Regarding the hydration heat of water vapor adsorbed on inorganic adsorbents, the hydration heat per unit amount of water vapor adsorbed is constant for a material having a solid three-dimensional structure, such as zeolite. However, for adsorbent having an amorphous structure, the pore size for adsorbing water vapor is widely distribute: therefore, the hydration heat changes depending on the relative vapor pressure during water vapor adsorption. As HAS-Clay (HAS: Hydroxyl Aluminum Silicate) is an adsorbent having an amorphous structure, the hydration heat changes depending on the relative vapor pressure when water vapor is adsorbed. The hydration heat per unit amount of water vapor adsorbed by HAS-Clay decreases with increasing relative vapor pressures. In this study, the hydration heat for each relative vapor pressure at the time of water vapor adsorption on HAS-Clay was obtained based on the hydration enthalpy value derived from the calorimetric experiment conducted using inorganic adsorption materials and water vapor adsorption isotherm of HAS-Clay.

In future smart grids, fast automated demand response (FastADR) may be realized through real-time pricing (RTP), in which the electricity rate significantly fluctuates at intervals of approximately 10 minutes. Among building load facilities, multi-type air-conditioners consume high amounts of power and are installed in many office buildings, so that they can be strong candidates for load control using FastADR. A control method for the multi-type air-conditioners that adapts to RTP and an algorithm that adjusts the trade-off between the electricity charge and room temperature comfort are required. By applying reinforcement learning to this algorithm, a control that automatically adapts to various air-conditioning environments in office buildings can be considered. However, in general, reinforcement learning requires several thousand trials to reach the practical control requirements. If we directly implement reinforcement learning in actual multi-type air-conditioners used in buildings, it will spend years of time for learning. Therefore, in this study, we constructed a reference virtual building using "air-conditioner emulator (AE)" which simulates the dynamic characteristics of power and room temperatures of building with multi-type air-conditioners, and tried to shorten the learning period through pre-learning using the reference virtual building. As the reference virtual building is a standard one for practical use, differences in the air-conditioning environment between the reference virtual building and an actual building will exist. It is necessary to clarify to what extent these differences in the air-conditioning environment can be tolerated. It is difficult to create various air-conditioning environments in a real building, and reproducibility cannot be ensured. Therefore, we used a derived virtual building instead of a real building, which was obtained by magnifying the cooling capacity, heat load, and heat capacity of the reference virtual building. We clarified the relationship between the magnification rates and the learning periods until the control performance reached practical performance requirements.

In rooms with large windows, the thermal environment tends to deteriorate during air-conditioned heating in winter. To improve the thermal environment near windows with low insulation performance, equipment such as perimeter fans and natural convection type heaters are used. In this study, temperature and velocity distributions using particle image velocimetry were measured for thoroughly understand ing the heat transfer phenomena and obtaining simulation verification data when the equipment was installed under the cooled window. In particular, the airflow rates during upflow and downflow near windows were estimated for verifying and developing vertical temperature distribution prediction models. The convective and radiative heat transfer coefficients were calculated from the measured room temperature and on the window surface heat flow rate.

In recent years, ceiling radiation air-conditioning systems are being increasingly used in office buildings in southern Japan to achieve energy conservation and comfort. Now, introduction of these systems are being introduced in cold regions. We conducted case studies on air-conditioning systems and window specifications for office buildings in cold regions during winter through CFD analysis, and compared the indoor thermal environments. The results, confirmed that a comfortable indoor thermal environment was formed by introducing the ceiling radiant heating and cooling system (water conveyance), and calm air flow was confirmed. Furthermore, with proper the heating in perimeter zone, air conditioning was unnecessary in interior zone, and this may lead to further energy savings.

In recent years, water usage equipment including sanitary appliances for water savings and hot water savings have advanced remarkably. However, the water supply unit (quantity) written in the SHASE (The Society of Heating, Air-Conditioning and Sanitary Engineers of Japan) Handbook, serving as the basis for planning and designing of building services, have not been reviewed for the past half century. Moreover, these units are considered outdated. The purpose of this study was to determine new water supply units for various building use. We first focused on hospitals. We propose new water supply units (quantity) for hospitals that are suitable for new hospitals in the future, using published values, standards, and measured values for water use in general hospitals. Furthermore, by identifying the range of the water supply units, which has been unclear until now, designers can design water supply systems that better match the characteristics of the building.