Residential Load Research Articles

Optimal Placement of Multiple Sources in a Mesh-Type DC Microgrid Using Dijkstra’s Algorithm

This research paper introduces an optimization methodology for the strategic electric sources’ placement at multiple positions in a DC islanded microgrid characterized by a mesh network, aiming to minimize line losses while considering minimal cable weight. The DC microgrid studied in this paper is composed of PV panels, batteries, a diesel generator, and 20 residential loads. Employing Dijkstra’s algorithm, a graph algorithm used in Google Maps, the study identifies the shortest path (resistance) between potential source nodes and various variable loads within a predefined electric distribution mesh network topology. This study focuses on active power considerations and offers valuable insights into the placement optimization of multiple sources’ positions in DC microgrid mesh networks. A key contribution of this paper lies in the ranking of source node positions based on minimal to maximal line losses, taking into consideration optimal cable weights, while using MATPOWER to validate sources’ ranking based on Dijkstra’s hypothesis. The research further includes a techno-economic study to assess the viability of sources’ placement at multiple positions within the mesh network, comparing it with the optimal placement scenario involving a single position for all sources. This methodology serves as a valuable resource for system designers and operators aiming to minimize line losses and optimize energy distribution in DC microgrids in a mesh topology.

Enhancing residential heating load prediction with advanced machine learning and optimization techniques

The effective prediction of heating load in residential buildings is crucial for optimizing energy consumption and enhancing thermal comfort. This study investigates three prominent regression techniques Extra Tree Regression, Decision Tree Regression, and Support Vector Regression utilized for this purpose. To improve their predictive performance, a novel metaheuristic optimization algorithm, the Tyrannosaurus Rex Optimization Algorithm, is employed. The dataset used includes features such as Relative Compactness, Surface Area, Wall Area, Roof Area, Overall Height, Orientation, Glazing Area, and Glazing Area Distribution. Each machine learning model (Extra Tree Regression, Decision Tree Regression, Support Vector Regression) is trained on historical heating load data and optimized using the Tyrannosaurus Rex Optimization Algorithm to adjust internal parameters. Comparative analysis demonstrates that the proposed Extra Tree + Tyrannosaurus Rex Optimization Algorithm model achieves superior performance with an R-squared value of 0.987 and a low Root Mean Square Error of 1.167. By leveraging the strengths of different machine learning algorithms, this research presents a reliable approach for accurately predicting heating load in residential buildings. This method holds promise for improving energy efficiency, enhancing thermal comfort, and promoting sustainable building design.

Assessment of techno-economic analyzes of grid-connected nuclear and PV/wind/battery/hydrogen renewable hybrid system for sustainable and clean energy production in Mersin-Türkiye

The main novelty in the presented paper is that it presents an energy analysis for a hybrid system that integrates nuclear power plants with wind/solar power plants for sustainable and clean energy production. In addition, excess energy is used to produce hydrogen. A techno-economic feasibility assessment is performed to ensure continuous electricity supply for hourly and peak residential load requirements in Mersin located in Türkiye. In the first stage of the study, suitable areas and sizes for solar and wind energy for energy production were determined with ArcGIS and Multi-Criteria Decision-Making methodologies. In the second stage, six scenarios were investigated in terms of electricity production, and excess energy was used for hydrogen production. The results brought out that more electricity was produced than the city's electricity consumption (5395 GWh) in all 6 scenarios considered. The maximum electricity production at a rate of 11900 GWh/yr and the maximum hydrogen at a rate of 132675 tons was produced in Scenario 4 with the cost of energy (0.31 $/kWh) and cost of hydrogen (12.6$/kgH2). In this scenario, the percentages of solar, wind and nuclear energy in total production 1 %, 12 %, and 87 %, respectively. According to the results, it was concluded that 236621 tons of carbon emissions were reduced when the city's energy needs were met according to the 6 scenarios in this study.

