Search

[Display omitted] • The state of art of wet substrate materials is presented. • Five materials were tested and compared to Kraft paper as wet substrate. • Absorption, evaporation and moisture-wicking capacities were determined according to norm NF ISO 20158, NF EN ISO 9073-6 and AATCC 198. • Taking into account the cost, our results suggest that Fibre fabrics (tight straight weave) is a good choice for evaporative cooling Pad purpose. Due to the ever-growing demand for air-conditioning to bring the indoor air temperature to a comfortable level despite the unreasonable electricity consumption, research is more oriented towards techniques in connection with new methods enabling more energy savings and less adverse environmental impacts. Dew point evaporative cooling systems hold among the most promising. Their ability to use the evaporation of water to absorb heat from the air in order to lower its temperature below its wet bulb level without adding humidity, is an extremely low-energy and eco-friendly cooling principle. The properties of the material used in the construction of the wet channel surface in an evaporative cooler, i.e, its moisture wicking ability, diffusivity and evaporation capacity, can alter cooling efficiency and performance considerably. The irregular fibres help to divert moisture and enlarge the wet area, thus promoting evaporation. A measuring device was mounted to test the capillary rise of evaporative fabrics, which is one of the techniques used to assess the liquid moisture transmission performance of fabrics. A variety of materials (fabrics) woven from different types of fibres have been experimentally tested and compared to Kraft paper, which is commonly used as a wet surface medium in evaporative cooling systems. The wicking rate of fabrics obtained from the test device correlated well with the NF EN ISO 9073-6. Consequently, the test device clearly demonstrated differences between the fabrics used in the study, and can be used to determine vertical wicking behaviour of fabrics. Most textile fabrics have been found to have superior properties and qualities in terms of moisture wicking ability than the Kraft Paper. Also, that the type of weaving and the compactness of the fabric improved the capillary rise. Compared to Kraft paper, the absorbency of some fabrics has been found 160% to 355% higher. An estimation evaluation regarding both moistures transfer and mechanical properties revealed that two of the fabrics were the most suitable for evaporative cooling applications. [ABSTRACT FROM AUTHOR]

Several studies have demonstrated the capability of data-driven modelling based on on-site measurements to characterise the thermal performance of building envelopes. Currently, such methods include steady-state and dynamic heating experiments and have mainly been applied to scale models and unoccupied test buildings. Nonetheless, it is proposed to upscale these concepts to characterise the thermal performance of in-use buildings based on on-board monitoring (OBM) devices which gather long-term operational data (e.g., room temperatures, gas and electricity consumption...). It remains, however, to be proven whether in-use data could be a cost-effective, practical and reliable alternative for the dedicated tests whose more intrusive measurements require on-site inspections. Furthermore, it is presently unclear what the optimal experimental design of the OBM would be and which data analysis methods would be adequate. This paper presents a first step in bridging this knowledge gap, by using on-board monitoring data to characterise the overall heat loss coefficient (HLC) [W/K] of an occupied, well-insulated single-family house in the UK. With the aid of a detailed building physical framework and specifically selected data subsets a sensitivity analysis is carried out to analyse the impact of the measurement set-up, the duration of the measurement campaign and the applied data analysis method. Although the exact HLC of the building is unknown and no absolute errors could hence be calculated, this paper provides a new understanding of the decisions that have to be made during the process from design of experiment to data analysis. It is demonstrated that such judgements can lead to differences in the mean HLC estimate of up to 89.5%. [ABSTRACT FROM AUTHOR]

This paper analyzes electricity use in buildings and explores key research questions related to assessing, rating, labeling, and benchmarking building energy use. The analysis reviews annual Energy Use Intensity (EUI) data from 3060 fully electric-powered buildings across six categories: residential, educational, offices, hospitals, hotels and restaurants, and wholesale and retail trade services. One-fifth of the analyzed buildings is derived from an energy dataset that offers a higher time-scale resolution, providing more detailed insights into energy usage patterns. The analysis employs various factors, including the seasonal variation factor (SVF), daily variation factor (DVF), monthly factor (M f), and hourly factor (H f), to gain a detailed understanding of actual energy usage at different time scales. The findings indicate that electricity use is less sensitive to seasonal variations than other energy carriers that provide heating and cooling. The paper provides hourly and monthly peak load percentages (PPL), which can indirectly assist in assessing hourly schedules for building energy performance simulations. The monthly energy use pattern varies between building categories, with high-energy months alternating between winter and summer. For non-residential buildings, high-energy hours correspond directly to the daytime, while for residential buildings, there is a shift towards the evening, and the daily amplitude is lower. Furthermore, the analysis of daily and monthly variability of heating energy carriers, even if non-electric, sheds light on the implications of electrification for these loads. • The study evaluates annual electricity data from 3060 fully electric-powered buildings. • Median EUIs are meaningful energy performance benchmarks by building category. • Residential and educational buildings exhibit EUIs of 90-120 kW h/m2. • Offices and hospitals demonstrate EUIs of 160-170 kW h/m2. • Monthly and hourly factors establish energy benchmarks for finer time resolution. [ABSTRACT FROM AUTHOR]

The accuracy of power load forecasting plays an important role in the development of the economy and the promotion of energy consumption and transformation. Aiming at the strong fluctuation and variability mechanism of electricity consumption, this paper proposes a novel hybrid method with the Long Short-term Memory (LSTM) on the dual optimization of particle swarm algorithm and butterfly algorithm, namely the EWT-MOLSTM-SVR model. Adaptive signal decomposition Empirical Wavelet Transform (EWT) decomposes the fluctuating electricity data into different modes, and uses the frequency domain analysis of signal analysis to classify and forecast. LSTM alone reveals that the advantage of the data variability mechanism is not significant. Particle Swarm Optimization (PSO) is used for law control, and Butterfly Optimization Algorithm (BOA) is used for parameter promotion to jointly optimize the expression effect. The PSO optimizes the LSTM by improving the convergence speed and global search ability of the BOA, and the optimal hyperparameters could be searched for forecast so that the accuracy of the forecast was improved. This paper adopts the data of American residential electricity data for empirical analysis. The experimental results show that the model is significantly better than Recurrent Neural Network (RNN), General Regression Neural Network (GRNN), PSO-SVM, and other models, which has a good effect on improving the forecast accuracy. The hybrid model reveals the fluctuation mechanism and variability of electricity data, deeply analyzes the economic behavior of electricity, and provides a new method for exploring residential energy consumption and optimizing residential buildings. [ABSTRACT FROM AUTHOR]

[Display omitted] • Regions where feed-in tariff higher than valley price should follow dual objective optimization. • The economy of the household PV-BES-EV system mainly depends on the purchase cost of EVs. • Decreasing electricity price will not undermine the effectiveness of the household PV-BES-EV system. • The energy bills and carbon emissions decrease significantly with the optimal scheduling schemes. The household PV-BES-EV system (photovoltaic - battery energy storage - electric vehicle system) connected to the grid can guarantee the electricity demand of the household and facilitate the flexible trade with the grid to achieve the goals of cutting energy bills and reducing carbon emissions. In related issues, past studies often assume that renewable energy is both economical and environmentally friendly, and thus regarded it as the primary power source. While in fact, in some regions where the feed-in tariff is higher than the sales price at a certain time, households may be driven by arbitrage opportunities to frequently purchase electricity from the grid at the certain time to gain benefits, resulting in far greater carbon emissions than expected within the household measurement boundary. To avoid the environmental burden caused by pure economic incentives and to provide practical and sustainable solutions consistent with various energy price policies, the optimal scheduling schemes of the household PV-BES-EV system that considers both economic and environmental incentives for different scenarios solved by the genetic algorithm is proposed in this paper. Household data in Shanghai, China and Hamburg, Germany were selected for modelling, representing the dual-objective optimization scenario with arbitrage opportunities and the single-objective optimization scenario with sales price higher than feed-in tariff, respectively. Results show that for urban and rural residents in Shanghai, the energy bill and carbon emission would decline about 105%, 48% and 88%, 62%, respectively. As for Hamburg, the accordingly results would be 69% and 64%. In addition, based on the cost-effectiveness of the system in long-term, this paper puts forward policy recommendations that align with local conditions to further improve the economy of the system. [ABSTRACT FROM AUTHOR]

