Applying support vector machines to predict building energy consumption in tropical region

The methodology to predict building energy consumption is increasingly important for building energy baseline model development and measurement and verification protocol (MVP). This paper presents support vector machines (SVM), a new neural network algorithm, to forecast building energy consumption in the tropical region. The objective of this paper is to examine the feasibility and applicability of SVM in building load forecasting area. Four commercial buildings in Singapore are selected randomly as case studies. Weather data including monthly mean outdoor dry-bulb temperature (T₀), relative humidity (RH) and global solar radiation (GSR) are taken as three input features. Mean monthly landlord utility bills are collected for developing and testing models. In addition, the performance of SVM with respect to two parameters, C and ɛ, was explored using stepwise searching method based on radial-basis function (RBF) kernel. Finally, all prediction results are found to have coefficients of variance (CV) less than 3% and percentage error (%error) within 4%.     

Sensitivity of the total heat loss coefficient determined by the energy signature approach to different time periods and gained energy

In order to identify buildings that have energy saving potential there is a need for further development of robust methods for evaluation of energy performance as well as reliable key energy indicators. To be able to evaluate a large database of buildings, the evaluation has to be founded on available data, since an in-depth analysis of each building would require large measurement efforts in terms of both parameters and time. In practice, data are usually available for consumed energy, water, and so on, namely consumption that the tenants or property holder has to pay for. In order to evaluate the energy saving potential and energy management, interesting key energy indicators are the total heat loss coefficient K_tot (W/K), the indoor temperature (T_i), and the utilisation of the available heat (solar radiation and electricity primarily used for purposes other than heating). The total heat loss coefficient, K_tot, is a measure of the heat lost through the building's envelope, whereas T_i and the gained energy reflect the user's behaviour and efficiency of the control system.In this study, a linear regression approach (energy signature) has been used to analyse data for 2003-2006 for nine fairly new multifamily buildings located in the Stockholm area, Sweden. The buildings are heated by district heating and the electricity used is for household equipment and the buildings’ technical systems. The data consist of monthly energy used for heating and outdoor temperature together with annual water use, and for some buildings data for household electricity are also available. For domestic hot water and electricity, monthly distributions have been assumed based on data from previous studies and energy companies. The impact on K_tot and T_i of the time period and assumed values for the utilised energy are investigated.The results show that the obtained value of K_tot is rather insensitive to the time period and utilised energy if the analysis is limited to October-March, the period of the year when the solar radiation in Sweden yields a minor contribution to heating. The results for the total heat loss coefficient were also compared to the calculations performed in the design stage; it was found that K_tot was on average 20% larger and that the contribution to heating from solar radiation was substantially lower than predicted. For the indoor temperature, however, the utilised energy had a large impact.With access to an estimate of K_tot and T_i, an improved evaluation of the energy performance may be achieved in the Swedish real estate market. At present the measure commonly used, despite the fact that monthly data is available, is the annual use of energy for space heating per square metre of area to let.     

The feedback of heat rejection to air conditioning load during the nighttime in subtropical climate

Taipei City, located in the subtropical zone, has a basin landform. The summer is always hot and humid and the air temperature right after sunset is typically higher than 30 °C. Heat rejection from residential buildings in urban area, equipped with lots of the window type air conditioners, not only increases the air temperature outside, but also burdens the cooling load. Based on the time schedule of air conditioner use of Taipei citizens, the heat rejection/building energy use and the air temperature distribution were evaluated, and finally the additional electric consumption of air conditioners was predicted. Two software, EnergyPlus (building energy program) and Windperfect (CFD, computational fluid dynamics software) were employed in this study. In the CFD simulation, the geometry of buildings that covers 700 m in diameter was created with GIS (geographical information system) and the total mesh number was more than 3 millions. Three specified temperatures (T_am, T_bu and T_ac) were used to describe the temperature distribution within the urban canopy by hourly time variation and spatial distribution with height and horizontal profile. The results revealed that the temperature gradually increased with height and the temperature next to the buildings was always higher than the ambient air. The feedback (penalty) of heat rejection to cooling load was found 10.7% during 19:01 to 02:00 h on the following day.     

