Electricity Consumption In Buildings Research Articles

An Empirical Evaluation of Deep Autoencoders for Anomaly Detection in the Electricity Consumption of Buildings

An Empirical Evaluation of Deep Autoencoders for Anomaly Detection in the Electricity Consumption of Buildings

Evaluation of household electricity consumption in multi-apartment buildings for optimization of rooftop PV systems

Empowering citizen energy communities (CECs) to produce, self-consume and share renewable energy can enhance household energy efficiency, support the adoption of renewable energy sources, reduce energy consumption and supply tariffs, and increase independence from fossil fuels. This study analyzes actual electricity consumption data from 31 dwellings in typical five-story multi-apartment buildings in Riga, Latvia, considering the potential for rooftop solar energy systems within CECs. The novelty of this study lies in the detailed analysis of household electricity consumption data and characteristics (e.g., household size, population, etc.) to predict household load which is essential for the optimal design of a multi-apartment building scale PV system. We evaluate the suitability of typical multi story buildings for rooftop photovoltaic (PV) systems, aiming to determine the potential PV performance relative to final electricity consumption. The study utilizes PolySun computer simulation software. Our findings indicate that a rooftop PV system on a typical 5 story multi-apartment building (60 apartments) can cover up to 77% of annual electricity consumption. This analysis is based on hourly real electricity consumption data. The research outcomes can inform optimization of on-site energy storage to mitigate the impact of PV systems on the grid. The proposed solution is applicable in many Eastern European countries and in some Western European countries that were influenced by the Soviet Union in the latter half of the 20th century.These regions have a similar stock of panel-constructed multi-apartment residential buildings and share comparable climatic conditions.

Statistical characterization of electricity use profile: Leveraging data analytics for stochastic simulation in a smart campus

On the path to energy transition, advanced metering infrastructures have been installed in distribution systems to support sustainability goals, generating a substantial volume of electricity consumption data that are essential for planning and management studies. Additionally, given the stochastic nature of electricity consumption, understanding and quantifying statistical properties such as data distribution, normality, stationarity, and autocorrelation are crucial for the development of more sustainable systems and the enhancement of building performance. In this context, this paper presents a statistical methodology for assessing key aspects of electricity consumption in buildings on a smart campus, which is an initiative originated on university campuses that integrates sustainable energy systems, efficient electrical infrastructure, and data-driven technologies to establish a sustainable learning environment. Using 28 months of electricity consumption data from a Brazilian smart campus, Electricity Use Profile models are developed and several hypothesis tests and probability distribution fittings are conducted to extract statistical features from the models of 128 buildings. The results indicate that each building exhibits unique statistical properties that cannot be generalized, emphasizing the need for data analysis for each building before using the data in decision-making processes.

Optimising Daylighting Performance Through Side light with Passive Devise Design in Buildings

Passive lighting design plays an important role in providing natural lighting to save electricity consumption in buildings. This study aims to investigate the performance of natural lighting and the potential of alternative designs through sidelights with 3 shading device models and light shelves with different sizes in north, west, east, and south orientations. This research method is quantitative, which describes the measurement results in existing conditions and radiance illuminance software simulations. The results showed that at the beginning of field measurements, it was known that there were rooms with lighting intensity too high, too low, and uneven distribution. The simulation results use Radiance Illuminance software to determine the illuminance value of light distribution into space in the morning, afternoon, and evening. Improvised designs are applied with shading device models on the building envelope, namely horizontal, vertical, and egg-crate models in spaces oriented toward the East and West. This study also applied an interior-exterior light self-model measuring 50 cm, 100 cm, and 150 cm of space with the orientation of window openings towards the North and South. The simulation results in this study, show that without a shading device and using a horizontal model, there will be a decrease of 0.4% at a distance of 2 m from window openings oriented to the East, while to the West will be able to increase the illuminance value by 11.5% at a distance of 10.5 m from the window opening. Buildings without using light self and using interior-exterior light shelves measuring 50 cm can increase the illuminance value by 0.5% at a distance of 10.5 m from window openings in a space oriented towards the North, while the orientation towards the south, which can be increased the illuminance value by 0.4% at a distance of 6 m from the window opening. Based on this research analysis, it concluded that the best-improvised design in rooms oriented toward North and South is a 50 cm interior-exterior light shelf and horizontal shading device models are the best for spaces oriented to the East and West. This study, also concluded that the orientation of the building, the passive device model, and the distance of the measuring point in the building envelope, affect the intensity and distribution of daylight entering the building.