Short-Term Residential Load Forecasting via Pooling-Ensemble Model With Smoothing Clustering

Short-Term Residential Load Forecasting via Pooling-Ensemble Model With Smoothing Clustering

Energy Consumption Management of Residential Appliances Based on Load Signatures Decomposition

Nowadays, energy consumption management techniques in the residential side have gained significant importance due to their considerable influence on the control of power flow in distribution networks and especially the possibility of managing a huge part of domestic electrical demand during peak-load hours. Since the customer’s data are recorded in an aggregated form, therefore, in order to apply control approaches, it is necessary to use load pattern evaluation techniques (load signature). These methods capable to decomposition effective features that help control approaches to implement with more accuracy. In this article, features of residential loads have been extracted by using the signature of the aggregated consumer’s demand. Then these features have been evaluated by methods such as logistic regression, k-nearest neighborhood, and decision tree. By assessing the results, it was determined that among the extracted features, the first two features (consumed power and injected harmonics) covered more than 89% of the variance of the entire set, and with the help of using the principal component analysis method, it was determined that by reducing the number of features to 2, a considerable amount of computation is reduced and only about 4% of the accuracy is reduced. Also, the convolutional neural network approach was used to estimate the type of load, and by identifying controllable loads and applying remote home energy management methods, it was found that by increasing participation up to 80%, more than 41% of peak-load consumption could be shifted to off-peak hours.

Impacts of Plug-in EVs and decentralized power generation on distribution system operation

Plug-in Electric Vehicles (PEVs) have gained popularity in recent years due to their environmentally friendly attributes. However, integrating PEVs into distribution systems poses novel challenges and opportunities. This research paper undertakes a comprehensive assessment of the repercussions of PEVs and distributed generation (DG) integration on the distribution system’s reliability, power loss, and voltage stability. In order to alleviate peak loads, select PEVs are deployed in Vehicle-to-Grid (V2G) mode. The integration of DG units, particularly solar Photovoltaic (PV) and wind turbine (WT) systems, is explored to mitigate the potential strain of PEV charging on the distribution system. The uncertainty of renewable DG and PEV charging-discharging, along with time-varying residential and commercial loads, are included in the analysis. Power interruption duration and frequency-based reliability index are considered in this work. The combined effects of PEVs’ Grid-to-Vehicle (G2V) and V2G mode and DG integration are comprehensively investigated, and the impact assessment is conducted on the modified IEEE 33 bus distribution system to ensure the practicality and applicability of the research findings.

Boosting prairie dog optimizer for optimal planning of multiple wind turbine and photovoltaic distributed generators in distribution networks considering different dynamic load models

Deploying distributed generators (DGs) supplied by renewable energy resources poses a significant challenge for efficient power grid operation. The proper sizing and placement of DGs, specifically photovoltaics (PVs) and wind turbines (WTs), remain crucial due to the uncertain characteristics of renewable energy. To overcome these challenges, this study explores an enhanced version of a meta-heuristic technique called the prairie dog optimizer (PDO). The modified prairie dogs optimizer (mPDO) incorporates a novel exploration phase inspired by the slime mold algorithm (SMA) food approach. The mPDO algorithm is proposed to analyze the substantial effects of different dynamic load characteristics on the performance of the distribution networks and the designing of the PV-based and WT-based DGs. The optimization problem incorporates various operational constraints to mitigate energy loss in the distribution networks. Further, the study addresses uncertainties related to the random characteristics of PV and WT power outputs by employing appropriate probability distributions. The mPDO algorithm is evaluated using cec2020 benchmark suit test functions and rigorous statistical analysis to mathematically measure its success rate and efficacy while considering different type of optimization problems. The developed mPDO algorithm is applied to incorporate both PV and WT units, individually and simultaneously, into the IEEE 69-bus distribution network. This is achieved considering residential, commercial, industrial, and mixed time-varying voltage-dependent load demands. The efficacy of the modified algorithm is demonstrated using the standard benchmark functions, and a comparative analysis is conducted with the original PDO and other well-known algorithms, utilizing various statistical metrics. The numerical findings emphasize the significant influence of load type and time-varying generation in DG planning. Moreover, the mPDO algorithm beats the alternatives and improves distributed generators' technical advantages across all examined scenarios.