This paper examines the extent to which different household socio-demographic and building characteristics contribute to high energy use in Social Housing. Using data from the sample created in the framework of the EU funded ICE-WISH project (social housing pilots in ten European countries, of around 25 dwellings each), an odds ratio analysis was performed to examine the effect of socio-demographic and building's structural determinants on high residential energy consumption. The analysis is performed separately for electricity use for lighting and appliances and for energy consumption for space and water heating. Results indicate that high electricity consumption in social housing is related to both socio-demographic characteristics of the occupants and characteristics of the buildings in which they live (OR = 5.6 for households with three occupants compared to single households and OR = 3 for households living in houses greater than 80 m 2 ), whereas high energy consumption for space and water heating is only related to building characteristics (OR = 1.15 for households with three occupants compared to single households and OR = 4.77 for households living in houses greater than 80 m 2 ). The outcomes of this paper call for future energy policy to intelligently incorporate traditional and behavior-based energy efficiency interventions for reducing residential energy consumption and CO2 emissions. [ABSTRACT FROM AUTHOR]

Non-Intrusive Load Monitoring (NILM) is the task of determining the appliances individual contributions to the aggregate power consumption by using a set of electrical parameters measured at a single metering point. NILM allows to provide detailed consumption information to the users, that induces them to modify their habits towards a wiser use of the electrical energy. This paper proposes a NILM algorithm based on the Deep Neural Networks. In particular, the NILM task is treated as a noise reduction problem addressed by using denoising autoencoder (dAE) architecture, i.e., a neural network trained to reconstruct a signal from its noisy version. This architecture has been initially proposed by Kelly and Knottenbelt (2015), and here is extended and improved by conducting a detailed study on the topology of the network, and by intelligently recombining the disaggregated output with a median filter. An additional contribution of this paper is an exhaustive comparative evaluation conducted with respect to one of the reference work in the field of Hidden Markov Models (HMM) for NILM, i.e., the Additive Factorial Approximate Maximum a Posteriori (AFAMAP) algorithm. The experiments have been conducted on the AMPds, UK-DALE, and REDD datasets in seen and unseen scenarios both in presence and in absence of noise. In order to be able to evaluate AFAMAP in presence of noise, an HMM model representing the noise contribution has been introduced. The results showed that the dAE approach outperforms the AFAMAP algorithm both in seen and unseen condition, and that it exhibits a significant robustness in presence of noise. [ABSTRACT FROM AUTHOR]

The work presented in this paper relates to a small scale district heating network heated by a gas-fired CHP. In most common situations, such a CHP is heat driven operated, meaning that the CHP will switch on whenever heat is needed, while not taking into account the demand of electricity at that time. In this paper however, an active control strategy is developed, aiming to maximize the profit of the CHP, selling its electricity on the spot market. The CHP will therefore switch on at moments of high electricity prices. Nevertheless, since there never is a perfect match between the demand of heat and the demand of electricity, thermal energy storage is included in the network to overcome this difference. Here, three different storage concepts are compared: (1) a central buffer tank next to the CHP; (2) small storage vessels distributed over the different connected buildings; and (3) the use of the thermal mass of the buildings as storage capacity. Besides the development of the control algorithms based on model predictive control, a simulation model of the network is described to evaluate the performance of the different storage concept during a representative winter week. The results show that the presented control algorithm can significantly influence the heat demand profile of the connected buildings. As a result, active control of the CHP can drastically increase the profit of the CHP. The concept with the distributed buffers gives the best results, whereas the profit for the thermal mass concept is only marginally smaller. Since in this latter case no significant investment costs are needed, the conclusion for this case study is that the use of thermal mass of buildings for demand side management in district heating systems is very promising. [ABSTRACT FROM AUTHOR]

In Australia, heatwaves are the deadliest natural hazard and a major driver of peak electricity demand. The disproportionately high peak demand increases electricity prices, causes occasional blackouts and exacerbates energy poverty, all of which limit one’s ability to use air conditioning. Meanwhile, increased energy efficiency of dwellings may decrease their heat stress resistance. This paper challenges whether the current Australian Nationwide Energy Rating Scheme encourages heat stress resistance. Cooling consumption, peak cooling demand and the risk of indoor overheating were assessed for a typical single-storey home in Adelaide and Sydney. Design options between 6 and 8 stars, plus two traditional, energy-inefficient homes were simulated with the AccuRate energy simulation program. This paper demonstrates that the NatHERS does not directly encourage heat stress resistance in new homes and can even deliver buildings with worse heat resistance and increased reliance on air conditioning than traditional, energy-inefficient buildings. This overreliance on air conditioning can present a public health hazard, consequently, the integration of heat stress resistance in the Nationwide House Energy Rating Scheme is recommended. Recommendations for the analysis and assessment of heat stress resistance are provided. [ABSTRACT FROM AUTHOR]

As the crucial foundation of distributed energy management systems, a reliable prediction of dynamic energy consumption is urgently needed in the building sector. Advances in metering technology and machine learning methods provide both opportunities and challenges. In particular, a clustering-based hybrid model is established in this paper to predict the day-ahead hourly electric demand in hotel buildings, which were initially classified as the non-stationary operated buildings (NOB) due to the irregular electric temporal features. Specifically, with multi-dimensional attributes of 100 days considered by fuzzy c-means (FCM) clustering, similar days and hours were extracted stepwise into several homogeneous groups. Afterwards, the online modified predictor was established by the combination of support vector regression (SVR) and wavelet decomposition, taking the corresponding training samples extracted by FCM as the input. Clustered results of similar days testified the clear identification of stationary subsequences (Group IV) and non-stationary subsequences (Group I). Based on the validation, it can be concluded that the introduction of days-clustering can always improve the accuracy of day-ahead prediction, while the recognition of similar hours is necessary only for non-stationary subsequences. Finally, this paper offers an online short-term prediction approach for model-based energy management, based on the limited and fluctuating historical records. [ABSTRACT FROM AUTHOR]

[Display omitted] • Present a comprehensive review of building electricity profile models. • Distinguish the differences of building electricity profile and prediction models. • Categorize the statistical and bottom-up models for electricity use profiles. • Investigate the validation metrics in accordance with the modeling methods. • Discuss the key future perspectives of building electricity profile analysis. The building sector contributes significantly to overall energy consumption and carbon emissions. Improving renewable energy utilization in buildings is of considerable importance for their carbon neutrality. Electricity is expected to become the major form of final energy consumption in buildings; as such, understanding the features of building electricity use profiles is pivotal for building energy system design and city-scale electricity distribution network planning. This study reviews the current status and future challenges of building electricity use profile models. The paper distinguishes between electricity use profile models and electricity use prediction models. Then it comprehensively reviews current research papers, focusing on four aspects, including the temporal and spatial distribution of datasets, the distinction between statistical and bottom-up modeling methods, mathematical and engineering domain validation metrics, and applications in building energy system design and energy policy making. Key perspectives for future research are discussed in the paper, including the temporal scales of models, the aggregation of electricity use profiles, occupant-driven environment-interactive model structures, as well as the fit-for-purpose model evaluation. This is intended to inform and inspire future research on building electricity use profiles, improve the in-depth understanding and quantitative descriptions of building electricity use, and support the design of future building systems and energy distribution networks. [ABSTRACT FROM AUTHOR]