Implementation and validation of optimal start control strategy for air conditioners and heat pumps

Commercial buildings are responsible for approximately 20 % of the total energy consumption and greenhouse gas emissions in the United States. Over 85 % of these buildings lack building automation systems, and many are small (<50,000 square feet), underserved, and use rooftop units (RTUs) for heating, ventilation, and air-conditioning needs. Because these buildings lack proper energy management systems, several operational deficiencies lead to excess energy consumption. Studies have shown that managing the RTUs’ heating and cooling set points, schedules, setbacks, and optimal start can result in a 20 % to 25 % reduction in electricity consumption in small commercial buildings. These buildings typically use fixed schedules to start the RTUs 60 to 120 min before occupancy begins, which results in excess energy consumption. This paper presents research that demonstrates and evaluates the performance of four optimal start methods, which utilize data-based modeling as a key element in facilitating adaptive control in response to time-varying inputs while requiring minimal sensor inputs. The evaluation found energy savings in two commercial buildings equipped with RTUs by periodically alternating four different optimal start models during the cooling and heating season. The resulting energy savings are positive for all models and range from 2 to 5 kWh/day/unit. The units on the east side of the building showed higher savings, while interior units showed greater variability in savings due to the differences in capacities and room sizes. Savings were considerably greater during the heating season compared to the cooling season. The performance of all four models on Mondays was poor; models suggested a shorter optimal start time, which resulted in relatively larger errors. The future work will look at using a different model for the days after weekends and holidays.

Beyond efficiency: Evaluation of the electricity saving potential of green residential buildings

While green building standards are widely promoted for their energy-saving potential, limited research has explored their actual impact on electricity consumption in residential buildings. This study addresses this gap, comparing certified green buildings to standard ones using a multifaceted approach that combines high-frequency smart meter data with comprehensive socio-demographic data and detailed descriptions of building properties. Metrics investigated included total household consumption, energy use intensity (EUI), and per capita consumption, in each case differentiating between overall appliance use and dedicated heating, ventilation and air conditioning (HVAC) consumption. While green buildings achieved a 16% reduction in EUI for HVAC compared to standard buildings and thus lowered peak electricity demand, total annual household electricity consumption remained similar. The variance in electricity consumption among households in the sample indicates that occupant behaviour is a significant factor. Counter-intuitively, the research found that occupants of green buildings did not exhibit greater environmental awareness or report more energy-efficient appliance use habits compared to those in standard buildings. It also highlights limitations of EUI as the primary metric for ’efficiency’, rather than total consumption. Furthermore, because consumption is normalized by floor area, this metric may in fact inadvertently incentivize larger apartments that might require even more energy overall. These findings demonstrate the need for a modified area-normalized EUI target incorporated in the building codes as well as a more comprehensive approach to develop complementary policy measures.

Gaussian-based plug load profile prediction in non-residential buildings archetype

This paper presents an innovative approach to addressing the prevalent challenge of simulation uncertainty in urban building energy modeling (UBEM), focusing on accurately determining occupant-related input parameters. Traditional UBEM methods typically rely on standard schedules to create archetype models, which often fail to reflect the variability observed in real-world scenarios. To overcome this limitation, this research introduces a novel framework for generating electricity use profiles in institutional building archetypes across various climate zones. This framework integrates k-means clustering with Gaussian processes, effectively incorporating uncertainties into the prediction models. The evaluation of this stochastic model suggests that the methodology can give acceptable predictions on the electricity consumption of institutional buildings. The model demonstrates robust predictive capabilities, achieving a CVRMSE as low as 11% on weekdays and 8.7% on weekends, reflecting its strong predictive performance. However, its performance varies among different clusters and time periods, with specific clusters displaying more significant predictive inaccuracies at particular times. These results emphasize the importance of fine-tuning models and offer opportunities for improvement in predicting urban building energy consumption. This can be achieved by incorporating sensor-derived data to develop more detailed building profiles that include variable electricity usage patterns. This methodology has been integrated into a UBEM tool, enabling the generation of more realistic electricity load profiles.

