Search

Building-related heating consumes 80% of final energy in the Swiss residential sector. Electrical resistance heating systems consumed 9.3 PJ and 7.9 PJ for space and domestic hot water heating respectively. This is 8.6% of the total electricity consumption in Switzerland. A successful retrofit action requires knowledge of energy consumed by these systems in the pre-retrofit stage. However, it is difficult to deduce the heating energy from the overall electrical load profile of each residence. One of the primary reasons is irregularities in the operation of the heating system. We address this by applying a data mining approach that group data points based on their concentration in the time axis. This is done by using DBSCAN (Density-based spatial clustering of applications with noise) which requires derivation of two clustering parameters, ‘Epsilon’ and ‘MinPts’. We present a method to automatically derive these two parameters based on a combination of histogram analysis and range search method. We also compare our results with those obtained by k-means and density peaks clustering methods. The proposed methodology can effectively disaggregate electrical load into space and domestic hot water heating for residences with an electrical resistance heating system.

Applied Energy, 294, ISSN:0306-2619, ISSN:1872-9118

This study aims to determine key building variables influencing energy consumption in air-conditioned office buildings. The study is based in Singapore which entails tropical climatic conditions. The analysis is based on assessment of several energy audit reports concerning pre- and post-retrofit data from 56 office buildings. A list of 14 building variables, extracted from these reports form the superset. These are systematically analyzed further to derive key variables influencing energy consumption and retrofitting decisions. For this purpose, a robust iterative process is developed utilizing k-means clustering. This process tests all combinations of the 14 variables against change in energy use intensity (EUI, measured as kWh/m2.year) for pre- and post-retrofit conditions. The results indicate that the best set of variables consists of: 1) gross floor area (GFA), 2) non-air-conditioning energy consumption, 3) average chiller plant efficiency, and 4) installed capacity of chillers. This information can be utilized to explore energy saving potential of office buildings that need to be retrofitted. The resultant clusters can also be used to benchmark buildings based on pre-retrofit conditions and energy saving potential.

Occupancy information is one of the crucial variables in modelling and predicting the energy use in buildings. However, the presence of occupants is often stochastic in nature. In the presented case of eight space typologies derived from three institutional building blocks, a substantial variation in the correlation between occupancy and energy consumption is found for different space types during the semester and semester breaks for various resolutions of day and time. Further, it has been identified that a weak correlation between occupants and energy use is due to the use of common plug and lighting loads such as office printers, projectors, lab instruments, and fluorescent lamps. However, in the spaces studied, such as offices and computer rooms, the control of plug and lighting loads can be at individual occupancy levels for avoiding energy wastages. To characterize occupancy and energy consumption patterns by different space types and occupant types, this study develops and presents an integrated, data-driven modelling framework and results for different space types (like classrooms, studios, computer rooms, office spaces and laboratories and time resolution (hourly to semester-long intervals) for the case study building. It is found that the Deep Neural Network (DNN) model exhibits a slightly better prediction accuracy than conventional machine learning/regression models such as gradient boosting, support-vector network, and feed-forward neural network. However, in terms of computation time, the gradient boosting model is found to be faster than the DNN model for comparable outcomes. The developed integrated model will better equip the building facility management for an occupant-oriented control of building systems for energy efficiency.

Clustering algorithms have been successfully applied in analyzing building energy consumption data. It has proven to be an effective technique to identify representative energy consumption patterns as well as being a pre-processing step for other techniques. In this paper, we propose a clustering method based on k-shape algorithm, which is a relatively novel method to identify shape patterns in time-series data. In the experiment, clustering is performed for each individual building according to its hourly consumption. The novelty of this paper is that a new k-shape algorithm is applied to detect building-energy usage patterns at different levels, and the clustering result is further utilized to improve the accuracy of forecasting models. Ten institutional buildings covering three different typologies are used as case studies and a set of hourly and weekly energy consumption data is further analyzed in this paper. The experimental results reveal that this proposed method can detect building energy usage patterns in different time granularity effectively and also proves that the forecasting accuracy of SVR model is significantly improved by utilizing the results of the proposed clustering method.

Electricity load forecasting is crucial for effective operation and management of buildings. Support Vector Regression (SVR) have been successfully used in solving nonlinear regression and time series problems related to building energy consumption forecasting. As the performance of SVR heavily depends on the selection of its parameters, differential evolution (DE) algorithm is employed in this study to solve this problem. The forecasting model is developed using weighted SVR models with nu-SVR and epsilon-SVR. The DE algorithm is again used to determine the weights corresponding to each model. A case of time series energy consumption data from an institutional building in Singapore is used to elucidate the performance of the proposed model. The proposed model can be used to forecast both, half-hourly and daily electricity consumption time series data for the same building. The results show that for half-hourly data, the model exhibits higher weight for nu-SVR, whereas for daily data, a higher weight for epsilon-SVR is observed. The mean absolute percentage error (MAPE) for daily energy consumption data is 5.843 and that for half-hourly energy consumption is 3.767 respectively. A detailed comparison with other evolutionary algorithms show that the proposed model yields higher accuracy for building energy consumption forecasting.