• A new hybrid model, namely CEEMDAN-SE-LSTM is proposed. • The model allows for more accurate forecasting of electrical loads. • The model is significantly better than models like EEMD-LSTM, CEEMDAN-LSTM, etc. Ultra-short-term power load forecasting refers to the use of load and weather information from the prior few hours to forecast the load for the next hour, which is very important for power dispatch and the power spot market establishment. Based on time series decomposition-reconstruction modeling and neural network forecasting, this study constructed a CEEMDAN- 1 1 May 13, 2014, was chosen as the starting time because the meteorological data (e.g., temperature, rainfall, wind speed, etc.) obtained by the authors for Changsha City started from that time; and to highlight the cyclical nature of the electricity load demand, given the lack of meteorological data afterApril 2018, May 13, 2017, was chosen as the cut-off date in this paper. However, we believe that the starting and ending time does not affect the idea and method of model construction in this paper. SE-LSTM model and used it to forecast the ultra-short-term electricity load in Changsha, China, considering meteorological and holiday factors. The article first decomposed the power load data from May 13, 2014, to May 13, 2017, at 24 time points per day for three years to obtain six component series, and then reconstructed them into a two-component series based on the sample entropy analysis to reflect the fluctuation and trend characteristics of the power load. Then, the LSTM neural network model was used to predict and superimpose the reconstructed component series to obtain the final prediction results. It was found that the RMSE, MAE, and MAPE of the CEEMDAN-SE-LSTM model were 62.102, 47.490, and 1.649 %, respectively, which were significantly better than those of the ARMA, LSTM single-prediction, EEMD-LSTM, and CEEMDAN-LSTM models. This study greatly improves the accuracy of ultra-short-term power-load forecasting, provides support for ultra-short-term power dispatching in Changsha, and provides a reference for other cities to develop short-term and ultra-short-term power load forecasting models. [ABSTRACT FROM AUTHOR]

Energy efficiency standards for household appliances have been adopted in many countries to mitigate negative externalities, reduce emissions, and save energy. Using Chinese provincial-level panel data from 2008 to 2018, this paper builds a difference-in-difference model to examine the mechanism by which energy-efficiency upgrades in China's air conditioners affect urban household electricity usage. The results show that the upgrades have brought about an obvious reverse effect, with urban household electricity usage increasing rather than decreasing. Reducing the unit costs of energy service has enhanced residents' sensitivity to cooling demand. Driven by the substitution and income effects, urban residents may increase the intensity and frequency of their AC use. Additionally, this reverse effect is heterogeneously dependent on the family size and urban residents' age structure rather than their educational level. When families are larger and have more children, the reverse effect is more pronounced. This paper's findings provide useful recommendations for policy-makers to formulate policies that are synergistic with energy efficiency standards for household appliances. [ABSTRACT FROM AUTHOR]

• Propose a seq2point non-intrusive conforming decomposition model with a deep feature-guided attention mechanism. • Variable sliding windows are used to obtain strongly correlated power series according to the operating cycles of different appliances. • Using word embedding technique to transform regression problem into multi-classification problem to improve the accuracy of power decomposition. • Snapshot ensemble techniques are used to further improve the generalization ability and accuracy of the model. Non-intrusive load monitoring (NILM), as an important part of intelligent electricity consumption, improves the cognitive level of the load by analyzing the bus power in a specific range to obtain refined electricity consumption data at appliance level in that range, such as the number of loads, load categories, load operating status and corresponding energy consumption. For a long time, the research on NILM algorithm has been quite in depth, especially with the deep learning method being used in NILM, breakthrough progress has been made. In this paper, an integrated policy non-intrusive load monitoring technique based entirely on deep feature-guided attention mechanism is constructed based on sequence-to-point deep neural networks. Firstly, according to the operating cycle characteristics of different appliances, the data are segmented with different sliding windows and the integerized power values are mapped to high-dimensional feature vectors by word embedding, which transforms the value regression problem of decomposed power into a multiclassification problem to find the probability of decomposed power values at each discrete integer power value. Then the encoder composed of deep residual network is used for multi-layer feature extraction, which is used to obtain richer feature information, and the extracted feature quantities will be sent to the decoder through two paths respectively, one is mapped to the power feature space through word embedding as the input to the decoder; the other is sent to the decoder based entirely on the attention mechanism through convolution as the checked vector K and the content vector V. The decoder obtains the information with the highest relevance to the current moment by assigning weights to the before and after information, and thus decodes the final load decomposition result. Finally, this paper introduces a snapshot technique to obtain an inexpensive multi-model integration strategy, which improves the generalization ability of the model and the accuracy of the load decomposition. The effectiveness of the proposed method is verified by experiments on the REDD dataset and UK-DALE dataset. [ABSTRACT FROM AUTHOR]

This paper provides an analysis of the appliance ownership and use factors contributing to high electrical energy demand in UK homes. The data were collected during a large-scale, city-wide survey, carried out in Leicester, UK, in 2009–2010. Annual electricity consumption and appliance ownership and use were established for 183 dwellings and an odds ratio analysis used to identify the factors that led to high electricity consumption. Many of the appliance ownership and use factors have not previously been studied for the UK domestic sector. The results of this study should be of key interest to government policy makers and energy supply companies interested in the underlying drivers of the highly positively skewed distribution of UK domestic electricity use. The study identifies those appliances that could be targeted for technical improvements or subjected to campaigns to encourage more energy efficient use in order to reduce electricity consumption among high demand households. This paper builds on earlier work by the current authors which identified the households (socio-demographic and dwelling characteristics) most likely to be high electricity consumers. The current work provides the basis for advice and guidance to those households that would enable them to, over time, reduce their electricity use. [ABSTRACT FROM AUTHOR]

Energy consumption of lift systems is one key issue in modern high-rise buildings, drawing property owners' attention and accounting for 6% of total electricity consumption by end-use in Hong Kong. A Benchmarking Energy Performance Indicator (BEPI), , measured in J.kg−1.m−1 and proposed by the first author of this paper for lift installations is mentioned in the Technical Guidelines of the Building Energy Code , as one emerging good engineering practice. BEPI is a fair parameter for comparing energy performance of lifts irrespective of their brand, model, travel, speed, capacity and control. It considers energy performance due to both the motor drive efficiency and traffic control intelligence. However, one main obstacle to make it statutory is the lacking of a generic monitor able to retrieve three sets of real time data (normally unavailable on an existing lift) to compute the BEPI, namely energy consumption, instantaneous car load and traveling distance, and to collect ample data statistically reliable enough to determine an acceptable threshold value. This paper describes the development of a proof-of-concept prototype BEPI monitor by giving full technical details on the rationale and history of development, the hardware components, software framework, and updated numerical results of one implementation inside the headquarters of Electrical & Mechanical Services Department (EMSD) of HKSAR. Experiences in installing and calibrating those external sensors that make the whole system non-disturbing to normal lift operation, as well as the procedures in computing the BEPI, are discussed in greater details. It has been shown that the BEPI is traffic dependent but the weekly average of it is relatively stable, confirming it may be robust enough for benchmarking purpose. The solution described in this paper may help roll out the monitoring of BEPI for all lifts inside and outside Hong Kong. [ABSTRACT FROM AUTHOR]

Twenty percent of the total energy consumption in the Netherlands comes from household electricity consumption. This comes from household electric appliances whose number has grown in recent years. The paper explores the effect of smart meter introduction, appliance efficiency and consumer behaviour on reducing electricity consumption in the Netherlands. It does so by combining two perspectives: a sociotechnical approach and a bottom up simulation approach. The range of scenarios explored through simulation in the paper provides an understanding of the interplay between efficiency, smart meter diffusion and consumer behaviour. The results show their effect on electricity consumption and suggest that further effort is required to control and reduce it. Insights from the paper suggest that future studies should disaggregate with respect to a number of factors. [ABSTRACT FROM AUTHOR]