Predictive Analysis of a Building’s Power Consumption Based on Digital Twin Platforms

Colleges and universities are large consumers of energy, with a huge potential for building energy efficiency, and need to reduce energy consumption to build a low-carbon, energy-saving campus. Predicting the energy consumption of campus buildings can help to accurately manage the electricity consumption of buildings and reduce the energy consumption of buildings. However, the electricity consumption of a building’s operation is affected by many factors, and it is difficult to establish a model for analysis and prediction. Therefore, in this study, the training building of the BIM education center on campus was selected as the research object, and a digital twin O&amp;M platform was established by integrating IoT, digital twin technology (DDT), smart meter monitoring devices, and indoor environment monitoring devices. The O&amp;M management platform can monitor real-time changes in indoor power consumption data and environmental parameters, and organize data on multiple influencing factors and power consumption. Following training, validation, and testing, the machine learning models (back propagation neural network, support vector model, and multiple linear regression model) were assessed and compared for accuracy. Following the multiple linear regression and support vector models, the backpropagation neural network model exhibited the highest accuracy. Consistent with the actual power consumption detection results in the BIM education center, the backpropagation neural network model produced results. Consequently, the BP model created in this study demonstrated its dependability and ability to forecast campus building power usage, assisting the university in organizing its energy supply and creating a campus that prioritizes conservation.

Utilizing PVSyst for planning a hybrid system rooftop solar power plant at Makassar Eye Hospital

Installing rooftop Solar Power Plants (SPP) represents the most efficient solution for reducing substantial electricity consumption in office buildings and hospitals, as Makassar Eye Hospital exemplifies. The research commences by addressing concerns about electricity availability and daily usage at the hospital while evaluating the solar radiation potential. Subsequently, the development of the SPP system begins. A location assessment is conducted using various tools to gather the necessary data for SPP design. In this study, the planning phase involves the utilization of PVSyst software for the technical analysis of electrical energy generation and system performance calculations. Additionally, the economic analysis includes evaluating the initial investment costs for establishing the SPP and considering prevailing market prices for components. The feasibility of the SPP investment is determined by calculating key financial metrics, including the Payback Period (PP), Net Present Value (NPV), Internal Rate of Return (IRR), Profitability Index (PI), and Return on Investment (ROI). According to the economic analysis findings, the initial investment cost for the SPP amounts to Rp. 544.031.733. The energy production value per kilowatt-hour (kWh) stands at Rp. 1.278, resulting in a Payback Period (PP) of 13,8 years, a Net Present Value (NPV) of Rp. 138.182.638, an Internal Rate of Return (IRR) of 11.51%, and a Return on Investment (ROI) of 25.4%. These results affirm the plan's feasibility, as the IRR value surpasses the Discount Rate (8.43%).

Geospatial modelling of seasonal water and electricity consumption in Doha's residential buildings using multiscale geographically weighted regression (MGWR) and Bootstrap analysis

Ensuring sustainable water and electricity consumption in urban residential buildings is a growing challenge worldwide, particularly in rapidly developing regions with harsh climates. This study examines the seasonal variation of water and electricity consumption in residential buildings in Doha, Qatar, exploring the interconnectedness of land use/land cover (LULC) and socio-demographic characteristics with household water and electricity consumption. For this purpose, we employed statistical analysis (i.e. Pearson correlation and Bootstrap analysis) and advanced geostatistical models, including Geographically Weighted Regression (GWR) and Multiscale Geographically Weighted Regression (MGWR), to analyze and monitor the spatial and seasonal variations of water and electricity consumption. The methods involved assessing the relationship between land surface temperature (LST), household water-electricity consumption, and analyzing the impact of demographic variables. Key findings indicate significant spatiotemporal variations in consumption influenced by changes in LULC and demographic characteristics such as household size and structure. The findings highlight the need for integrated urban planning and energy policies that consider the impacts of LULC and demographic changes to enhance energy efficiency and sustainability in urban settings. Furthermore, the results underscore the importance of addressing the complex interplay between urban development and resource consumption in policy-making.