This paper presents a detailed investigation and analysis of the energy consumption characteristics of three institutional buildings in Singapore. Building information, energy consumption data of the air-conditioning system, and energy consumption data of the plug loads were collected separately. Identification models are developed to predict the real daily energy consumption data. Developed models involve three specific functions to represent the variability of the daily occupancy, the additional occupancy due to visitors and the variation of outdoor air temperature. The performance of the developed identification model is very satisfactory and fits very well with the real energy consumption data. Based on the identification model, key factors which are influencing the energy consumption in institutional buildings are identified as the variability of the daily occupancy, hence a methodology is developed to calculate the occupancy in the building. The identified parameters are used as inputs into deterministic energy simulation programs, like Energy Plus, to perform detailed energy analysis. The whole methodology developed has improved significantly the accuracy of the energy simulation modeling of institutional buildings and has permitted to understand and evaluate the major energy characteristics of these buildings.

This study presents a methodology to forecast diurnal cooling load energy consumption for institutional buildings using data driven techniques. The cases for three institutional buildings are examined. A detailed analysis on their energy consumption data for two years shows that there is a high variation in diurnal energy consumption. This is largely attributed to the university scheduling and vacation periods. To reduce the degree of variation, the energy consumption data is divided into classes. These class numbers are then taken as inputs for the forecasting model which is developed using Artificial Neural Networks (ANN). The results show that the ANN is able to train and forecast the next day energy use based on five previous days’ data with good accuracy. The model development, along with ANN architecture used in this case is discussed in detail. As a next step, the forecasted output is taken back as an input with a view to forecast the output of the following day. This step is repeated and the model exhibits an R2 of more than 0.94 in forecasting the energy consumption for the next 20 days. It is also noted that such a methodology can be positively extended to other institutional buildings.

The thermal performance gap in buildings is defined as the difference between the theoretical and the actual energy consumption for heating, and is known to undermine energy retrofit strategies and policies. This study examines the performance gap in retrofitted buildings using the Swiss Cantonal Energy Certificate for Buildings (CECB) database, using a sample of 1172 buildings for which both theoretical and actual metered consumption were known. We found an average negative performance gap of –23% for pre-retrofit buildings (actual consumption smaller than calculated) and instead a good approximation of actual consumption with theoretical consumption after retrofitting (a positive gap of 2%). A regression analysis on the energy performance certificate input parameters characterizing the building led to the conclusion that these are poor predictors of actual consumption compared to the theoretical calculation: parameters such as the energy label and the thermal proprieties of the envelope (U-values) have minor explanatory power for the actual consumption despite explaining a high degree of change in the theoretical consumption. Analysis of the indicator Energy Savings Deficit (ESD) shows an overestimation (of 37%) of the achievable savings on the basis of the theoretical consumption, whereas the prediction of savings using measured consumption before retrofit resulted in a good agreement with the actual savings (3.6% overestimation). This implies that energy savings can be estimated rather accurately by comparing the actual current consumption with the expected theoretical consumption defined by the certificate after retrofit., Energy and Buildings, 224, ISSN:0378-7788, ISSN:1872-6178

A high-quality building energy retrofit analysis requires knowledge of building characteristics like the type of installed heating system. This means auditing the building in person or conducting a detailed survey, which is not readily scalable for many buildings. This paper presents a data-driven methodology to identify building characteristics from raw smart meter data sets to allow large scale, high-quality building retrofit analysis. We use the concept of energy signatures, a scatter plot with outside air temperature on the x-axis and electricity consumption on the y-axis, which condenses each building’s electricity use into one highly informative graph. Using a Support-Vector Regression model we extract the shape of each signature and cluster them subsequently. Dynamic time warping is used to align the signature shapes of all buildings. In two case studies, consisting of smart meter data sets from 408 and 480 buildings respectively, we show that our clusters correlated well to the heating system type and the building type by comparing to building-level metadata or demographic data.

Air pollution and climate change increased the importance of renewable energy resources like solar energy in the last decades. Rack-mounted PhotoVoltaics (PV) and Building Integrated PhotoVoltaics (BIPV) are the most common photovoltaic systems which convert incident solar radiation on facade or surrounding area to electricity. In this paper the performance of different solar cell types is evaluated for the tropical weather of Singapore. As a case study, on-site measured data of PV systems implemented in a zero energy building in Singapore, is analyzed. Different types of PV systems (silicon wafer and thin film) have been installed on rooftop, facade, car park shelter, railing and etc. The impact of different solar cell generations, arrays environmental conditions (no shading, dappled shading, full shading), orientation (South, North, East or West facing) and inclination (between PV module and horizontal direction) is investigated on performance of modules. In the second stage of research, the whole PV systems in the case study are simulated in EnergyPlus energy simulation software with several PV performance models including Simple, Equivalent one-diode and Sandia. The predicted results by different models are compared with measured data and the validated model is used to provide simulation-based energy yield predictions for wide ranges of scenarios. It has been concluded that orientation of low-slope rooftop PV has negligible impact on annual energy yield but in case of PV external sunshade, east facade and panel slope of 30–40° are the most suitable location and inclination.