Buildings in the Hong Kong Special Administrative Region, China, with sub-tropical climates are subjected to high cooling requirements throughout most of the year. Among the electricity consumption in commercial buildings, about 60% is attributable to the operation of air-conditioning systems for comfort cooling. This is about 11.5% of the final energy requirement in Hong Kong, which is much more significant compared with the 6.1% final consumption for commercial space heating and cooling in the member countries of the International Energy Agency. Here the cooling load is particularly dependent on the construction of the building envelope and the weather conditions. To limit the amount of heat gain through the building envelope is obviously an important step for reducing the cooling energy consumption.

Following several year’s preparation work by the former Hong Kong Government, a Building (Energy Efficiency) Regulation dealing with envelope construction designs stipulating a limit on the overall thermal transfer value (OTTV) has been enacted since 1995. The OTTV is used to describe the maximum rate of thermal transfer permissible into the building through its wall or roof, due to solar heat gain and outdoor-indoor temperature difference. It was initiated by ASHRAE as a cooling criteria to limit the amount of heat gain through the building fabric, hence the energy consumption for cooling could be reduced. Other Southeast Asia countries, e.g. Singapore and Malaysia, are also concerned with the utilization of building energy, particularly on the cooling load resulting from the external heat gains through the building envelopes in air-conditioned buildings. An overall thermal transfer value is stipulated in their building control regulation to evaluate the design on the thermal performance of the building envelopes.

Almost all commercial buildings in Hong Kong are high-rise. There are two common constructions: the traditional one with concrete frame and load-bearing external wall, and the other with steel or concrete frames plus a non load-bearing external wall. While concrete wall structures with ceramic tile finish or stucco finish are typical for older buildings, curtain walls are common for commercial buildings since the early 1980s. These buildings vary in the types of fenestration, window-to-wall area ratio, materials of opaque wall, with or without insulation, and light or heavy weight constructions.

A total of 64 buildings were surveyed to study the construction characteristics of existing commercial buildings in Hong Kong. The construction details were obtained from in-situ investigation, architectural drawings and from other references for some cases. Most of these buildings were built in the past 10 years, and some were under construction at the time of the study.

The findings show that the external walls can be classified into five types based on the construction methods, designated type I to V as illustrated in Fig. 1. On the other hand, local offices are also graded as A, B or C by the Rating and Valuation Department. Grade A offices have high-quality finishes and good management, Grade C offices have basic finishes and require only minimal management, and Grade B offices are average ones. Type I external-wall construction is traditional, and is found in most Grade C and some Grade B buildings. It has a superficial mass commonly exceeding 300 kgm^?2 of envelope area, and is characterized by high wall U-values and tinted glass. Twenty per cent of the surveyed buildings belong to this category.

Grade A buildings typically have external wall construction of types II, III, IV or V, which in effect are various forms of curtain wall and account for 80% of the surveyed buildings. Types II and III are medium to heavy-weight curtain wall structures, having an inner heavy-weight concrete layer with spandrel glass or granite panel facade, with or without a fiberglass insulation layer in between. These are most common and account for 68% of the studied buildings. Types IV and V are non-load-bearing constructions, characterized with a light-weight, thick fiberglass insulation layer giving low U-value, highly reflective vision glass, attractive spandrel glass or sometimes an aluminum wall panel facade, and an innermost layer of gypsum board.

A similar investigation for local buildings has been reported but it provides less details, e.g. HKIE OTTV ad-hoc Committee; or for the residential sector, e.g. Lam. The distribution of the sample of surveyed buildings in this study over the range of each key building fabric parameters, including window-to-wall ratio, fenestration-shading coefficient, and opaque wall U-value, are shown in Fig. 2, Fig. 3 and Fig. 4. Many commercial buildings, particularly those with curtain walls, are characterized by large window areas. Window-to-wall ratios (WWR) of 0.35–0.7 are common, with the 50% level at 0.47 (i.e. half of the studied buildings have window-to-wall ratios below 0.47, and the other half above). The large window areas are very often coupled with highly-reflective glass having a low shading-coefficient, though the reflective glass is chosen by architects mainly out of aesthetic consideration rather than reduction of solar gain. While most buildings have a shading coefficient of between 0.25 and 0.5, the 50% level is at 0.39, which implies that half of the buildings have a shading coefficient of less than 0.39. Notwithstanding the low shading-coefficient, most of the buildings have single-glazed windows. Among the studied buildings, only five use double-pane glazing, which reflects that this practice is not common locally.

The U-values of the external walls of the studied buildings can be grouped into two categories. The insulated-wall category consists of about 40% of the buildings and each has U-value between 0.5 and 1 Wm^?2 K^?1. The other 60% have non-insulated walls, where the U-value can be less than 2 Wm^?2 K^?1 if the wall is massive and has an air layer, but can be as high as 3.4 Wm^?2 K^?1, which is typical for the usual concrete wall with ceramic tile finish. The 50% level is at U-value of 1.9 Wm^?2 K^?1. This generally high wall-U-value differs very much from the standard wall constructions in regions with cold climates, where insulation is important for the reduction of heat loss.

Use of heavy-weight concrete (typically 2400 kgm^?3) for the construction of slab and external walls is common in local practice. The majority of the studied buildings have external wall masses in the range of 300–450 kgm^?2 of envelope area, owing to a layer of 100–150 mm concrete. A light-weight construction with no concrete layer, but having insulation, is becoming popular in the latest building practices. The mid-level of wall superficial mass densities is 360 kgm^?2, i.e. 50% of the studied buildings have wall masses below this value.

The effect of building-envelope parameters on the heat gain and chiller load can be studied by parametric analysis and has been reported by Chow and Chan.

The hour-by-hour heat conduction and solar radiation through the building walls and windows were determined with the building energy simulation tool DOE-2.1D. The transient heat conduction through the wall is solved using the response-factor method and principle of superposition, with time-series impulse transfer function to determine the transient behavior of a building. One-dimensional heat flow is assumed for the heat flows by conduction through the considered wall with finite thermal-capacity. The heat flux responses to temperature excitation at the outside surface are determined at hourly intervals.