Evaluation of adaptive thermal comfort models in moderate climates and their impact on energy use in office buildings

Thermally Activated Building Systems (TABS) are regarded as top-cooling systems rather than full air-conditioning systems. Therefore, adaptive thermal comfort models (ASHRAE55, ISSO74 or EN15251) are supposed to be applicable to TABS buildings, although the comfort model conditions are not necessarily satisfied. This paper investigates whether, for a moderate climate and with the heating and cooling set points chosen according to the adaptive models, the building’s energy use reduces. After all, applying adaptive models, if appropriate, is thought to lower energy use because higher maximum operative zone temperatures T_op,max are allowed, compared to the conventional ISO7730 model. For purpose of generality, a building with an ideal heating and cooling system is considered. Analysis of moderate climate weather data reveals a low energy reducing potential for the ASHRAE55 and ISSO74 model, because high reference outdoor temperatures hardly occur. EN15251 on the other hand, allows very high T_op,max and will lower the cooling need.A 2-zone building simulation demonstrates a higher cooling need for ASHRAE55 and ISSO74, compared to ISO7730. Because cooling is needed during the whole year, the lower winter T_op,max of these adaptive models causes these unexpected results. With real data of warmer years or varying gains, this conclusion holds.     

Future trends of building heating and cooling loads and energy consumption in different climates

Principal component analysis of dry-bulb temperature, wet-bulb temperature and global solar radiation was considered, and a new climatic index (principal component Z) determined for two emissions scenarios – low and medium forcing. Multi-year building energy simulations were conducted for generic air-conditioned office buildings in Harbin, Beijing, Shanghai, Kunming and Hong Kong, representing the five major architectural climates in China. Regression models were developed to correlate the simulated monthly heating and cooling loads and building energy use with the corresponding Z. The coefficient of determination (R²) was largely within 0.78–0.99, indicating strong correlation. A decreasing trend of heating load and an increasing trend of cooling load due to climate change in future years were observed. For low forcing, the overall impact on the total building energy use would vary from 4.2% reduction in severe cold Harbin (heating-dominated) in the north to 4.3% increase in subtropical Hong Kong (cooling-dominated) in the south. In Beijing and Shanghai where heating and cooling are both important, the average annual building energy use in 2001–2100 would only be about 0.8% and 0.7% higher than that in 1971–2000, respectively.     

Principal component analysis of electricity use in office buildings

Principal component analysis was conducted on five major climatic variables—dry-bulb temperature, wet-bulb temperature, global solar radiation, clearness index and wind speed. Twenty-eight year (1996–2000) long-term measured weather data were considered. A two-component solution was obtained, which could explain 80% of the variance in the original weather data. Monthly electricity consumption data recorded during a 5-year period (1979–2006) were gathered from 20 fully air-conditioned office buildings with centralised HVAC systems in subtropical Hong Kong. Electricity use per unit gross floor area ranged from 163 to 389 kWh/m². These consumption data were correlated with the corresponding principal components using linear multiple regression techniques. The coefficient of determination (R²) varied from 0.76 to 0.95 indicating reasonably strong correlation. It was found that the regression models developed could give a reasonably good indication (mostly within 3%) of the annual electricity use, but the monthly estimates might differ from the actual consumption by up to 9%. Attempt was also made to develop a general regression model for the 20 buildings, which had an R² of 0.84 with a maximum mean-biased error of 18.6% and a maximum root-mean-square error of 21.4%.     

Simulation of the urban thermal comfort in a high density tropical city: Analysis of the proposed urban construction rules for Dhaka, Bangladesh

The thermal comfort in urban canyons of a high density city is a very challenging issue for urban planners and designers, especially in hot humid tropical zone. The present study aims to evaluate the effects of a newly promulgated building construction rules in respect of thermal climate for Dhaka, Bangladesh, a region characterized by high density area with tropical climate. Three different urban canyons from three areas were considered which almost represents the whole city. Measurements were carried out to assess the existing thermal climate in the city canyons. Further, three model canyons were configured according to the new rules of ground coverage, floor area ratio (FAR) and site setback. A three-dimensional numerical model with high spatial and temporal resolution was used to investigate the microclimatic changes within the urban environments. Model calculations were run for a typical summer day. The thermal climate was evaluated based on air temperature T _a, surface temperature T _s, relative humidity RH, and wind speed in the existing and model canyons. For thermal comfort assessment the temperature humidity index (THI) was considered in this study. The results showed that the outdoor spaces of the study areas are thermally uncomfortable and the new building construction rules hardly improve the conditions in general. However, THI decreases in a canyon where sky view factor (SVF) decreases in the model canyon than the existing one. In addition, the surface temperature T _s and air temperature T _a do not show the same picture. Furthermore, the wind speed increases in all the model canyons as the site setback allows wind flow uniformly inside the canyons.     