Utility-based context-aware multi-agent recommendation system for energy efficiency in residential buildings

A significant part of CO2 emissions is due to high electricity consumption in residential buildings. Using load shifting can help to improve the households’ energy efficiency. To nudge changes in energy consumption behavior, simple but powerful architectures are vital. This paper presents a novel algorithm of a recommendation system generating device usage recommendations and suggests a framework for evaluating its performance by analyzing potential energy cost savings. As a utility-based recommender system, it models user preferences depending on habitual device usage patterns, user availability, and device usage costs. As a context-aware system, it requires an external hourly electricity price signal and appliance-level energy consumption data. Due to a multi-agent architecture, it allows for easy integration of new agents, enabling seamless functionality expansion, or the disabling of existing agents to tailor the system to specific needs. Empirical results show that the system can provide energy cost savings of 18% and more for most studied households.

Analisis Jejak Karbon Dalam Proses Pembelajaran Kelas

The environmental issue should have already become a primary concern and a crucial focal point for sustainability. One of the influencing factors is carbon dioxide (CO2) emissions, which have a significant impact on climate change. Electricity consumption contributes the largest share of carbon emissions in the energy sector, followed by transportation activities. This study aims to compare the levels of carbon emissions generated and the costs incurred in the face-to-face and online learning processes involving electricity consumption and transportation activities. The research employs a qualitative approach with a case study method. The population in this study consists of active accounting students at Bunda Mulia University, Serpong Campus, and no sampling is conducted to obtain the entire data, which includes primary data from observation and questionnaire. The tools used in this study are Microsoft Excel and jejakkarbonku.id for processing research samples. The results of the study indicate that electricity consumption is higher in face-to-face learning compared to online learning. Additionally, a greater level of carbon emissions is produced from face-to-face learning than online learning, with building electricity consumption surpassing transportation activities. The costs incurred for face-to-face learning are also higher than those for online learning. Despite face-to-face learning being the largest contributor to consumption, carbon emissions, and costs, it is perceived as more effective than online learning. Therefore, there is a need for solutions to address environmental issues while maintaining the effectiveness of education.

Geometrical optimization of solar venetian blinds in residential buildings to improve the economic costs of the building and the visual comfort of the residents using the NSGA-II algorithm

The entering sunlight from the building's windows mainly affects the heating and visual comfort of the occupants. The applications of Venetian blinds are a solution to improve the heating and visual comfort of the occupants. However, reducing the sunlight that enters the space can result in a rise in the building's electricity consumption. While most studies focus on the electricity production of solar panels, present study aims to examine the effect of solar venetian blinds on the indoor visual and thermal comfort of the occupants and optimize their geometry considering different geographical specifications. In the present paper, efforts are made to numerically install solar panels on Venetian blinds and analyze the effect of changing the geometrical parameters of solar Venetian blinds and the building's window dimensions on visual comfort and net electricity. Therefore, the target functions in the present paper are an improvement percentage in the daylight glare index and an improvement percentage in the net electricity costs for the analyzed building. As a result, five cities in Iran that have different climatic conditions are targeted to model the building. EnergyPlus software is employed to conduct the energy-based calculations, and the design variables and target functions are defined using JEPLUS software. The outputs are next inserted in JEPLUS+EA software to process a multi-objective optimization using the NSGA-II algorithm. The results demonstrate that the visual comfort and net electricity can be optimized by ranges of 10–100% and 1.5–10%, respectively. Furthermore, Venetian blinds are proven to have higher reception of sun radiations and better efficiency in southern cities and they can have a more proper performance while being installed for windows of southern building wall.

