Power Consumption In Buildings Research Articles

Estimating Spatio-Temporal Building Power Consumption Based on Graph Convolution Network Method

Buildings are responsible for around 30% and 42% of the consumed energy at the global and European levels, respectively. Accurate building power consumption estimation is crucial for resource saving. This research investigates the combination of graph convolutional networks (GCNs) and long short-term memory networks (LSTMs) to analyze power building consumption, thereby focusing on predictive modeling. Specifically, by structuring graphs based on Pearson’s correlation and Euclidean distance methods, GCNs are employed to discern intricate spatial dependencies, and LSTM is used for temporal dependencies. The proposed models are applied to data from a multistory, multizone educational building, and they are then compared with baseline machine learning, deep learning, and statistical models. The performance of all models is evaluated using metrics such as the mean absolute error (MAE), mean squared error (MSE), R-squared (R2), and the coefficient of variation of the root mean squared error (CV(RMSE)). Among the proposed computation models, one of the Euclidean-based models consistently achieved the lowest MAE and MSE values, thus indicating superior prediction accuracy. The suggested methods seem promising and highlight the effectiveness of GCNs in improving accuracy and reliability in predicting power consumption. The results could be useful in the planning of building energy policies by engineers, as well as in the evaluation of the energy management of structures.

Predicting the electric power consumption of office buildings based on dynamic and static hybrid data analysis

Currently, building energy consumption accounts for a considerable proportion of the world's energy consumption, e.g., 35 % in China, which has become a key concern with its rising proportion. Accurate prediction of building hourly energy consumption is key to realizing green, low-carbon, and energy-saving modern buildings. However, most of the current prediction methods are only based on dynamic data from IoT (Internet of Things) systems that may contain a large number of outliers, resulting in lower accuracy and reliability. To solve this problem, with the help of rich data provided by BIM (Building Information Modelling), we propose an approach named dynamic and static hybrid data analysis (D&S-HDA) that can provide a solution with better prediction accuracy. Technically, the novelties of D&S-HDA are fourfold. Firstly, we establish the D&S-HDA framework to obtain more accurate prediction outcomes, where the electricity consumption prediction results based on dynamic data analysis are weighted with the estimation values based on static data analysis. Secondly, in the dimension of dynamic data analysis, we design an improved temporal convolutional networks (TCNs) for parallelly outputting dynamic data samples to make the prediction results more intuitive. Thirdly, in the dimension of static data analysis, a novel concept of building hourly power consumption coefficient(BHPCC) is proposed, and its calculation method using static data analysis is designed to estimate hourly electricity consumption. Finally, in order to validate the effectiveness of our D&S-HDA approach, we design multidimensional evaluation metrics and conduct multi-group comparative experiments under different conditions, including climate models, electricity usage behaviour and time scales. The experimental results reveal that the proposed D&S-HDA approach outperforms current mainstream works in terms of root mean square error (RMSE), mean absolute error (MAE) and mean absolute percentage error (MAPE), with values of 1.5248, 1.0693 and 2.9505, respectively, which shows the efficiency and feasibility of the proposed D&S-HDA.

Power management of lighting loads combined with green building integrated photovoltaics as a solution for developing more sustainable and smarter cities

This study provides a strategy for managing lighting loads in conjunction with building-integrated photovoltaics (BIPVs) to control power and energy consumption in green buildings. The suggested approach increases the flexibility of lighting loads while considering energy production fluctuations of green buildings' integrated photovoltaics. This technique applies a mathematical model to define the relationship between the produced energy by building integrated photovoltaics and lighting demands. The control process is then performed by adjusting lighting loads based on the specified algorithm. Also, the suggested technique maintains minimum lighting levels and satisfies users' visual comfort. The proposed strategy can also be applied to a wide variety of demands in a building. By categorizing loads as fundamental and flexible, comparable patterns can be defined to control other flexible loads in a complex. Results from the system simulation confirm the precision of this technique. According to the results, for variations in BIPV production, the control system applies the necessary settings based on the defined algorithm and regulates energy consumption and lighting levels. This method is recommended to develop more sustainable and smarter cities.

