisfaction on thermal comfort within an air-conditioned reading room of a library confirm the viability of the proposed real-time computerized implementation of adaptive indoor comfort temperature via the embedded system for a conventional air-conditioning unit in practical uses.5821

Keywords: Adaptive control; Air-conditioning unit; Embedded systems; Indoor comfort temperature

1 Introduction
Thermal comfort within air-conditioned spaces affects occupants’ productivity and health. The climate of Thailand is tropical-hot and humid. Like other countries, variations of year-round temperature and humidity with extensive ranges impose challenges to efficiently maintain thermal comfort for adaptable occupants within air-conditioned spaces. Currently, control methodologies and computing technologies are key successful elements for improving such systems' efficiency in all steps of energy conversion. In turn, researchers’ attentions have been drawn to thermal acceptability of indoor environments, which is particularly intended in an energy-intensive environmental control strategy for an air conditioning system [1]. This means that alternative schemes for an air-conditioning control system are to be tightly developed to guarantee satisfaction of most occupants' human comfort all the operating time so that, in sequence, energy is consumed efficiently. Energy is efficiently consumed by an air-conditioning system if both thermal comfort and air quality are satisfactory, via supplying make-up air to an air-conditioned space over the operating time. Conventionally, air quality of an air-conditioned space can be improved through demand-driven air ventilation with a suitable intake amount of fresh air from the outdoor environment as shown in Fig. 1.
   It has been shown experimentally that the obligation on the air quality can be fulfilled independently by optimally supplying fresh air together with return air [2]. With this implication, the complicated problem of handling an air-conditioning system can be simplified to just govern physical conditions of the supply air for thermal comfort, such as temperature, humidity of the supply air, etc. Nowadays, a conventional air-conditioning system available in the market has been widely implemented with a practical basis of temperature recommendation. This scheme being applied to a typical air-conditioning system means that the temperature of the indoor air is to be regulated at a fixed reference temperature under dynamically environmental changes and adaptable occupants, which actually is quite ideal.
   According to ASHRAE Standard 55-1981 [3], thermal comfort is defined as that condition of mind which expresses satisfaction with the thermal environment. Thermal-comfort perceptions of environments are affected by six thermal variables: (1) air temperature, (2) radiant temperature, (3) relative humidity, (4) air velocity, (5) metabolic rate of occupant in various activities, and (6) clothing [4]. In standard studies [5,6], this systematic development of thermal comfort science is based on the principle of heat balance to the human body within a climate chamber in steady conditions of those thermal variables. From experimental studies, the resulting equation of predicted mean vote (PMV) can be used to estimate human mean response to thermal environment from those six thermal variables. The thermal-response index of PMV consists of synonyms to categories: cold (-3), cool (-2), slightly cool (-1), neutral (0), slightly warm (1), warm (2) and hot (3). Theoretically, a minimum rate of dissatisfaction exists at thermal neutrality (PMV=0) and the dissatisfaction percentage increases exponentially as the thermal response index changes away from neutrality. It should be noted that the last two thermal variables above, that is, metabolic rate and clothing, can be adapted freely by occupants themselves to be neutral according to human feeling, local climate or environmental constraints such as season and geography. Therefore, there still has been argument that PMV-based climate control established by thermal balance in steady state conditions does not reflect real-life conditions, particularly when occupants’ adaptability to dynamical environment takes place all the time. In other words, occupants are active recipients of thermal environment; therefore, they react in such a way that their comfort can be restored when any climate change occurs, such as to produce discomfort [7]. 嵌入式系统对空调舒适温度控制的英文文献和翻译:http://www.youerw.com/fanyi/lunwen_3023.html