Electricity consumption of the Chinese building materials industry accounted for 8.4% of industrial and 6.2% of national electricity usage in 2011. The purpose of this paper is to estimate the future electricity intensity and conservation potential of the Chinese building materials industry. This paper adopts a cointegration method to establish a long-run equilibrium relationship between electricity intensity and factors including technology, power tariff, enterprise scale and value-added per worker. Electricity conservation potential in the Chinese building materials industry is predicted to be 90.5 billion kW h in 2020 under the moderate scenario, which is more than the total electricity consumption of Malaysia in 2007; and 150.9 billion kW h under the advanced scenario, which is more than the total electricity consumption of Saudi Arabia in 2005. Conserved electricity in building materials industry would account for 1.2% and 2.0% of national electricity consumption under the two scenarios, respectively. More importantly, we find that the electricity intensity gap between Chinese and the world's leading building materials industries could be significantly narrowed by 2020 if aggressive energy conservation policies were implemented. Finally, based on the results of our study, future policy priorities and directions are suggested. [ABSTRACT FROM AUTHOR]

• Operating time of SMEs is estimated using a SOM and hourly demand data. • The estimated operating time is used for hourly AC demand disaggregation. • The average errors of yearly AC demand estimation are better than 10%. • Hourly disaggregation highlights season and time period of AC demand peaks. Despite the growing penetration of smart meters and energy management systems in recent years, few tools have been proposed for energy demand data analysis and visualization to promote energy conservation in small- and medium-sized enterprises (SMEs). In this paper, we propose a new method consisting of operating time estimation and simplified disaggregation based only on the SME's hourly electricity demand data and actual outdoor temperature data. The method first estimates the operating time using a self-organizing map, which categorizes the hourly data as operating, half-operating, or non-operating time and eliminates manual parameter adjustment that requires user trial and error. Then, using the estimated operating time, our simplified disaggregation method estimates hourly air conditioning (AC) demand considering the variation in each office's temperature at which air conditioners are switched on. The method was evaluated using a public business energy management system dataset from Japan, and the average errors of yearly AC demand estimation were better than 10%. The outputs of the proposed method may provide information that is more relevant, reliable, and specific to each customer and efficiently promote energy conservation measures in SMEs. [ABSTRACT FROM AUTHOR]

• 24 design principles for energy feedback relevant within a local social network. • End-user consultation revealed empirical support for 15 principles. • There appears to be no 'one size fits all' type of energy feedback. • The general public needs to be educated around the concept of load shifting. Engaged households are needed for the future 'smart grids' to function, but it is difficult to engage residents in their household electricity consumption. Building on earlier research suggesting that a local social network could provide a suitable context for energy feedback, this paper examines how feedback presented in a social network should be designed in order to be better understood by the residents and encourage long-term engagement. A review of the literature on design principles and cases where they had been implemented identified 24 principles. A prototype feedback design adapted for a local social network was made, based on design principles deemed suitable for the screens designed. End-user feedback on the prototype was collected through a stakeholder consultation workshop. In the workshop discussions, 17 identified principles were mentioned and there was support for 15 of these, one was both supported and contested and two were contested. Based on comments and suggestions from end-users, a revised version of the feedback prototype was made for implementation in a pilot study in Sweden and Portugal. [ABSTRACT FROM AUTHOR]

Urban expansion, driven by population and economic growth, has been a major contributor to increased levels of energy consumption across the globe. Beirut, Lebanon's capital, is no exception in facing a surge in its demand for electricity as it expands. However, with frequent power outages, underpinning a largely problematic power sector, Beirut's demand for electricity is becoming a real hurdle that impedes the city's economic growth and development. This paper introduces a near-city-scale building energy model, BEirut Energy Model BEEM, which estimates the building stock's electricity consumption in two different districts in Beirut. The methodology uses rule-based expert data for an archetypal classification of the buildings based on their functions and periods of construction with their corresponding attributes including the number of floors, number of apartments, and bimonthly electricity consumption to generate a 3D model for 3630 buildings coupled to the hourly weather conditions and topographic map, which is then simulated in EnergyPlus. The predicted consumption of 2311 buildings is then calibrated with actual available metered data, to adapt the model to Beirut's occupancy and users' behaviors. Calibrated results are mapped to reveal the spatiotemporal distribution of energy peak demands which provide insights for future interventions. An analysis of the spatial distribution of electricity use demonstrates a spatial clustering that underlies urban energy demand which can be used for smart grid zoning. [ABSTRACT FROM AUTHOR]

• This paper presents a methodology to assist building managers detect and identify operating anomalies. • The data generated consist of three consumption sectors (water, gas and electricity). • The methodology employed is based on combining data mining with machine learning, in the aim of creating a model that allows defining every day's status (atypical vs. frequent). • The expert flow manager receives an alert and analyzes the situation, with the possibility of implementing actions to correct the problem should an anomaly be detected. Building facility managers are increasingly equipping their buildings with extensive sets of sensors. This article aims to develop an analysis decision-making methodology based on the production of statistical indicators. The tracking of such indicators allows detecting any systems performance problems. The automatic pinpointing of malfunctions can serve to activate alerts. Our approach focuses on the processing of data stemming from secondary schools managed by departmental services in the Pas-de-Calais, where 117 secondary school buildings have been instrumented with various sensors and supplying data since 2015. This article starts with a close-up on data mining for water, gas and electricity consumption. Data mining and machine learning methods, including the Clustering approach (K-Means), have been used to extract information from the measurements conducted in 2015 and 2016. This information is used to classify the 2017 measurements according to supervised approaches (SVM). The specificity of this work is to delve deeper into the analysis by combining into the same algorithm a set of various sensors related to both energy use and building occupancy. The data classification results have allowed highlighting "atypical" operations during the daytime, through interpreting data classification results in an effort to define the status of every day in year 2017. [ABSTRACT FROM AUTHOR]

Understanding the demand flexibility of city-scale buildings under different demand response (DR) strategies is crucial to balance the power supply and demand in the grid. This paper presented a case study to evaluate the electricity consumption reduction potential for buildings in Changsha under different DR strategies. First, sixty-six prototype energy models for twenty-two building types of three vintages were selected to represent 59,332 buildings in Changsha. The total electricity consumption of the building stock was obtained by aggregating the prototype electricity use intensity with their representative building area. Three DR strategies were formulated, including temperature reset, lighting control, and plug load control. Second, a toolkit, AutoBPS-DR, was developed to automatically integrate individual DR strategy as well as their combination into the baseline energy models. Five scenarios were analyzed, including the baseline case, three cases with a single DR strategy, and a case with three DR strategies combined. The results showed that compared to the baseline scenario, the electricity consumption reductions of the case buildings for a 3-h DR event under temperature reset, lighting control, plug load control, and combined control scenarios were 603.97 MWh (7.94%), 441.08 MWh (5.80%), 663.26 MWh (8.72%) and 1673.32 MWh (22.00%), respectively. In the plug load control scenarios, high-rise apartments made the largest contribution with a reduction of 134.15 MWh. Medium office buildings had the most reduction under the other three scenarios, with a reduction of 122.16 MWh for temperature reset, 107.90 MWh for lighting control, and 357.16 MWh for combination control. This study demonstrated a significant DR potential for buildings in Changsha to overcome the electricity shortage period. [ABSTRACT FROM AUTHOR]

With rising energy prices, it is becoming increasingly attractive for households to form an energy community (EC) and become an active member of the energy system. Therefore, the choice of the right sharing coefficient, which defines how the produced energy is shared, is important for the expected profits. The aim of this paper is to analyse and compare two existing sharing coefficients with two new sharing coefficients. Therefore, we define two EC setups, one with only residential members and one with residential and commercial members, and investigate the impact of the sharing coefficients by using a data driven simulation model. For the analysis, we use a Monte Carlo approach in combination with load profiles of EC members simulated via time series models to account for random variation in electricity demand. The applied methodology gives additional insights on the profitability and the distribution of savings. The analysis shows that sharing coefficients that benefit small consumers are essential to incentivise private members to form an EC with commercial members. It is shown that when such an EC is formed, higher savings, a higher degree of self-sufficiency and a more efficient use of the produced electricity can be achieved. [ABSTRACT FROM AUTHOR]