A collaborative matching method for multi-energy supply systems in office buildings considering the random characteristics of electric vehicles

Multi-energy supply systems for office buildings have been rapidly developed. Although some combinations of distributed energy sources have been applied in office buildings, there is a lack of research on the collaborative matching of office building multi-energy supply systems that integrate random electric vehicles (EVs), stationary battery (STB), photovoltaic (PV), and the power grid (PG). Therefore, in this study, we propose a method for the collaborative matching of supply and demand for multi-energy systems integrated with random EVs, STB, PV, and PG in office buildings to minimize the net operating cost of the system. To this end, we first constructed an objective function, including system operating cost and operating revenue. We then modeled the supply and demand sides of the system. On the supply side, the travel parameters of EVs around the office building were investigated, and the randomness of EVs was quantified using the Monte Carlo algorithm. On this basis, the charging and discharging capacity of EVs was determined. Additionally, an STB charging and discharging model and PV power generation model were established. On the demand side, a prediction model of office building power consumption was constructed using the Random forest (RF) algorithm. Next, the energy supply and demand of office buildings for the following 24 h were matched collaboratively using nonlinear programming. Finally, the proposed collaborative matching method was applied to a real case study. The results showed that the matching method can reduce the net operating cost of the system by up to 92% compared with the Basic scenario.

Multi-Task Learning and Temporal-Fusion-Transformer-Based Forecasting of Building Power Consumption

Improving the accuracy of the forecasting of building power consumption is helpful in reducing commercial expenses and carbon emissions. However, challenges such as the shortage of training data and the absence of efficient models are the main obstacles in this field. To address these issues, this work introduces a model named MTLTFT, combining multi-task learning (MTL) with the temporal fusion transformer (TFT). The MTL approach is utilized to maximize the effectiveness of the limited data by introducing multiple related forecasting tasks. This method enhances the learning process by enabling the model to learn shared representations across different tasks, although the physical number of data remains unchanged. The TFT component, which is optimized for feature learning, is integrated to further improve the model’s performance. Based on a dataset from a large exposition building in Hangzhou, we conducted several forecasting experiments. The results demonstrate that MTLTFT outperforms most baseline methods (such as LSTM, GRU, N-HiTS) in terms of Root Mean Squared Error (RMSE) and Mean Absolute Percentage Error (MAPE), suggesting that MTLTFT is a promising approach for the forecasting of building power consumption and other similar tasks.

WITHDRAWN: Analysis on high-efficiency chiller plant's characteristics

Building's power consumption counts a large ratio of the social total energy cost and chiller plants cost a large amount of power for public buildings. With the globe's carbon neutralization desire becoming more important, high-efficiency chiller plant (HECP) is the first thing for the design and management of building's central air condition system. With this goal in sight, this study analyses the chiller plant's efficiency which is composed by single chiller based on the system's view and presents the relative importance of 8 variables on the chiller plant's efficiency. To simplify the analysis, the concept of pump's head in power is introduced. HECP's constrains on chiller's COP, chilled and cooling water temperature differences and pump's head in power are presented. The energy consumption ratio of HECP's every component is also studied. The theoretical study results were verified by the operation data of one hotel building's chiller plant located in Chongqing municipality. The research results by this paper are significant for the design and operation of HECP and are further beneficial for the global carbon neutralization.

Analyzing Long-Term and High Instantaneous Power Consumption of Buildings from Smart Meter Big Data with Deep Learning and Knowledge Graph Techniques