Demand Management for Manufacturing Loads Considering Temperature Control under Dynamic Electricity Prices

Demand response (DR) can provide extra scheduling flexibility for power systems. Different from industrial and residential loads, the production process of manufacturing loads includes multiple production links, and complex material flow and energy flow are closely coupled, which can be seen as a typical nondeterministic polynomial-time (NP) hard problem. In addition, there is a coupling effect between the temperature-controlled loads (TCLs) and the manufacturing loads, which has often been ignored in previous research, resulting in conservative electricity consumption planning. This paper proposes an optimal demand management for the manufacturing industry. Firstly, the power consumption characteristics of manufacturing loads are analyzed in detail. A state task network (STN) is introduced to decouple the relationship between energy and material flow in each production link. Combining STN and production equipment parameters, a general MILP model is constructed to describe the whole production process of the manufacturing industry. Then, a mathematical model of the TCLs considering a comfortable human degree is established. Fully considering the electricity consumption behavior of equipment and TCLs, the model predictive control (MPC) method is adopted to generate the optimal scheduling plan. Finally, an actual seat production enterprise is used to verify the feasibility and effectiveness of the proposed demand management strategy.

Heat Pumps with Smart Control in Managing Australian Residential Electrical Load during Transition to Net Zero Emissions

Australia, like many other countries around the world, is undergoing a transition toward net zero emissions. It requires changes and development in many sectors, which not only bring benefits but also challenges. The rapid growth in renewable energy sources (RESs) is necessary to decarbonise electricity generation but negatively affects grid stability. Residential buildings also contribute to this issue through specific load profiles and the high penetration of rooftop photovoltaic (PV) installations. Maintaining grid balance will be crucial for further emissions reductions. One of the potential solutions can be the replacement of conventional heating and cooling systems in houses with solutions capable of storing energy and shifting the electrical load. As presented in this paper, heat pumps and hydronic systems can significantly improve the electrical load of a typical South Australian household when they are controlled by algorithms reacting to the current grid conditions and household-generated electricity compared to conventional solutions. TRNSYS 18 simulations of air source and ground source heat pump systems with smart control based on measured electricity consumption and domestic hot water usage data showed the possibility of total energy consumption reduction, shifting the load from peak periods towards periods of excessive RES generation and increasing self-consumption of rooftop PV electricity. These improvements reduce the amount of emissions generated by such a household and allow for further development of other sectors.

Enhancing multivariate, multi-step residential load forecasting with spatiotemporal graph attention-enabled transformer

Short-term residential load forecasting (STRLF) holds great significance for the stable and economic operation of distributed power systems. Different households in the same region may exhibit similar consumption patterns owing to the analogous environmental parameters. Incorporating the spatiotemporal correlations can enhance the load forecasting performance of individual households. To this end, a spatiotemporal graph attention (STGA)-enabled Transformer is proposed for multivariate, multi-step residential load forecasting in this paper. Specifically, the multiple residential loads are cast to a graph and a Transformer with a graph sequence-to-sequence (Seq2Seq) structure is employed to model the multi-step load forecasting problem. Gated fusion-based STGA blocks are embedded in the encoder and decoder of the Transformer to extract dynamic spatial correlations and non-linear temporal patterns among multiple residential loads. A transform attention block is further designed to transfer historical graph observations into future graph predictions and alleviate the error propagation between the encoder and decoder. The embedding of multiple attention modules in the Seq2Seq framework allows us to capture the spatiotemporal correlations between residents and achieve confident inference of load values several steps ahead. Numerical simulations on residential data from three different regions demonstrate that the developed Transformer method improves multi-step load forecasting by 14.7% at least, compared to the state-of-the-art benchmarks.

Constructing Australian Residential Electricity Load Profile for Supporting Future Network Studies

Constructing Australian Residential Electricity Load Profile for Supporting Future Network Studies

Efficient energy management of domestic loads with electric vehicles by optimal scheduling of solar-powered battery energy storage system