• ESS degradation cost based on battery lifetime is considered for power management of nanogrid with PV system. • Multi-objective optimization representing power management determines individual power consumption of schedulable electric appliances and ESS charging/discharging. • P2P power trading is performed between nanogrids based on cooperative game model. • ESS degradation cost is decreased by 32.75% by the proposed power management. This study presents a peer-to-peer (P2P) trading strategy for power management of nanogrids with renewable energy sources (RESs), taking battery lifetime of energy storage system (ESS) into consideration. Nanogrid requires more analytical power management as individual control of each electric appliance should be carried out, considering its rated power. With the increasing penetration of distributed energy resources, P2P trading can be applied to nanogrids equipped with RESs and ESSs. For the power management of a nanogrid, multi-objective optimization is used to minimize grid power consumption, electricity cost, and delay of using electric appliances in appliance scheduling. In this paper, a battery aging model or degradation cost model is considered in the multi-objective optimization to prolong the ESS lifetime. The electricity from RESs and ESSs used for P2P trading is determined to increase the profitability of nanogrids. Depending on the amount of electricity allocated to P2P trading, a nanogrid can be a seller or buyer in P2P trading. P2P trading is carried out based on the cooperative game theory to maximize the welfare of both sellers and buyers. As compared to power management involving P2P trading without considering ESS lifetime, the proposed one in consideration of ESS lifetime decreases the electricity cost of nanogrid by 2.86% on average, and the degradation cost of ESS by 32.75% on average. [ABSTRACT FROM AUTHOR]

Decarbonization goals in the United States electricity sector are increasing the levels of renewable energy generation in the electricity supply system, and are driving increased attention to building electrification, which will increase the magnitude and shift the timing of the electricity system peak. These changes are motivating new approaches to coordinate building electricity demand with low-carbon renewable generation, elevating the importance of demand flexibility (DF) in buildings and the need to quantify the temporal impacts of DF. In this paper, we first characterize the hourly predictive accuracy of six commonly used baseline models in an application context of quantifying building-level load shift. Our analysis revealed insights such as hours of the day (afternoons), periods of the week (weekends), and seasons (summer) that were predicted with more accuracy than other time periods. In addition, the analysis showed tendencies toward overprediction or underprediction of load. Secondly, we provide the first published investigation of baseline erosion from repeated dispatch of building load shifting. We observed that as the baseline period is pushed back further from the prediction day, the distribution of errors across baseline model predictions increases, with notable inflection points near the three-week erosion point for two of the three models. [ABSTRACT FROM AUTHOR]

Abstract: Voltage rise is one of the main factors which limits the capacity of Low Voltage (LV) network to accommodate more Renewable Energy (RE) sources. This paper proposes a robust and effective approach to coordinate customers’ resources and manage voltage rise in residential LV networks. PV is considered as the customer RE source. The suggested coordination approach in this paper includes both localized control strategy, based on local measurement, and distributed control strategy based on consensus algorithm. This approach can completely avoid maximum permissible voltage limit violation. A typical residential LV network is used as the case study where the simulated results are shown to verify the effectiveness of the proposed approach. [Copyright &y& Elsevier]

Abstract: A stochastic model is proposed for fluctuations in electricity demand that are associated with individual user's consumption choices. Electricity consumption is modeled as a function of social activities of consumers. The dynamics of these activities are modeled as a Markov chain. Markov models are simplified models that capture the stochasticity to the unmodeled dynamics typically attributed to white noise disturbances. Additional uncertainties are also accrued in the process of calibrating the transition rates of these chains from finite samples. In this paper, these uncertainties are accounted for by considering random transition matrices. Such formalism can also reflect the fluctuations in the environment in which the chain evolves. We also discuss a third interpretation where uncertain transitions, in a multiscale setting, reflect the fine-resolution information that is lost in the process of state aggregation. As numerical demonstration, we study the activity modeling of a heterogeneous population. Activity uncertainties are propagated onto the energy demand. Demand uncertainties, in turn, are propagated onto a global performance metric. Such uncertainty management framework bridges between the actual drivers of the energy consumption and the system health. Subsequent decisions can be robustly supported based on the results of the quantitative model proposed in this paper. [Copyright &y& Elsevier]

This paper empirically evaluates the extent of energy resilience achieved in a socially-deprived community in Oxford, through deployment of solar photovoltaic (PV) systems and smart batteries (internet enabled and controllable) across a cluster of 82 dwellings (households). The methodological approach comprised dwelling and household surveys, along with high frequency monitoring of household electricity consumption, solar PV generation, battery charge and discharge data. In the monitored households, average daily electricity consumption was found to be positively related with dwelling size, number of occupants and number of appliances used. Although 117 MWh of PV electricity was generated within a year across 74 dwellings, peak generation did not match peak consumption, demonstrating the need for battery storage. Home batteries were found to increase self-consumption of PV electricity and offset grid demand through discharge of stored PV electricity marginally at an average of 6%, depending on the size of the PV system, surplus PV electricity available and size of the battery. Aggregating solar generation and storage at a community level showed that peak grid electricity demand between 17:00 and 19:00 was reduced by 8% through the use of smart batteries across 74 dwellings. In future, a local energy sharing scheme could be developed, wherein not all dwellings would need to have solar PV systems, but rather have internet enabled batteries that could be monitored and controlled virtually. [ABSTRACT FROM AUTHOR]

Highlights • A quantitative comparison of time-use diary and electricity demand data is presented. • Household occupants report common energy-related activities relatively accurately. • Electricity demand data cannot be used to infer use of appliances with thermostats. • Digital activity reporting techniques are recommended for future time-use studies. Abstract Understanding demand flexibility in the residential sector depends on understanding the causal link between household occupants' activities and resulting electricity demand. Self-reported electricity use via time-use diaries is often used as a direct descriptor of occupants' activities and has been integrated into residential electricity demand simulation models. Conversely, smart meter electricity demand data is increasingly used to infer occupants' activities. Underlying both these approaches are a number of unverified assumptions about people's perceptions of their energy use, the accuracy with which they report these activities and the physical operation of electrical devices. This paper carries out a comparison between self-reported energy-related activities and monitored electricity demand in 15 households over a week-long time period, with focus on electric hot water cylinders and heat pumps as appliances with large potential for demand flexibility. This comparison quantifies the extent to which self-reported activity is a predictor of electricity demand and conversely, whether electricity demand can accurately identify occupant activity. Results show that, although there is significant variation across households, self-reported activity tends to be a reasonably good predictor of electricity demand. However, due to the intervention of thermostat-controlled devices, electricity demand is not a good indicator of occupant activity. [ABSTRACT FROM AUTHOR]

Abstract: Due to the scientific and economic growth of recent years, the on-going innovation in electrical products has enabled people to live more comfortably and conveniently. The standby power function is one such innovation. Unfortunately, as an increasing number of products are equipped with standby power, more energy is wasted and more carbon dioxide is emitted into the atmosphere. This paper combines a two-step proportional stratified sampling technique with systematic random sampling to select representative households in Taiwan. We adopted the measurement method of the International Electro technical Commission (IEC) in measuring the overall use of standby power in Taiwanese households. We then proposed a model for the estimation of standby power consumption, before calculating annual standby power consumption among all households in Taiwan. This paper could serve as a reference for the government and relevant authorities in the establishment of corresponding standards, providing suggestions for the minimization of standby power consumption and the unnecessary waste of energy. [Copyright &y& Elsevier]