Bin weather data for five climatic zones of Morocco

ABSTRACT

The Bin method is one of the well-known and accurate steady-state approach to predict heating and cooling energy consumption of buildings. The application of this method requires detailed ambient temperature bin data. In this paper, the dry-bulb temperature bin data for five climatic zones of Morocco was generated. It was calculated from ?12°C to 45°C with 3°C increments in six daily 4?h shifts (1–4, 5–8, 9–12, 13–16, 17–20 and 21–24?h) for Marrakech, Ifrane, Fez, Errachidia and Agadir. The bin data given in this paper may have the positive impact on building energy efficiency in Morocco.     

A methodology for a thermal energy building audit

The present paper introduces a new method for the certification of the energy consumption of a building recording its “energy behavior”. The method utilizes energy indices such as Index of Thermal Charge or Index of Energy Disposition to simulate the heat losses of the building and the heat flow because of the temperature difference (ΔT) from the inner to outer space.The present method and the algorithm that is implemented could be used as a part of a building energy audit or as a single audit method. Additionally it could be used for the inspection of the energy efficiency in public or municipal buildings. The forenamed method is currently under investigation by the present research team.     

An investigation of atmospheric turbidity of Thai sky

An investigation of atmospheric turbidity has been undertaken for tropical Thai sky. Values of turbidity indices, namely, Linke factor (T_L), Angstrom coefficient (β) and illuminance turbidity factor (T_il) are derived directly from measurements taken by pyrheliometer, Volz sunphotometer and beam illuminance meter. Monthly mean values and frequency of occurrence of the value of each turbidity index are used to characterize variations of atmospheric turbidity. Simple polynomial equations are developed for computing values of Linke factor and illuminance turbidity factor as functions of solar altitude angle. Using the values of Linke factor and illuminance turbidity factor obtained from the models developed, values of beam normal irradiance and illuminance can be calculated accurately under clear sky conditions. Values of daylight illuminance are useful for daylighting application that contributes to energy conservation for buildings. Knowledge of the size of beam normal irradiance is useful for calculation of cooling load in air-conditioning buildings in tropical climate.     

Passive cooling for air-conditioning energy savings with new radiative low-cost coatings

Passive cooling is considered as an alternative technology to avoid unwanted heat gains, to reduce urban heat islands and to generate cooling potential for buildings (limiting air-conditioning energy). According to materials and surface treatments, the roof can represent to be a major heat gain source from opaque elements of the building fabric, heating up the outer surface and increasing heat flow by conduction. This paper presents low-cost new radiative materials (1 ∉/m²) allowing to limit heat gains during diurnal cycle for hot seasons. To evaluate the relevance of these new substrates, their reflective UV-VIS-IR behavior are studied and compared to classical roofed materials available in industrial and developing countries. A 48 m² experimental roof having different surfaces (plate steel sheets, fiber cement, terra cotta tiles and corrugated sheets) allows to determine the temperature ratio δ between uncoated and coated materials. Up to 34% surface temperature gains are obtained for white coated CS, 25% for FC and ∼18% for TCT and PSS. According to uncoated materials for a surface temperature T₀ = 60 °C, simulations showed that the low-cost white opaque reflective roofs (50 m²) presented in this study would reduce cooling energy consumption by 26-49%.     

An analysis of bioclimatic zones and implications for design of outdoor built environments in Egypt

Climate considerations are essential dimensions in the assessment of quality of outdoor built environments. This paper provides an analysis of bioclimatic classification of Egypt to help the environmental design of wide range of purposes, including: climate responsive design; energy conservation and thermal comfort in the outdoor built environments. The analysis of this classification uses a bioclimatic approach in which the comfort zone and monthly climatic lines were determined and plotted on the psychrometric chart. Since the mean radiant temperature (T_mrt) is the most important input parameter for the energy balance in outdoor environments, the charts apply the ASHRAE 55-2004 standard considering the operative temperature as a function of T_mrt. Analysis for each bioclimatic zone determines the potential of passive design strategies to maintain thermal comfort in outdoor spaces and to contribute to energy efficient built environment. Finally, this study suggests a design guideline matrix for landscape architectural design for the different bioclimatic zones.     