Optimal load scheduling strategy for isolated building-integrated DC microgrid using binary PSO to maximize user comfort and minimize peak-to-average ratio

Due to the growth of electricity consumption in residential and commercial buildings, there is a growing burden on utilities in urban areas. Sustainable buildings are being developed by integrating renewable energy generation to make an eco-friendly environment. An isolated building-integrated (IBI) DC microgrid fed by a hybrid solar PV (photo voltaic) generator-thermo electric generator (TEG) is one viable option for obtaining self-sustainable energy for the building. The battery is an important component in such IBI DC microgrid fed by the hybrid solar PV generator-TEG for providing reliable power to the loads. However, the operation and planning of the IBI DC microgrid are challenging, and therefore it needs a suitable load scheduling strategy. This paper proposes the optimal load scheduling strategy (OLSS) under demand response to schedule the household loads in the building. The proposed OLSS aims to balance two conflictive objectives, user comfort and peak-to-average ratio (PAR), using a binary particle swarm optimization algorithm (BPSO). The proposed OLSS optimizes user comfort and PAR and reduces peak demand and battery usage. It also helps in the optimum utilization of power generated from hybrid solar PV generator-TEG by shifting the loads during sun hours. Reducing battery usage reduces the conversion loss and extends the battery life. To validate the proposed OLSS, the investigations are carried out using the data obtained across different seasons. The results show that the proposed OLSS schedules the household loads with the desired user comfort and reduces PAR by 22 % in future buildings.

Investigating the Impact of Data Normalization Methods on Predicting Electricity Consumption in a Building Using different Artificial Neural Network Models.

The study investigates the impact of data normalization on the prediction of electricity consumption in buildings using four multilayer Artificial Neural Networks (ANN) algorithms: Long Short-Term Memory Networks (LSTM), Levenberg-Marquardt Back-propagation (LMBP), Recurrent Neural Networks (RNN), and General Regression Neural Network (GRNN). Four data normalization approaches, Min-Max Scaling, Mean, Z-score, and Gaussian function were assessed on experimental datasets. The LSTM algorithm, when combined with Min-Max normalization, showed the most favorable predictive capabilities, with a low Coefficient of Variation of the Root Mean Square Error (CVRMSE) of 10.3 and Normalized Mean Bias Error (NMBE) of 0.6. The remaining three normalization approaches showed satisfactory concordance with empirical data, but with slight disparities in precision. The LMBP model, when using Z-score normalization, had favorable performance in forecasting electricity consumption, but the discrepancies across the models were not significant. The Recurrent Neural Network (RNN) model, when used with Gaussian normalization, exhibited the most favorable performance, with the lowest Coefficient of Variation of Root Mean Square Error (CVRMSE) at 11.8 and Normalized Mean Biased Error (NMBE) at 0.6. The Generalized Regression Neural Network (GRNN) model, trained on unprocessed data, exhibited superior performance, with the lowest Coefficient of Variation of Root Mean Square Error (CVRMSE) at 19.2 and NMBE at 1.0. In conclusion, the study highlights the significant influence of data normalization on the predictive capabilities of various ANN models, suggesting that careful use of data normalization techniques can significantly improve the accuracy of electricity consumption forecasting in buildings.

Electric Vehicle Charging Infrastructure Study for Apartment Buildings

Changes in the engine type of cars aimed at climate goals also have an impact on the habit of electricity consumption at home. Charging electric cars from the residential grid also increases the total electricity consumption of the residential building. If all apartment owners in an apartment building want to charge their electric car at the same time, this may create a situation where the main protection of the apartment building is activated. The purpose of this research is to identify possible options to avoid this problem, based on different real-life data. For example: adding an energy storage to the electrical system of an apartment building is seen as one solution. Also supplement of an energy management software to the electrical and battery bank system to enables residents to schedule and prioritize their electric car charging times to avoid simultaneous high-demand periods. This system can also be used to optimize electricity distribution and consumption in apartment buildings under volatile electricity price conditions.