A critical element of assessing a building's suitability for Demand Side Response (DSR) is understanding its turndown potential to ensure that DSR participation will be financially viable. While research has been undertaken on site level DSR estimation methods, there is currently no research that compares the outcomes of these methods. This paper compares four non-domestic energy estimation methods used for understanding the DSR potential of electrical appliances in a building to provide insights about uncertainty levels based on input requirements. Each method is deployed to estimate the DSR potential of HVAC chiller assets at two UK hotels over two years. The results show the methods have a range of error levels from the highest Mean Average Percentage Error (MAPE) of 159% to the lowest MAPE of 39%. The input requirements followed a general trend of more complex informational inputs resulting in lower error values. The outcomes of this research enable users to make informed decisions in selecting DSR estimation methods based on information availability and acceptable estimation error levels. [ABSTRACT FROM AUTHOR]

Abstract: This paper investigates the development of a low cost air cooling system used for small scale applications. Its operation is based on the rational and efficient use of energy available in conventional freezers, used for other purposes than air conditioning, to cool small areas. In this paper, the proposed air cooling system is applied to small kiosk prototype (called “periptero”), found very often in the streets of Greek cities and villages, used to sell goods to people. The proposed air cooling system is using the low temperature air inside the freezers to cool the air inside the kiosk and provide a better living environment for the owner, particularly during summer period. The basic principle of the system operation and the experimental installation will be described and the experimental results will be presented and discussed. [Copyright &y& Elsevier]

• Building codes require about 20 powered safety devices in new U.S. homes. • Health and safety devices consume about 1% of U.S. residential electricity. • Safety devices in the U.S. are typically exempted from efficiency regulations. • Per-unit consumption is small, but each home will have many units. • Home medical devices are a growing use of energy. Miscellaneous electrical loads (MELs) are responsible for a significant fraction of energy consumption in buildings. This paper addresses one category of MELs: Safety, Security, and Health Devices (SSHDs). Common SSHDs include: electrical life safety equipment, smoke alarms, radon mitigation fans, and home oxygen concentrators. The installation or use of these devices is dictated by building codes, health providers, insurance companies, and other entities—none of which would ordinarily consider energy efficiency a priority. For this project, the most important residential devices were explored in terms of their governing regulations, functions, technologies, and energy use. The power consumption of 41 life safety SSHDs were measured. Individual power consumption of life safety devices is very small but many are typically required in every home. Devices performing the same functions have wide ranges in power use. Opportunities for reducing SSHD energy use were investigated, the results are summarized, and recommendations are made for further actions that could be taken to reduce SSHD energy use. SSHDs identified in this paper are responsible for at least one percent of current U.S. residential electricity consumption, but that number will climb steadily as existing buildings are upgraded and new types of SSHDs appear. [ABSTRACT FROM AUTHOR]

[Display omitted] The school building sector has a pivotal role to play in the transition to a low carbon UK economy. School buildings are responsible for 15% of the country's public sector carbon emissions, with space heating currently making up the largest proportion of energy use and associated costs in schools. Children spend a large part of their waking life in school buildings. There is substantial evidence that poor indoor air quality and thermal discomfort can have detrimental impacts on the performance, wellbeing and health of schoolchildren and school staff. Maintaining high indoor environmental quality whilst reducing energy demand and carbon emissions in schools is challenging due to the unique operational characteristics of school environments, e.g. high and intermittent occupancy densities or changes in occupancy patterns throughout the year. Furthermore, existing data show that 81% of the school building stock in England was constructed before 1976. Challenges facing the ageing school building stock may be exacerbated in the context of ongoing and future climate change. In recent decades, building stock modelling has been widely used to quantify and evaluate the current and future energy and indoor environmental quality performance of large numbers of buildings at the neighbourhood, city, regional or national level. Building stock models commonly use building archetypes, which aim to represent the diversity of building stocks through frequently occurring building typologies. The aim of this paper is to introduce the Data dRiven Engine for Archetype Models of Schools (DREAMS), a novel, data-driven, archetype-based school building stock modelling framework. DREAMS enables the detailed representation of the school building stock in England through the statistical analysis of two large scale and highly detailed databases provided by the UK Government: (i) the Property Data Survey Programme (PDSP) from the Department for Education (DfE), and (ii) Display Energy Certificates (DEC). In this paper, the development of 168 building archetypes representing 9,551 primary schools in England is presented. The energy consumption of the English primary school building stock was modelled for a typical year under the current climate using the widely tested and applied building performance software EnergyPlus. For the purposes of modelling validation, the DREAMS space heating demand predictions were compared against average measured energy consumption of the schools that were represented by each archetype. It was demonstrated that the simulated fossil-thermal energy consumption of a typical primary school in England was only 7% higher than measured energy consumption (139 kWh/m2/y simulated, compared to 130 kWh/m2/y measured). The building stock model performs better at predicting the energy performance of naturally ventilated buildings, which constitute 97% of the stock, than that of mechanically ventilated ones. The framework has also shown capabilities in predicting energy consumption on a more localised scale. The London primary school building stock was examined as a case study. School building stock modelling frameworks such as DREAMS can be powerful tools that aid decision-makers to quantify and evaluate the impact of a wide range of building stock-level policies, energy efficiency interventions and climate change scenarios on school energy and indoor environmental performance. [ABSTRACT FROM AUTHOR]

This paper introduces short-term load forecasting (STLF) using Generative Adversarial Networks (GAN). STLF was explored using several Artificial Intelligence based methods that offered excellent results. However, the usage of GAN models in this field is very limited, and usually works on creating synthetic load profiles to increase load datasets. This paper investigates the application of GAN for load forecasting by generating daily load profiles. Predicting the daily load is a challenging task that requires accurate and stable models that can capture seasonality and variation in load data. This paper presents a conditional GAN (cGAN) architecture, that uses only four exogenous variables (maximum and minimum temperature, day of the week and month), to predict a daily profile (24 h of the day). Several types of GAN have been compared such as Deep Convolutional GAN, Least Squares GAN and Wasserstein GAN. The generated load profiles were tested on one year of data and compared to the real load profiles. The proposed GAN models provided excellent predictions, averaging a Mean Absolute Percentage Error (MAPE) of 4.99%. [ABSTRACT FROM AUTHOR]

The paper suggests a methodology for energetic buildings diagnosis for the Perugia University [Energy Buildings 1436 (2002) 1; B. Puttinger, Guideline for implementing energy audits in public buildings, Landes Energie Verein; R. Recalcati, G. Dall’O’, Procedure for buildings energetic certification, Punto Energia, Brescia (1998)].All of the university buildings have been put into six poles according to their geographical location.First of all, the entire area was subdivided into blocks according to didactics, office, and research areas. On the basis of this division, the utilization coefficient of the specific areas was determined for each pole.The paper suggests a method for pole electric consumption dividing into the three terms related to didactics, office, and laboratory block by the definition of the three “area indexes” Cdid.A, Coff.A, and Clab.A which represent the specific consumption per square meter of dedicated area. Their values, and in particular Clab.A value, vary according to the nature of the activities carried out in each pole. Starting from “area indexes”, the three “user indexes” (Cstu.U, Cper.U, and Cres.U) were then defined considering students, personnell, and researcher, separately.Area and user indexes’s yearly evaluation, dependent on the users number per typology and the blocks area, allows to identify the increasing or decreasing characteristic of the intensity of the activities carried out in each pole, to determine if consumptions depend mainly on class activities being carried out or on the use of electric heater and air conditioning systems.Finally, thermal consumptions were analyzed by evaluating the yearly energy amount and the specific consumption of each pole according to age and state of the buildings. [Copyright &y& Elsevier]