In the context of the growing emphasis on energy conservation and carbon reduction, the widespread deployment of smart meters in residential and commercial buildings is instrumental in promoting electricity savings. In Taiwan, local governments are actively promoting the installation of smart meters, empowering residents to monitor their electricity consumption and detect abnormal usage patterns, thus mitigating the risk of electrical fires. This safety-oriented approach is a significant driver behind the adoption of smart meters. However, the analysis of the substantial data generated by these meters necessitates pre-processing to address anomalies. Presently, these data primarily serve billing calculations or the extraction of power-saving patterns through big data analytics. To address these challenges, this study proposes a comprehensive approach that integrates a relational database for storing electricity consumption data with knowledge graphs. This integrated method effectively addresses data scarcity at various time scales and identifies prolonged periods of excessive electricity consumption, enabling timely alerts to residents for specific appliance shutdowns. Deep learning techniques are employed to analyze historical consumption data and real-time smart meter readings, with the goal of identifying and mitigating hazardous usage behavior, consequently reducing the risk of electrical fires. The research includes numerical values and text-based predictions for a comprehensive evaluation, utilizing data from ten Taiwanese households in 2022. The anticipated outcome is an improvement in household electrical safety and enhanced energy efficiency.

A comparative techno-economic assessment of bidirectional heavy duty and light duty plug-in electric vehicles operation: A case study

The proliferation of electric vehicles (EVs) all around the world offers both challenges and opportunities to build a sustainable city and transportation system. Bidirectional charging capabilities at workplace charging facilities (e.g., as part of a microgrid) have made the overall economic optimization more attractive on one hand, but also more complex on the other hand. This paper investigates the cost optimization problem for bidirectional charging at a workplace microgrid connected to two different buildings to determine the optimal framework for a combination of both heavy-duty and light-duty electric vehicles (HDEV and LDEV). A deep learning-based model has been developed to forecast the 15-minute solar generation and building power consumption. Real-time travel profile data has been used to represent the temporal uncertainty of electric vehicle charging. The cost optimization problem is formulated as a Mixed Integer Programming (MIP) model which also addresses battery life degradation. Furthermore, a comprehensive economic analysis has been carried out to analyze the payback period, peak reduction, and cost savings for two different buildings at the same workplace with both on-board and off-board charger configurations. It has been found that HDEV is a better cost-effective solution in comparison to LDEV in terms of energy cost reduction and payback periods. Net metering capability leads to higher energy savings and peak reductions in most cases.

Electricity Consumption Optimization Using Thermal and Battery Energy Storage Systems in Buildings

Energy storage system (ESS) plays a key role in peak load shaving to minimize power consumption of buildings in peak hours. This paper proposes a novel energy management unit (EMU) to define an optimal operation schedule of ESSs by employing metaheuristic and mathematical optimization approaches. The proposed EMU uses a thermal energy storage system (TESS) and a battery energy storage system (BESS) to store the energy in off-peak periods and discharge it in high load demands. We formulate the charging/discharging schedule of TESS and BESS as an optimization problem. Then, particle swarm optimization (PSO) is employed to obtain the optimal schedule due to its computational time efficiency. The mathematical approach is also applied to prove the convexity of the problem and the uniqueness of the solution. Due to the different characteristics of the building loads, this paper divides the total load into shiftable and fixed loads. Moreover, to model the building components and loads, grey-box modeling is adopted. Results show that employing a combination of TESS and BESS achieves peak load shaving while reducing 42.2% of the required BESS capacity compared with the case where the BESS is only used. In addition, the results indicate the effectiveness and robustness of the proposed algorithm.

Analysing the Residential Electricity Consumption using Smart Meter

A massive amount of electricity usage may be accessed on an everyday and hourly basis due to the advancement of smart power measuring technology. Electricity demand management and utility load management are made easier by energy usage forecasts. The majority of earlier studies have concentrated on the power consumption of business clients or residential buildings, or they have experimented with individual household electricity usage using behavioral and occupant sensor information. This study used smart meters to examine energy usage at a single household level to enhance residential energy services and gather knowledge for developing demand response strategies.The power usage of various appliances in a single household is estimated, by utilizing Autoregressive Integrated Moving Average (ARIMA) modeling technique, which is applied to daily, weekly, and monthly information granularity. To select the household’s energy consumption dataset for this study, a multivariate time-series dataset describing the four-year electricity usage of a household is provided. The use of Exploratory Data Analysis (EDA) is utilizedfor the selection of features and data visualization. The correlation coefficients with the daily usage of the household have been computed for the characteristics prepared for the forecast. The top three major determinants with the top three positive significance are "temperature," "hour of the day," and "peak index." A single household's usage is inversely related to the variables having negative coefficients. It should be noticed that the correlations among a household's attributes with usage vary from one another. Finally, the power prediction is analyzed in a single household.