Drivers and spatial patterns of carbon emissions from residential buildings: An empirical analysis of Fuzhou city (China)

Global economic development will continue to harm the natural environment. China is the largest contributor to global carbon emissions, with cities and the construction sector playing a major role. Understanding carbon emissions from buildings in cities can inform mitigation measures. This study conducted a comprehensive accounting of carbon emissions from lighting and appliance electricity consumption in residential buildings. The investigation covered Fuzhou city in Fujian Province of China in 2011–2021. ArcGIS generated and analyzed the spatial patterns of carbon emissions by districts. The STIRPAT model identified the leading carbon-emission drivers. The findings indicate: (1) fine-grained spatial distribution of carbon emissions demonstrates a marked concentration in the city center and contiguous districts, low emissions in surrounding districts, and a gradually increasing trajectory; and (2) GDP per capita, urbanization rate, and resident population are the principal drivers of carbon emissions, with every 1 % increase in GDP per capita raising carbon emissions by 0.66 %. Relevant practical energy-saving and emission-reduction measures are distilled from the results. The findings provide a scientific basis for decision-makers to formulate emission-reduction targets and strategies for Fuzhou's residential buildings, a theoretical basis for promoting regional low-carbon development, and a reference for other regions with a similar developing economy.

School Electricity Consumption in a Small Island Country: The Case of Fiji

Electricity consumption in buildings is one of the major causes of energy usage and knowledge of this can help building owners and users increase energy efficiency and conservation efforts. For Pacific Island countries, building electricity demand data is not readily accessible or available for constructing models to predict electricity demand. This paper starts to fill this gap by studying the case of schools in Fiji. The aim of the paper is to assess the factors affecting electricity demand for grid-connected Fijian schools and use this assessment to build mathematical models (multiple linear regression (MLR) and artificial neural network (ANN)) to predict electricity consumption. The average grid-connected electricity demand in kWh/year was 1411 for early childhood education schools, 5403 for primary schools, and 23,895 for secondary schools. For predicting electricity demand (ED) for all grid-connected schools, the stepwise MLR model shows that taking logarithm transformations on both the dependent variable and independent variables (number of students, lights, and air conditioning systems) yields statistically significant independent variables with an R2 value of 73.3% and RMSE of 0.2248. To improve the predicting performance, ANN models were constructed on both the natural form of variables and transformed variables. The optimum ANN model had an R2 value of 95.3% and an RMSE of 59.4 kWh/year. The findings of this study can assist schools in putting measures in place to reduce their electricity demand, associated costs, and carbon footprint, as well as help government ministries make better-informed policies.

Evaluation of the Effects of Window Films on the Indoor Environment and Air-Conditioning Electricity Consumption of Buildings

The objective of this study was to evaluate the effects of window films on indoor environmental conditions and electricity consumption of air conditioning. The research focused on the performance of different window films (HAG, RG), taking into account variations from different building orientations. The findings of this research indicated that building orientation could significantly influence the duration of direct sunlight entering the interior, with the areas closer to the glass being more susceptible to the effects of outdoor temperature and solar radiation. The clear glass with heat-absorbing film (HAG) and reflective film (RG) both reduced the indoor temperature and indoor illuminance while increasing indoor comfort. The RG could accumulate less heat on the glass surface compared with the HAG. The glass temperature of the RG will be lower than the HAG. The electricity-saving ratios of the HAG were 1.4%, 1.9%, 1.4%, and 1.2%, respectively, when facing the east, south, west, and northwest orientations compared with the clear glass (OG). The electricity-saving ratios of the RG were 3%, 4.2%, 4.2%, and 10.3%, respectively.

Evaluating the impact of the COVID-19 pandemic on the geospatial distribution of buildings' carbon footprints associated with electricity consumption

The carbon footprint (CF) linked to electricity consumption in buildings has become a significant environmental issue because of its significant role in greenhouse gas emissions. This study seeks to assess and examine the CF of electricity consumption in buildings across various building types. Additionally, this paper aims to investigate the impact of the COVID-19 pandemic on the CF of buildings. The investigation involves a comparative analysis between the CF values observed and predicted during the years affected by the pandemic. Additionally, the study evaluates the influence of the pandemic on the accuracy of CF model predictions by employing three distinct machine-learning models. Spatial analyses were conducted to identify clustering patterns of CF and identify areas of both high and low CF concentrations within the study area. The findings demonstrate significant disparities in the CF of electricity consumption across distinct building types, with residential buildings emerging as the largest contributors to carbon emissions. Moreover, the pandemic has had a notable impact on CF patterns, leading to alterations in the areas identified as hotspots and cold spots during the pandemic years compared to the pre-pandemic period, based on building types.