• Energy Research

202407 bcch ; Accepted Manuscript ; RGC ; Published ; Green (AAM)

Abstract Methods to estimate RTTV (roof thermal transfer value, or equivalent) currently adopted by numerous South East Asian countries have different inherent limitations in accurately evaluating the thermal performance of cool roofs. These existing methods either use a fixed value to represent the solar reflectance effect or assume a linear correlation between annual-averaged conduction heat gain and solar absorptance, which are shown to be inaccurate. In this paper a new RTTV model is proposed with new formulation for modelling the equivalent thermal resistance increment due to the solar reflectance effect on opaque roofs using the previously developed CRHT (Cool Roof Heat Transfer) model. The new formulation is incorporated into the U-value estimation of the heat conduction gain component. The new RTTV model is validated against computational simulations and experiments on an air-conditioned test building with flat concrete roof in Singapore. The Current RTTV model adopted in Singapore gives large error in RTTV estimation for the test building at high-reflectance (over-estimates by 20 times at reflectance = 0.90) whereas the new RTTV model gives much more accurate estimations (maximum error within 12%). The proposed method for new RTTV model formulation is also applicable to other similar models, and is not limited by climate conditions.

Abstract Double-skin roof is a popular passive cooling solution to curb heat gain into buildings and cool roof is another emerging solution. This study proposed a novel CRHT (cool roof heat transfer) model for double-skin roof which is able to model the heat transfers for a double-skin roof combined with cool roof. The CRHT model was validated against experiments performed in two identically-configured, naturally ventilated apartments in Singapore. CRHT predictions match with experimental measurements with reasonable accuracy. White-color cool coating on a flat double-skin roof reduces the daily heat gain by 0.21 kWh/m 2 (or 51%), resulting peak indoor air temperature reduction by 2.4 °C on a sunny day. Furthermore, thermal performance of cool roof is compared with double-skin roof using the CRHT model. In the roof setup of the current study, double-skin roof is about 6% more effective than cool roof in reducing annual heat gain into the apartment during day time. However, the extra insulation of double-skin roof hinders the heat loss during night time, ensuing cool roof is almost equally effective in reducing net annual heat gain. The proposed CRHT model is generally applicable to any climate conditions as demonstrated by applying it for Mediterranean climate of Athens, Greece.

Abstract Machine-learning (ML) –based building models have been gaining popularity in constructing model predictive control (MPC) for building energy management applications. However, ML-based building models are usually nonlinear so to capture the building dynamics, leading to high computation load for MPC, prohibiting its application for real-time building control. This study proposes a ML-based MPC with an instantaneous linearization (IL) scheme, which employs real-time building operation data to linearize the nonlinear ML-based building model for constructing a linear MPC at each control interval. The proposed ML-based MPC with IL system is implemented to control an air conditioning system in an office of a general hospital building located in Singapore for experimental evaluation of its control performance. The ML-based MPC with IL is compared to a ML-based MPC that directly uses a nonlinear ML-based building model and the original reactive-control-based thermostat of the office. Results show that the ML-based MPC with IL significantly reduced the computation time (by more than 70 times) as compared to the ML-based MPC while retained most of the advantages of the ML-based MPC. The ML-based MPC with IL and the ML-based MPC achieved 31.6% and 26.0% reductions, respectively, in cooling energy consumption as compared to the original thermostat. Meanwhile, both the MPC systems significantly improved indoor thermal comfort for the office as compared to the original thermostat. The study demonstrated that using IL for ML-based MPC could substantially improve computation efficiency with no obvious performance degradation in terms of thermal comfort and energy saving.

Abstract Sub-ambient cooling can be achieved through radiative coolers that selectively emit radiation within the atmospheric window (8–13 μm) to outer space and suppress absorption/emission of other wavelengths. This study explores the feasibility of adopting radiative cooling in the hot and humid climate of Singapore through both numerical and experimental approaches. A theoretical simulation based on the heat transfer balance is first proposed to obtain the cooling power of the radiative cooler considering different solar spectral irradiance and total water vapor column. The larger solar irradiance in Singapore, especially within the ultraviolet and visible light spectrum where the absorbance of the material is relatively high, could counteract its cooling effects. Moreover, the increased atmospheric radiation induced by higher humidity and temperatures in Singapore could worsen cooling performances of the radiative material. Next, experimental investigations were conducted by measuring the steady-state temperatures of two radiative coolers (photonic radiative cooler and enhanced specular reflector film) under three typical weather conditions in Singapore, namely clear, partly cloudy and cloudy skies. While both radiative coolers were unable to achieve daytime cooling performance on a clear day, the enhanced specular reflector (ESR) film with higher solar reflectance can reach sub-ambient temperatures on a cloudy day. When it comes to night-time, the steady-state temperature of the photonic radiative cooler and ESR film was about 3.5 °C and 5 °C lower than ambient, respectively.