Analysis of the Minimum Electrical Load of Residential Apartment Buildings in Urban Power Supply Systems

At present, a significant part of the studies of a residential apartment building electrical load concerns only the analysis of the maximum load at the residential building power supply input and on the buses of the transformer substation supplying power to residential buildings. However, for a complete analysis of the residential apartment building power consumption, it is necessary to evaluate not only the maximum input power, but also the minimum load. The article presents the results from an analysis of electricity consumption in residential apartment buildings with electric stoves (taking the city of Moscow as an example). The power consumption in eight residential apartment buildings recorded by electric meters in the input switchgears of these buildings at 30-min intervals for several years was used as input data for the analysis. The dynamics of minimum electrical load variation during a year is considered; the dependence of the minimum power on the outdoor temperature is estimated, and the time of the day, day of the week, and month in which the residential building minimum electrical load is observed are determined. The distribution law of a random variable represented by the time in which the absolute annual minimum of the residential building electrical load is observed and its characteristics: mathematical expectation and standard deviation are determined. The results of a comparative analysis of the absolute annual minimum and maximum (design) load at the input of residential apartment buildings are presented.

An outlier management framework for building performance data and its application to the power consumption data of building energy systems in non-residential buildings

The existence of outliers and noise is one of the major causes of poor quality of the building performance data. Methods to address such abnormal data have received significant attention from several research interests including data-driven fault detection, diagnosis, and estimation (FDD&E), performance optimization, predictive modeling, pattern recognition of building power consumption, etc. However, there are many factors that may eventually lead to unreliable FDD&E results or data cleaning effects. For instance, the misconception of the statistical definition of outliers, the confusion about the concepts of outliers and noise, and the neglect of the potential impact of the outlier processing way on the study results, etc. This study gives an interpretation to the statistical definition of outliers using the domain knowledge in building energy systems. The outliers in the building performance data are classified into three categories. Based on such classification, an outlier management framework is developed to achieve reliable outlier detection results and accurate outlier estimates. In Case study 1, under the proposed outlier management framework, about 81.2%–88.6% of the power consumption observations can be identified as outliers or normal observations. In terms of the outlier estimation accuracy, the RMSE of the chiller/chilled water pumps/cooling water pumps power outlier estimation results can be limited to less than 5.2, 0.5 and 0.6 kWh. The results of Case study 1 show that the proposed outlier management framework is generic. The framework can help reducing unnecessary monitoring cost of the HVAC system. In Case study 2, proper outlier management method helps improve the chiller power prediction results. The RMSE of the Basic group, Method 1 group and Method 2 group power prediction results are 0.55, 0.54 and 0.2 kWh. The results of Case study 2 show that the proposed outlier management framework can contribute to improve the results reliability of building performance analysis related studies. In this study, an outlier management framework is developed for the building performance data. The proposed framework aims to improve the quality of the data cleaning work. In the following research, we would dedicate to improve the efficiency of the data cleaning work. In other word, to develop specific online outlier management algorithms for specific studies under the proposed framework.

Toward an Intelligent Campus: IoT Platform for Remote Monitoring and Control of Smart Buildings

With the growing need to obtain information about power consumption in buildings, it is necessary to investigate how to collect, store, and visualize such information using low-cost solutions. Currently, the available building management solutions are expensive and challenging to support small and medium-sized buildings. Unfortunately, not all buildings are intelligent, making it difficult to obtain such data from energy measurement devices and appliances or access such information. The internet of things (IoT) opens new opportunities to support real-time monitoring and control to achieve future smart buildings. This work proposes an IoT platform for remote monitoring and control of smart buildings, which consists of four-layer architecture: power layer, data acquisition layer, communication network layer, and application layer. The proposed platform allows data collection for energy consumption, data storage, and visualization. Various sensor nodes and measurement devices are considered to collect information on energy use from different building spaces. The proposed solution has been designed, implemented, and tested on a university campus considering three scenarios: an office, a classroom, and a laboratory. This work provides a guideline for future implementation of intelligent buildings using low-cost open-source solutions to enable building automation, minimize power consumption costs, and guarantee end-user comfort.