The increasing adoption of electric vehicles (EVs) and variable energy usage patterns substantially strain the electrical grid; indeed, optimal energy management, monitoring, and utilization are required for the reliable operation of the grid. This paper introduces a novel model design of a solar-powered battery energy storage system (SPBESS) as a viable substitute for conventional demand-side management (DSM) and time of use (ToU) pricing schemes, intending to optimize energy management and utilization with IoT monitoring. In addition, the IoT-based prototype has been developed to monitor and control the proposed system. To validate the proposed SPBESS model, the study examines two distinct cases: one encompassing domestic loads and the other integrating EV loads alongside household demand. Using ToU pricing, it determines the optimal charging and discharging strategies for the SPBESS, evaluates their implications for the system configuration and grid ToU pricing, and quantifies the annual reductions in power purchases, electricity bills, energy costs, and carbon emissions. Moreover, the study comprehensively analyzes the optimal power exchanges between the photovoltaic system, battery energy storage system, and the grid, precisely considering the specific load requirements and grid ToU pricing. The conducted research findings manifest considerable decreases in energy costs, with the domestic load condition witnessing a reduction from $0.312 per kilowatt-hour to $0.245 per kilowatt-hour and the scenario involving EV and residential loads experiencing a decline from $0.27 per kilowatt-hour to $0.197 per kilowatt-hour. Furthermore, the annualized cost savings for cases 1 and 2 amount to $1,336 and $4,544, respectively, yielding substantial reductions in polluting gas emissions. Besides, the IoT-based constructed prototype model design for real-time power parameters monitoring and control shows the importance of IoT-based monitoring for remote load control, allowing users to maximize energy efficiency, resource utilization, and system visualization.

Optimal multi-objective sizing of renewable energy sources and battery energy storage systems for formation of a multi-microgrid system considering diverse load patterns

Optimal microgrid design is pivotal in planning active distribution networks (ADNs) with intermittent renewable energy sources (RESs) and battery energy storage systems (BESSs). This paper introduces an innovative approach to clustering existing ADN systems, incorporating RESs and BESSs into a set of microgrids (MGs) termed a multi-microgrid (MMG). The approach accommodates diverse load patterns, considering the dynamic nature of residential, commercial, and industrial loads. The proposed methodology employs an innovative k-means algorithm that utilizes end node characteristics based on the graph partitioning method and the Silhouette technique, transforming the existing system into an MMG framework. Furthermore, an iterative pareto-fuzzy (IPF) method is suggested to determine optimal RES and BESS sizes, considering reliability, total cost, and the unutilized surplus power function. The framework integrates total line losses into the generation adequacy assessment of energy management strategy (EMS) for each MG within the MMG system. The design and performance analysis of the proposed multi-objective optimization algorithm is tested on the IEEE 33-bus distribution system. The implementation of the independent operation index (IOI) and round-trip efficiency (RTE) metrics demonstrates the effectiveness of the proposed approach in constructing an MMG, with achieved values exceeding 90 percent for IOI and 77 percent for RTE.

Economic operation of residential load using IOT-based renewable energy management system

Economic operation of residential load using IOT-based renewable energy management system

A probabilistic model for residential community heating load calculation based on occupancy patterns

Ensuring precise prediction of community heating loads is crucial for optimizing the performance of district heating systems. However, the currently prevalent methods, like the simultaneous usage coefficient method and the area index method (AIM), which heavily rely on practical engineering experience, may not be suitable for ultra-low energy residential buildings (UERBs). Applying these methods to UERBs may result in overcapacity of the heating system, which increases energy consumption and costs. This study aims to enhance the accuracy of heating load calculations for UERB communities. A probabilistic model for heating load calculation was proposed, considering the “part time, part space” spatiotemporal distribution characteristics of residents. Typical occupancy patterns in cold zones of China were identified through questionnaire surveys and K-means cluster analysis, and typical models of UERBs were established, with both the building models and survey data validated. The residential building energy model (RBEM) was applied to a case community in Dalian, analyzing the impact of occupancy patterns on heating loads. The results indicate that the RBEM method increased the system's heating load factor by 6.9 % and reduced its design capacity by 32.9 % compared to the AIM. The proposed method might guide the development of management strategies in the energy planning stage of district energy systems.