Assessing dynamic efficiency of air curtain in reducing whole building annual energy usage

The efficiency of air curtain in reducing infiltration and associated energy usage is currently evaluated statically by using an efficiency factor, η _air, based on single steady/static condition, which is often not the case for actual buildings under variable weather conditions and door usages. Based on a new method to consider these dynamic effects on air curtains, this study uses a dynamic efficiency factor η _B in terms of whole building site end-use energy to assess the efficiency of air curtains when compared to single doors (i.e. without air curtains) and vestibule doors. Annual energy simulations were conducted for two reference building models considering their specific door usage schedules in 16 climate zone locations in the North America. The variations of the proposed efficiency factor for different climate zones illustrated the dynamic impacts of weather, building, unit fan energy and door usage frequency on air curtain efficiency. A sensitivity study was also conducted for the operation temperature conditions of air curtain and showed that η _B also considers these operational conditions. It was thus concluded that using whole building site end-use energy to calculate the efficiency factor, ηB, can provide more realistic estimates of the performance of air curtains operations in buildings than the existing static efficiency factor.     

Renovation of existing office buildings in regard to energy economy: An example from Ankara,Turkey

Unit energy consumption of existing buildings in Turkey is excessive. While average energy consumption of residential buildings in Europe is 100 kWh/m² per year, it is about 200 kWh/m² per year in Turkey. The principle reason for this, is that there was not any regulation on thermal insulation issues until recent years. However, the fiscal value of total energy consumption in residential buildings is about $2.5 billion. Recent research has shown that 40% of this energy consumption could be saved, provided that using energy efficiently. Furthermore, every reduction in energy-usage has a significant influence on environmental protection and CO₂ emissions. This study has focused on energy efficiency in a building of public sector that had been inaugurated in 1988 in Ankara. During the pre-investigative step, it has been determined that 47% of total energy consumption of the building could be saved.     

Parametric analysis of alternative energy conservation measures in an office building in hot and humid climate

The growth of demand for electrical energy in the rapidly expanding towns, cities and industries exceeds the growth of the power being made available. Therefore, energy conservation is becoming an increasingly important issue in Saudi buildings. The objective of this study was to investigate the impact of alternative energy conservation measures on energy requirements in office buildings in hot–humid climates. The study was conducted on a five-story office building located in Dammam, Saudi Arabia, which has been in use since 1998. Different types of HVAC systems were selected and different feasible and practical operational energy conservation measures (ECMs) were evaluated using the energy simulation software of Visual DOE 4.0. Previous studies conducted in this area were reviewed. Data was collected through review of design drawings, building audit and the analysis of 4 years of utility bills. All the collected data was analyzed and the utility bills data was used to calibrate the base case of the existing building using Visual DOE energy simulation software. Conclusions and recommendations were developed for conserving energy using various appropriate ECMs in office buildings in hot and humid climates.     

Methodology to estimate building energy consumption using EnergyPlus Benchmark Models

The evaluation of building energy consumption usually requires building energy profiles on an hourly basis. Computer simulations can be used to obtain this information but generating simulations requires a significant amount of experience, time, and effort to enter detailed building parameters. This paper presents a simple methodology to estimate hourly electrical and fuel energy consumption of a building by applying a series of predetermined coefficients to the monthly energy consumption data from electrical and fuel utility bills. The advantage of having predetermined coefficients is that it relieves the user from the burden of performing a detailed dynamic simulation of the building. The coefficients provided to the user are obtained by running EnergyPlus Benchmark Models simulations; thus, the simulation process is transparent to the user. The methodology has been applied to a hypothetical building placed both in Atlanta, GA, and in Meridian, MS, and in both cases, errors obtained for the estimated hourly energy consumption are mainly within 10%.     