Short-term load forecasting for residential buildings is of great significance in ensuring the safe and economic operation of the power grid. However, most existing prediction methods focus solely on the temporal characteristics of individual buildings and ignore the spatial correlation between different buildings in a neighborhood. To tackle this problem and realize joint prediction for multiple buildings, a novel multi-task learning model with a selected-shared-private mechanism is proposed in this paper. Firstly, the temporal and spatial characteristics hidden in electricity consumption patterns are analyzed and represented by auto-correlation and cross-correlation, respectively. Then, a correlation-oriented combination strategy is proposed to build input feature set for the prediction model, and the temporal convolutional network is adopted to extract features. Furthermore, a novel selected-shared-private mechanism is designed for multi-task learning to improve the prediction accuracy, which can selectively utilize information from related tasks while learning private features. The proposed model is compared with other methods on the public dataset, and the results demonstrate that the proposed model can make satisfactory joint predictions for multiple buildings with more than 35% accuracy improvement over the support vector machine. • A correlation-oriented combination strategy is proposed to build the optimal input set. • A novel selected-shared-private mechanism is designed for multi-task learning. • The proposed model realizes superior joint load prediction for multiple residential buildings. [ABSTRACT FROM AUTHOR]

• Analysis of 15-minute sampled residential electricity consumption data of Dubai. • Households are classified using cooling consumption included in electricity bills. • K-Means clustering is used to group households based on their hourly consumption. • Consumption patterns helped to identify occupant behaviours in summer and winter. • Consumption patterns were cross-referenced to building type and occupant profiles. Analyzing residential load profiles and usage patterns is critical to making better decisions for demand-side management initiatives and designing strategies to get more people interested in energy savings. Therefore, analyzing load profiles, it is crucial to know how hourly consumption varies during summers, winters, weekdays, and weekends. In addition, determining the influence of occupancy, dwelling size, and building topologies on consumption is equally important to build predictive models. Therefore, this paper presents detailed research on the electricity consumption and profiles of the residential sector in Dubai based on the dwellings' characteristics and 15-minute resolution smart meter data. The data utilized includes the cooling systems, the number of occupants, and the physical characteristics of a dwelling, such as its typology, sizes, and number of bedrooms. First, using the K-Means clustering method, the authors grouped the households based on their consumption profiles. Second, the consumption patterns of each group of households were identified and organized based on similar consumption profiles over 24 h. Third, the authors applied several classification algorithms to assess the potential of using dwellings and occupants' characteristics to predict the patterns with which each household is associated. The analysis of consumption patterns showed that 43% of households with cooling included in their bills had the global peak demand at midnight during weekdays and weekends in summer. However, the global peak demand for cooling-excluded households occurs from 7:00 to 10:00 pm on weekdays and, in some specific cases, on the weekends, as early as 10:00 am. Finally, the methods used for classification were able to identify key characteristics driving the patterns of electricity demand and were well suited to this predictive modeling context. [ABSTRACT FROM AUTHOR]

The current energy crisis has drastically altered forecast electricity plans and budgets for European university campuses. This situation heightens the need to analyze their electrical consumption, with two main goals: identifying their patterns and promoting the development of renewable installations for these consumers. Previous research has focused only on aggregated demand data, with the studies being based on estimations and forecasts, and focused mainly on single buildings. Moreover, there is a lack of scientific papers that provide a replicable codebase for electricity analysis. Our work presents a novel methodology to overcome these research gaps, proposing the first comprehensive, replicable and scalable codebase to analyze electricity consumption in universities. It is based on three steps. The first comprises automated data collection of real electricity measurements at each electricity supply point. The second develops the complete analysis of electricity consumption. The last step parameterizes this consumption by identifying seasonal and daily profiles. The research was applied to the University of Castilla-La Mancha, campus Albacete (Spain) case study. The results revealed the 4 highest electricity-demanding buildings: Biomedical Complex, Higher Technical School of Industrial and Computer Engineering, Vice-rectorate and Library, and Higher Technical School of Agricultural and Forestry Engineering. The results are thus of great value for other educational buildings. [ABSTRACT FROM AUTHOR]

Optimal planning and operation studies of modern energy systems are tied up with medium to long-term predictions of energy demand. Deep learning algorithms have recently become significantly useful in this regard because of their capacity to learn load patterns with a high degree of fluctuation and uncertainty. Yet, empirical evaluation of different configurations of deep learning models, for enhancing prediction accuracy, has been mostly ignored in the preceding literature. In this paper, different architectures of deep recurrent neural networks (DRNNs) are explored and adjusted specifically for the purpose of performing medium- and long-term predictions of energy demands. We intend to create a bespoke DRNN for heating and electricity consumption prediction with a 1-h resolution. Moreover, hyperparameter optimization, which is a time-consuming and rigorous task in deep learning algorithms due to their abundance, dependence on the particular application, and empirical nature, is studied comprehensively. In this respect, the space of possible configuration variables is thoroughly explored to identify a collection of hyperparameters that enables the DRNN model to more accurately forecast energy consumption. We also benchmark the proposed algorithm's performance against other data-driven models, namely support vector machine (SVM) and gradient boosting (GB) regression. The proposed model outperforms SVM by 5.4% and GB regression by 7.0% in terms of energy forecasting accuracy. [ABSTRACT FROM AUTHOR]

Under the background of global warming, the world is experiencing many extreme climate events. The temperature fluctuation associated with climate change is often considered as a primary driving force of electricity consumption. To better explore the impact of temperature on electricity consumption, this paper aims to analyze the question from the perspective of income growth, i.e., the moderating effect of income growth on the response of urban residential electricity to temperature changes. Using the unbalanced panel data of 23 cities in the Yangtze River Delta Urban Agglomeration from 2004 to 2015, we construct the response model of residential electricity consumption to income growth, heating degree days, cooling degree days and their interactions. The results show that heating demand, cooling demand and income growth significantly increase residential energy consumption. However, urban residential disposable income growth weakens the positive effect of heating demand and cooling demand on residential electricity consumption. Because, urban residents have obvious demand for improving indoor heating comfort and cooling comfort, the moderating effect of income is shown as the crowding-out effect. Residents may adjust the structure of heating energy consumption (from electricity to natural gas) and upgrade their electrical equipment to high energy efficiency equipment, thereby reducing residential electricity consumption. This study also indicates that urban residential energy consumption is very sensitive to changes in permanent populations. The main contribution of this study is proposing the hypothesis of the negative moderating effect of income growth on temperature changes and expanding the response model of residential electricity consumption to temperature changes. Finally, this paper also proposes the adaptive strategies and policy implications for climate change. [ABSTRACT FROM AUTHOR]

• This paper determines the relationship between urban form and building energy use. • The surface-to-volume ratio and obstruction angle of buildings are important. • Consideration of the heating energy type and building use can yield reliable result. • The urban heat island effect exerts an impact on the electricity use in some months. • Correlation variation is caused by the building permit date. In the climate change era, urban geometry serves to ensure sustainability for urban planners since it notably affects building energy consumption. However, previous studies have mainly conducted experimental investigations under hypothetical conditions. Without empirical evidence, it is difficult to implement these findings under real conditions. This paper examined empirical data to determine the urban geometry effect on building energy consumption in a large, complicated urban area. Urban geometric parameters, including the surface-to-volume ratio, obstruction angle, and building orientation, were measured across Seoul, and their correlations with building energy use were analyzed. The results show that the surface-to-volume ratio and obstruction angle are important, despite the empirical data complexity. Additional data analysis indicates that the heating energy type and building use consideration yields more reliable correlation coefficient trends. The potential impacts of natural ventilation and the urban heat island (UHI) effect on building energy consumption were examined. The UHI effect, rather than shadows on buildings, predominantly impacts the building electricity use in August and September. The correlation variation caused by the building permit date was investigated. The results in this paper could help urban planners and designers develop blueprints to control urban geometry and establish relevant design guidelines. [ABSTRACT FROM AUTHOR]