Dual-Band Electrochromic Optical Modulation Improved by a Precise Control of Lithium Content in Li4+xTi5O12.

Dual-band electrochromic smart windows that can dynamically and independently control incident solar irradiation and visible light are envisioned as intelligent technology to reduce power consumption of buildings. However, there is still a great challenge to put the dual-band electrochromic technology into practice due to some limits in material systems and preparation techniques. Herein, a new electrochromic material of Li4Ti5O12 is developed to implement the dual-band optical modulation behavior, which could be further improved by a precise control of the lithium content in the active material. It could separately modulate the light and heat based on regulation of the transmittance of visible and near-infrared light. This enables Li4Ti5O12 to operate in three distinct modes of bright, cool, and dark, so as to meet various indoor needs. The optical transmittance contrast reaches over 60% at both visible- and near-infrared-light regions between different modes, and a large range of apparent temperature adjustments (7 °C) could be achieved. The prototype device based on dual-band electrochromic Li4Ti5O12 is further developed into a smart window of a house model, which exhibits good optical and thermal modulation behaviors in response to a high-temperature environment. This work provides a new material system for achieving dual-band electrochromic optical modulation toward smart energy-saving window applications.

Genetic Algorithm for the Optimization of a Building Power Consumption Prediction Model

Accurately predicting power consumption is essential to ensure a safe power supply. Various technologies have been studied to predict power consumption, but the prediction of power consumption using deep learning models has been quite successful. However, in order to predict power consumption by utilizing deep learning models, it is necessary to find an appropriate set of hyper-parameters. This introduces the problem of complexity and wide search areas. The power consumption field should be accurately predicted in various distributed areas. To this end, a customized consumption prediction deep learning model is needed, which is essential for optimizing the hyper-parameters that are suitable for the environment. However, typical deep learning model users lack the knowledge needed to find the optimal values of parameters. To solve this problem, we propose a method for finding the optimal values of parameters for learning. In addition, the layer parameters of deep learning models are optimized by applying genetic algorithms. In this paper, we propose a hyper-parameter optimization method that solves the time and cost problems that depend on existing methods or experiences. We derive a hyper-parameter optimization plan that solves the existing method or experience-dependent time and cost problems. As a result, the RNN model achieved a 30% and 21% better mean squared error and mean absolute error, respectively, than did the arbitrary deep learning model, and the LSTM model was able to achieve 9% and 5% higher performance.

Study on occupant behaviour using air conditioning of high-rise residential buildings in hot summer and cold winter zone in China

Over the past two decades, the electricity consumption of residential air conditioning (AC) has increased by 13.4 times. Moreover, carbon emissions from building operations remain high. Occupant behaviour related to AC operation is the most significant factor affecting the energy consumption of residential ACs. Hence, in this study, occupant behaviour towards using air conditioning in high-rise residential buildings was analysed to provide suggestions for increasing the AC usage efficiency. Annual data on the AC power consumption, indoor temperature, and humidity of typical high-rise residential buildings in hot summer and cold winter zones in China were collected. Five target parameters—AC opening rate, AC standby rate, start–stop time, start–stop temperature, and stable temperature—were statistically analysed. The results reveal that the AC start–stop time follows a normal distribution. The start and stop times of the AC during the cooling period are 22:32 and 7:05 the next day, respectively, whereas during the heating periods these values were 20:42 and 23:02 or 7:56 the next day, respectively. The standard deviation of the time distribution of AC starting or stopping at night is larger than that during the day. The AC start-up temperature followed an exponential distribution. The indoor temperature during the cooling and heating periods is more than 29.42 °C and less than 11.08 °C, respectively. The opening rate of the room AC is 80 %.