Optimum insulation thicknesses for building walls with respect to cooling and heating degree-hours in the warmest zone of Turkey

Thermal insulation is one of the most effective energy conservation measures for cooling and heating in buildings. Therefore, determining and selecting the optimum thickness of insulation is the main subject of many engineering investigations. In this study, the determination of optimum insulation thickness on external walls of buildings is comparatively analyzed based on annual heating and cooling loads. The transmission loads, calculated by using measured long-term meteorological data for selected cities, are fed into an economic model (P₁−P₂ method) in order to determine the optimum insulation thickness. The degree-hours method that is the simplest and most intuitive way of estimating the annual energy consumption of a building is used in this study. The results show that the use of insulation in building walls with respect to cooling degree-hours is more significant for energy savings compared to heating degree-hours in Turkey's warmest zone. The optimum insulation thickness varies between 3.2 and 3.8 cm; the energy savings varies between 8.47 and 12.19 $/m²; and the payback period varies between 3.39 and 3.81 years depending on the cooling degree-hours. On the other hand, for heating load, insulation thickness varies between 1.6 and 2.7 cm, energy savings varies between 2.2 and 6.6 $/m², and payback periods vary between 4.15 and 5.47 years.     

Evaluation of four control strategies for building VAV air-conditioning systems

It is necessary to adopt appropriate control strategies to save energy and improve the indoor air quality (IAQ). On the validated TRNSYS simulation platform, four different control strategies are investigated to examine the indoor air temperature, energy consumption, CO₂ concentration and predicted mean vote (PMV) for the variable air volume (VAV) systems in an office building in Shanghai. As an original scheme, Strategy A using constant outdoor air intake fraction shows high energy consumption, low CO₂ concentration and acceptable thermal comfort. By using minimum outdoor air ventilation based on dynamic occupancy detection, Strategy B can provide more than 15% energy saving, acceptable PMV value but high CO₂ concentration in breathing zone. By using indoor air temperature reset, Strategy C presents the most energy savings beyond 20% reduction, low CO₂ concentration but poor thermal comfort. In mild seasons, combining enthalpy-based outdoor airflow economizer cycle with supply air temperature reset, Strategy D can achieve 9.4% energy savings and the lowest CO₂ concentration. Taken together, each strategy covers some strengths as well as some weaknesses. How to comprehensively assess a control strategy for all specific objectives should be considered in future studies.     

Robustness of a methodology for estimating hourly energy consumption of buildings using monthly utility bills

The evaluation of building energy consumption under typical meteorological conditions requires building energy profiles on an hourly basis. Computer simulations can be used to obtain this information, but generating simulations requires a significant amount of experience, time, and effort to enter detailed building parameters. This paper considers a simple methodology for using existing EnergyPlus benchmark building energy profiles to estimate the energy profiles of buildings with similar characteristics to a given benchmark model. The method utilizes the building monthly energy bills to scale a given benchmark building energy profile to approximate the real building energy profile. In particular, this study examines the robustness of the methodology considered with respect to the parameter discrepancies between a given building and the corresponding EnergyPlus benchmark model used to estimate its profile. Test buildings are defined by perturbing several combinations of the parameters defined in the benchmark building model. The test buildings examined are similar to the EnergyPlus, medium sized office, benchmark building in Baltimore, MD, and a total of 72 distinct test building configurations are examined. The analysis reveals that the methodology can significantly reduce the errors introduced by discrepancies from the EnergyPlus benchmark model.     

气候变化对中国寒冷和夏热冬暖城市建筑能耗的影响

气候变暖已对建筑全生命周期的运行状况产生了不可忽略的影响,准确评估气候变化下的建筑能耗对建筑方案设计和既有建筑的节能改造具有重要意义。进行气候变化下建筑能耗的精确预测,必须拥有未来的逐时气象数据。以寒冷地区北京和夏热冬暖地区广州为研究对象,将挑选的两个城市典型气象年为基线气候,结合全球模式下的预测气象数据,应用变形法修正TMY的气象参数,得到直至本世纪末的10个节点年逐时气象文件,并进行了全年能耗模拟,预估了两个城市的办公建筑在气候变化下建筑能耗的变化趋势。结果表明:在两种预测排放情景下,干球温度、含湿量和太阳辐射均呈增加趋势;北京采暖能耗显著降低、制冷能耗增加,总能耗减少,广州采暖能耗降低、制冷能耗显著增加,总能耗增加。