Abstract In this work, an analytical Complex Fast Fourier Transform (CFFT) method is used and modified to predict the transient roof temperature and transmitted heat flux through the multilayer roofs of naturally ventilated rooms. A field experiment is carried out on two full-scale roofs to validate the CFFT model. The mean bias error (MBE) and cumulative variation of root mean square error (CVRMBE) in the ceiling temperature prediction using CFFT model are found less than 4% during both sunny and rainy days. After validation, a parameter study is conducted to investigate the impacts of rooftop surface solar reflectivity (from 0.1 to 0.9) and thermal resistance (from 0.1 to 2.5 m 2 K/W) on the thermal performance of two types of concrete-based roofs, namely the unventilated and ventilated roofs. Compared to the roofs with solar reflectivity of 0.1, increasing the solar reflectivity by 0.1 reduces the daily heat gain by 11% in both the unventilated and ventilated roofs during a typical weather day in Singapore. Compared with the unventilated roofs, the individual uses of roof ventilation and 2.5-cm expanded polystyrene (EPS) foam insulation reduce the daily roof heat gain by 42% and 68% respectively, and the daily roof heat gain reductions increase to 73% and 84% in the ventilated roofs incorporated with 2.5-cm EPS foam and radiant barrier respectively.

Abstract Active chilled beams (ACB) are gaining popularity worldwide as a potentially energy-efficient air-conditioning technology for buildings. However, the control of ACB system is challenging, as it needs to handle multiple cooling coils and the relatively slow response to cooling load dynamics. This paper reports the implementation of a model predictive control (MPC) system for an ACB system, which employs a linear white-box building model for building energy and indoor condition predictions. A multiple-objectives function is employed in the MPC controller to optimize the energy efficiency in air-conditioning system and indoor thermal comfort while fulfilling the constraints of indoor comfort range (−0.5

The concept of ‘effective’ thermal resistance could facilitate in-depth understanding of the impact of passive substrate properties such as surface radiative and thermo-physical (which are not directly measurable using instrumentations in terms of R-value). A simple to-use and concise single performance factor has been formulated in this study to comprehend the effective thermal resistance provided by the enhanced surface radiative and thermo-physical properties of passive envelope materials. The derived expression is validated against measurements in real residential apartments located in Singapore. The derived effective thermal resistance expression is function of solar radiative properties, thermo-physical properties and weather parameters, and hence contains much more information than the traditionally estimated R-value. The effective thermal resistance is found to be dynamic in behaviour i.e., thermal resistance (or heat flow character) of the envelope material varies with transient weather conditions. Increasing roof surface radiative properties i.e., solar reflectance (from 0.1 to 0.8) alone has advantages during both daytime and nighttime with daily integrated-heat gain reduction by 60-68%. Whereas increasing the other thermo-physical properties of the envelope i.e., adding insulation or thermal mass (with a layer of phase change material-modified skim coat) has advantage only during daytime, but penalty during nighttime for the hot climates. The effect of increasing the solar reflectance by 0.7 for an insulated gray aluminum metal roof (with 20-mm polystyrene) is almost equivalent to effectively further adding 40-mm thick polystyrene. The application of proposed approach has been demonstrated by investigating the effect of passive envelope properties for different roof assemblies under four different climates. Using this approach, the accuracy of estimation of heat flux through roof, an indicator of the roof thermal efficiency, which was found to have improved by up to 78% against commonly found any steady-state method of heat flux estimation.

Abstract A model predictive control system with adaptive machine-learning-based building models for building automation and control applications is proposed. The system features an adaptive machine-learning-based building dynamics modelling scheme that updates the building model regularly using online building operation data through a dynamic artificial neural network with a nonlinear autoregressive exogenous structure. The system also employs a multi-objective function that could optimize both energy efficiency and indoor thermal comfort, two often contradicting demands. The proposed model predictive control system is implemented to control the air-conditioning and mechanical ventilation systems in two single-zone testbeds, an office and a lecture theatre, located in Singapore for experimental evaluation of its control performance. The model predictive control system is compared against the original reactive control system (thermostat in the office and building management system in the lecture theatre) in each testbed. The model predictive control system reduces 58.5% cooling thermal energy consumption in the office and 36.7% cooling electricity consumption in the lecture theatre, as compared to their respective original control. Meanwhile, the indoor thermal comfort in both testbeds is also greatly improved by the model predictive control system. Developing a model predictive control system using machine-learning-based building dynamics models could largely cut down the model construction time to days as compared to its counterpart using physics-based models, which usually take months to construct. However, the machine-learning-based modelling approach could be challenged by lack of building operational data necessary for model training in case of model predictive control development before the building has become operational.