Considerable energy is used for cooling buildings in hot climates like those found in Gulf Cooperation Council (GCC) countries. Designing building envelopes for the purpose of saving energy can necessarily help reduce the building's operating expenses and therefore energy demand over time. For roofing systems, a research project was recently conducted to assess experimentally and numerically the performance of the roofing systems. This paper presents the experimental results of this project. Cool/reflective/white roofs use reflective materials or coatings to reflect a portion of the incident solar radiation. This results in lowering the surface temperature of the cool roofs compared to black roofing systems. As such, cool roofs help reduce the cooling loads during the summer season. This paper presents the test results of characterizing the short-wave solar reflectivity of a reflective roofing material that can be used in cool roofs of Saudi buildings. A Reflective Coating Material (RCM) that is commercially available in the GCC markets was used in this study. Also, the reduction in the solar reflectivity due to dust and dirt accumulations on the surface of the RCM was investigated in this study when this coating was exposed to a dusty and polluted climate of Jubail Industrial City (JIC). As well, the effect of applying different cleaning operations on the solar reflectivity of the RCM was investigated. The test results showed that dust/dirt can significantly contribute in reducing the short-wave solar reflectivity of the RCM. For the RCM subjected to the natural weathering conditions at different exposures times, the results showed that a maximum enhancement of 5% in the short-wave solar reflectivity was achieved, as a result of conducting air plowing process that simulates the wind action. Additionally, a maximum enhancement of 45% in the short-wave solar reflectivity was achieved as results of conducting water cleaning/rinsing process that simulates the rain action and as well conducting washing with dishwashing detergent that simulates homemade cleaning. Furthermore, conducting professional cleaning has resulted in insignificant enhancement in the short-wave solar reflectivity compared to conducting homemade cleaning. Finally, this paper provides a test procedure for characterizing the dust concentration/intensity on the surface of the RCM at different exposure times to the natural and polluted climate. This test procedure can easily be applied on different types of reflective roofing materials and membranes that can be used in cool roofs. [ABSTRACT FROM AUTHOR]

[Display omitted] • Discussed the current status of cooing technologies in data centers. • Introduced and reviewed existing phase change cooing in data centers. • Compared and summarized the characteristics of phase change cooling systems. • Analyzed the application limitations of phase change cooling technologies. • Explored the future research direction of phase change cooling technologies. With the advent of big data era, data centers (DCs) related energy use accounts for approximately 3% of the global electric power consumption. As the augmentation of data-processing performance and thermal density in DCs, its energy use will undoubtedly continuously burst. Meanwhile, the potential safety hazards, caused by IT equipment local overheating, threaten the safe operation and restrict the further development of DCs. Thus, efficient cooling approaches should be applied in DCs to ensure its safe operation and optimize its thermal environment and cooling efficiency. Phase change cooling (PCC) technology is regarded as one of the effective and widely-used cooling methods, which have been applied in DCs for several years. In this paper, the up-to-date PCC technologies are reviewed and summarized, as well as the latest progress in DC cooling field. Four main PCC technologies are discussed in this paper, namely independent heat pipe cooling, integrated heat pipe cooling, two-phase immersion cooling, and cold storage systems. Finally, the shortcomings of the current researches on the PCC methods are summarized, while some suggestions for future researches are provided to promote the application of PCC technologies and achieve safe operations and energy savings in data centers. [ABSTRACT FROM AUTHOR]

• Changes in peak demand are often larger than changes in annual demand. • Ventilation rate is the parameter with the largest contribution for HVAC demand. • Equipment and lighting density have substantial contributions for total demand. Previous dynamic building simulation studies have not often focused on analysing the sensitivity of peak loads to input parameters. However, these peak loads may have a critical impact on system design capacities and power network operation. This study aims to examine the sensitivity of cooling demand related results (total electricity demand, HVAC end-use and space cooling) in a large office building using two global sensitivity analysis methods: Morris elementary effect and Sobol indices. More specifically, this paper examines the implications of different type of climates to the uncertainty in these different cooling output results and the sensitivity of the different parameters for each result. Moreover, this paper investigates the difference between the effects of annual and peak analysis for cooling demand of office buildings, which can provide insight on cooling demand from the perspectives of total cooling energy and system capacity for building cooling systems, respectively. This study has found that generally, the changes are more significant for peak demand than for annual demand. The coefficient of variation for the total peak demand is around 25% and 21% for total annual demand. This study identifies that the ventilation rate is the parameter that contributes the largest for the uncertainty in electricity demand of the HVAC end-use, between 50% and 70% of the change (S T), both for annual and peak demand. Regarding the effect on total electricity demand, ventilation rate is still one of the most critical factors, but equipment and lighting densities also become a significant contributor to the sensitivity of the total demand. [ABSTRACT FROM AUTHOR]

This paper presents a building stock energy model for the estimation of hourly electricity consumption for a large group of residential buildings. A Monte Carlo model stochastically generates a large sample of dwellings representative of the building stock and the correspondent number of user profiles, statistically supported by a web survey about the use of energy in dwellings for space heating and cooling. The model uses hourly energy balance equations to estimate energy needs and calculates the mean annual electricity consumption for regularly occupied dwellings with an error below 3%. Model is also validated against independent smart-metered data of about 250 dwellings. Hourly electricity consumption results feature an overall normalised mean absolute error of 11% and normalised root mean square error of 16%. The maximum relative difference is  ± 72% and the maximum absolute error is ≃217 Wh/h. The model is considered to be able to predict hourly electricity consumption accurately. [ABSTRACT FROM AUTHOR]

An increasing interest in the representation of the intra-day variability observed in domestic electricity consumption patterns has driven the development of various modelling frameworks that incorporate consumer behavioural patterns as a key element for the simulation of electricity consumption. Some of the existing models produce reasonable representations of the broader characteristics of user activity patterns, typically rendered as dwelling occupancy patterns. However, when these activity patterns are used to produce estimates of the electricity demand loads associated with such activities, little attention is paid to the links between activities and the actual use of the electric equipment responsible for the production of the actual demand loads. Instead, the simulation of demand loads from activity patterns is ruled by simplifying assumptions that mask the reality behind those links. This paper therefore seeks to unpack the relationship between activity and electricity demand profiles by focussing on the underlying activity patterns in more detail, and how these relate to the usage patterns of the associated appliances. These relationships are studied based on currently available datasets. The analysis of the activities associated with the use of more than one appliance revealed the differences between the likelihood of each appliance being activated throughout the day relative to user engagement in the activity the appliance is associated with. In practice, what this shows is how each appliance's share of the demand load associated with the activity varies throughout the day. As a result, the power consumption associated with a particular activity is subject to the same kind of variability; activity-related demand. The results of this analysis can be used in conjunction with current or new modelling approaches with a view to linking the user activity patterns with the simulation more of realistic electricity demand loads. [ABSTRACT FROM AUTHOR]

Non-intrusive load monitoring methods aim to disaggregate the total power consumption of a household into individual appliances by analysing changes in the voltage and current measured at the grid connection point of the household. The goal is to identify the active appliances, based on their unique fingerprint. An informative characteristic to attain this goal is the voltage–current trajectory. In this paper, a weighted pixelated image of the voltage–current trajectory is used as input data for a deep learning method: a convolutional neural network that will automatically extract key features for appliance classification. The macro-average F -measure is 77.60% for the PLAID dataset and 75.46% for the WHITED dataset. [ABSTRACT FROM AUTHOR]

While urban form affects building energy consumption, the pathways, direction and magnitude of the effect are disputed in the literature. This paper uses a unique dataset to examine the effect of urban form on residential electricity consumption in Ningbo, China. Using survey and utility bill data of 534 households in 46 neighborhoods in the city, we model the electricity use of households using a multi-level regression model. We find that neighborhood street configuration and tree shade are important in controlling residential electricity consumption and, consequently, greenhouse gas emissions. Our results suggest that seasonality and dwelling type condition the effect of neighborhood densities on electricity consumption. Neighborhood density is associated with household electricity consumption in summer months, while there is no such association in winter months. As neighborhood density increases, households in slab and tower apartments in dense urban neighborhoods consume more electricity in summer months, which can be partly explained by exacerbated heat island effect. Interestingly, the neighborhood density is negatively associated with summer electricity consumption for single-family houses, suggesting that the effect of neighborhood density is different for different types of dwelling units. [ABSTRACT FROM AUTHOR]