A Review on Certainty of load monitoring and analysis for Home Energy Systems

One important aspect of energy management is energy monitoring. Consequently, before planning some technical measures to reduce energy use, premises must be monitored for power consumption. Through Load Monitoring (LM), the most recent advancements in appliance energy management is presented in this paper. Various methods of Home Energy Management (HEM) using LM have been analyzed and categorized in an effort to investigate the most recent trend in energy management for researchers in the field. The researchers' various contributions have been highlighted, as have some methods for lowering a building's power consumption to save money and improve the environment. According to the findings of this study, there are problems with load management and monitoring that require attention; issues like the need for a surveillance system capable of that can recognize as many different types of loads as possible and more accurate recognition. Additionally, additional efforts are required to implement LM in appliance energy management. Last but not least, it is necessary to promote a culture of energy management among those who use electricity, whether in businesses, offices, or homes. The study will assist local researchers in gaining a clear understanding of the area's most recent trends.

Large-scale photovoltaic system in green building: MPPT control based on deep neural network and dynamic time-window

Under the background of smart city, the concepts of “green building” and “net-zero energy building” become more and more popular for reducing the building power consumption. As a result, the technologies related to the design and intelligent control of building integrated green energy system develop rapidly in recent years. In this study, the topological structure of large-scale building integrated photovoltaic (BIPV) system is analyzed, and a novel data-driven maximum power point tracking (MPPT) methodology is developed. To be specific, several characteristic-variables for achieving efficient MPPT of large-scale BIPV system are proposed, and the data-driven MPPT model based on deep neural network (DNN) is developed. Then, the developed characteristic-variables and DNN model are verified by a comprehensive set of numerical experiments. The optimal DNN structure is also verified in detail in this study. In addition, in order to dynamically track the degradation of photovoltaic module and overcome its influence on DNN model, the time-window mechanism of BIPV knowledge-base is introduced, and the optimal length of time-window for different DNN structures is verified by numerical experiments. Experimental results show that the DNN model with developed characteristic-variables and time-window mechanism achieves accurate and robust forecasting performance on the MPPT of large-scale BIPV system.

A Survey on Deep Learning for Building Load Forecasting

Energy consumption forecasting is essential for efficient resource management related to both economic and environmental benefits. Forecasting can be implemented through statistical analysis of historical data, application of Artificial Intelligence (AI) algorithms, physical models, and more, and focuses on two directions: the required load for a specific area, e.g., a city, and the required load for a building. Building power forecasting is challenging due to the frequent fluctuation of the required electricity and the complexity and alterability of each building’s energy behavior. This paper focuses on the application of Deep Learning (DL) methods to accurately predict building (residential, commercial, or multiple) power consumption, by utilizing the available historical big data. Research findings are compared to state-of-the-art statistical models and AI methods of the literature to comparatively evaluate their efficiency and justify their future application. The aim of this work is to review up-to-date proposed DL approaches, to highlight the current research status, and to point out emerging challenges and future potential directions. Research revealed a higher interest in residential building load forecasting covering 47.5% of the related literature from 2016 up to date, focusing on short-term forecasting horizon in 55% of the referenced papers. The latter was attributed to the lack of available public datasets for experimentation in different building types, since it was found that in the 48.2% of the related literature, the same historical data regarding residential buildings load consumption was used.

Dual-Band Electrochromism: Plasmonic and Polaronic Mechanisms

Electrochromic windows that dynamically absorb incident solar irradiation are envisioned as smart technology devices that can reduce power consumption in buildings and accelerate progress toward sustainable development. While regulating the visible light in interior spaces to suite the comfort of occupants, electrochromic windows can also offer thermal control by controlling the flux of solar energy, including infrared radiation. The ability to selectively modulate visible and infrared solar transmittance is denoted as dual-band electrochromism. Dual-band control can be achieved by employing materials that absorb light in different spectral regions when electrochemically charged, with light absorption due to either a polaronic or a plasmonic mechanism, depending on the charging conditions. In this Perspective, we discuss how adopting a plasmonic or a polaronic approach can affect the performance of an electrochromic device in terms of different quantitative figures-of-merit and the challenges and opportunities for designing better dual-band electrochromic devices in the future.