Demand response of large-scale residential load to concentrated variable renewable energy

Large-scale concentrated Variable Renewable Energy (VRE) sources pose a challenge to the balance and security of electricity systems. Studies have shown that the demand side may offer a greater responsiveness based on the shiftability of loads. However, there is currently no known literature on the demand response (DR) of large-scale shiftable loads in the residential sector. Firstly, there is no sufficiently large database of residential appliance-level load. Secondly, scheduling a large number of small residential shiftable loads simultaneously is a major challenge. Ultimately, the DR of large-scale residential buildings encounters the challenge of ensuring sufficient backup shiftable load to effectively cope with real-time VRE fluctuations during the response period. To address these issues, this paper proposes an appliance-level load data generation method based on diffusion neural networks. Meanwhile, a clustering method based on Generalized End to End (GE2E) loss neural networks is proposed to solve the complexity obstacles in simultaneous load swarm scheduling. Thirdly, a two-stage scheduling approach is proposed to address the challenges in the large-scale DR period through constructing day-ahead and real-time response models. The simulation results show that large-scale residential shiftable loads have a strong response capability to VRE fluctuations, and their response cost is lower than the conventional fuel-fired-power-plant-based-response within a certain user preference cost range. In practice, the clustering and scheduling methods proposed in this paper can be applied after the non-intrusive load monitoring (NILM) method is deployed to decompose the load data.

Characterizing residential sector load curves from smart meter datasets

Understanding how and when residential electricity is used throughout the day is integral to the successful implementation of potential residential demand management strategies. Our analysis characterizes the daily hourly load profiles of approximately 160,000 residential electricity customers across the Southern California Edison (SCE) service area during the period spanning 2015 to 2016 and 2018 to 2019 across weekends, weekdays, seasons, and climate zones. We find that total daily electricity usage was highest in the hottest months of the year compared to milder months, particularly for households located in the hottest climate zones. The most energy-consumptive hours occurred during the mid-afternoon during the hottest months, in contrast to early evening high consumption in cooler months. We find that customers with average daily consumption at or above the 80th percentile cumulatively consume over 40% of electricity during the hottest months of the year residential load, while the bottom half of customers cumulatively consume <25% of the total residential load. The disparities in electricity usage across SCE households are higher in the mid-day, especially in milder months across all regions, and in mild climate zones compared to hotter climate zones since loads are not as dependent on high HVAC loads.

Techno-economic optimization of a hybrid energy system with limited grid connection in pursuit of net zero carbon emissions for New Zealand

As New Zealand moves towards net zero carbon emissions by 2050, multiple factors must be considered including increased electrical load due to electrification, variability of renewable energy generators, required storage capacity, system economics and limitations on grid transmission capacity. Complex and region-specific interactions between the various design choices involved are likely to require an understanding of a range of optimal and near-optimal designs for proposed micro-grid systems as opposed to a single optimal point. This work develops a novel multiscale optimization algorithm for optimization from a univariate capacity optimization approach to a multivariate one. This enhanced algorithm is applied to a grid-connected hybrid energy system consisting of local wind and solar generation, battery storage, and a limited grid connection for industrial and residential loads. This analysis is repeated for current 2023 and forecasted net zero 2050 grid conditions. Development of local generation allows for a 36.8% reduction of levelized cost of electricity in the 2023 case and a 38.6% reduction in the 2050 case. This results in a projected reduction of 19920 tonnes of CO2/yr. The algorithm and methodology developed are broadly applicable to optimization of next-generation energy grids.

A risk warning method for steady-state power quality based on VMD-LSTM and fuzzy model

The risk warning for steady-state power quality in the power grid is essential for its prevention and management. However, current risk warning methods fall short in predicting the power quality trend while accounting for potential risks. Consequently, this study introduces a novel steady-state power quality risk warning method utilizing VMD-LSTM and a fuzzy model. Firstly, a power quality index prediction method based on variational mode decomposition (VMD) and long short-term memory (LSTM) is proposed. This approach significantly enhances prediction accuracy. Secondly, a power quality risk warning method incorporating kernel density estimation (KDE) and a fuzzy model is proposed, which systematically addresses the uncertainty associated with power quality risks. To validate the effectiveness and practicality of the proposed method, experiments are conducted using field monitoring data from a residential load in southern China. The results affirm the reliability and applicability of the proposed method. The simulation results show that the median error of prediction of power quality indexes by the proposed method is 5.03 % during the evaluated time period, and the prediction accuracy is mostly maintained above 90 %.

Privacy-Preserving Distributed Learning for Residential Short-Term Load Forecasting

Privacy-Preserving Distributed Learning for Residential Short-Term Load Forecasting