Electricity Consumption Characteristics Research Articles

Behavioral analysis of electricity consumption characteristics for customer groups using the k-means algorithm

In the fierce competition of the electricity market, how to consolidate and develop customers is particularly important. Aiming to analyze the electricity consumption characteristics of customer groups, this paper used a k-means algorithm and optimized it. The number of clusters was determined by the Davies-Bouldin index (DBI). An improved Harris Hawks optimization (IHHO) algorithm was designed to realize the initial cluster center selection. Based on data such as electricity purchase and average electricity price, electricity customer groups were clustered using the IHHO-k-means algorithm. The IHHO-k-means algorithm achieved the best clustering effect on Iris, Wine, and Glass datasets compared with the traditional k-means and PSO-k-means algorithms. Taking Iris as an example, the optimal value of the IHHO-k-means algorithm was 96.538, with an accuracy rate of 0.932, precision and recall rates of 0.941 and 0.793, respectively, an F-measure of 0.861, and an area under the curve (AUC) value of 0.851. In the customer dataset, the number of clusters determined by DBI was 4. The power customers were divided into four groups with different characteristics of electricity consumption, and their electricity consumption behaviors were analyzed. The results prove the reliability of the IHHO-k-means algorithm in analyzing electricity consumption characteristics of customer groups, and it can be applied in practice.

Economic optimization scheduling of virtual power plants considering an incentive based tiered carbon price

Against the backdrop of China's carbon peak and carbon neutrality goals, the use of virtual power plants for electricity economic dispatch has gradually become a research hotspot. The virtual power plant includes both supply and demand sides. If the inherent conflicts of interest between both parties cannot be effectively addressed, there are significant hidden dangers in the sustainable operation of the virtual power plant. On the basis of existing research, this article constructs a regional virtual power plant. With the goal of maximizing supply and demand utility, a two-stage Stackelberg dynamic game model was established between virtual power plant operators and users in the electricity market and carbon trading market mechanisms, and an incentive based hierarchical carbon price constraint was designed in the optimization model. In the model constructed in this article, the operator of a virtual power plant will consider environmental benefits based on expected returns, adjust the output of controllable units to change the electricity supply strategy, and formulate a pricing strategy. At the same time, users will adjust their electricity consumption strategy based on their own electricity consumption characteristics, and both will achieve a game equilibrium. Based on the optimization results, this article conducted a sensitivity analysis on the incentive coefficient of graded carbon prices and obtained the optimal incentive coefficient under the case conditions. The research results indicate that the model can coordinate and meet the interests and needs of both the power generation and consumption sides, and the implementation of incentive based hierarchical carbon pricing strategies has certain advantages compared to traditional carbon pricing strategies in different user types scenarios. The research conclusion of this article has good reference value for the sustainable green operation of virtual power plants.

Spatialization of electricity consumption by combining high-resolution nighttime light remote sensing and urban functional zoning information

ABSTRACT An accurate and timely spatialization of electricity consumption is fundamental for energy management and sustainable urban development. While previous research has relied heavily on statistical data or moderate-resolution nighttime light data, this study has presented a new spatialization method by combining high-resolution Luojia 1–01 nighttime light and urban functional zoning information. The total electricity consumption volume can be allocated to each land use pixel based on its strong linear relationship with nighttime light. Specifically, urban functional zoning data were connected with the corresponding economic sectors to differentiate the complex relationships between electricity consumption and nighttime light within different zones. The digital number value of every pixel can be multiplied by its corresponding electricity consumption coefficient. A number of comparisons have indicated that the proposed method can more accurately characterize the spatial distribution of electricity consumption. Our results can exhibit much clearer outlines and more detailed internal characteristics. More importantly, urban functional zoning information can be used to distinguish the electricity consumption of various economic sectors at fine scales. The proposed method is expected to capture the spatial characteristics of electricity consumption in a timely and accurate manner. The findings can help local authorities formulate sustainable energy utilization and emission reduction strategies.

Transient dataset of household appliances with Intensive switching events

With the development of Non-Intrusive Load Monitoring (NILM), it has become feasible to perform device identification, energy consumption decomposition, and load switching detection using Deep Learning (DL) methods. Similar to other machine learning problems, the research and validation of NILM necessitate substantial data support. Moreover, different regions exhibit distinct characteristics in their electricity environments. Therefore, there is a need to provide open datasets tailored to different regions. In this paper, we introduce the Transient Dataset of Household Appliances with Intensive Switching Events (TDHA25). This dataset comprises switch instantaneous data from 10 typical household appliances in China. The TDHA dataset features a high sampling rate, accurate labelling, and realistic representation of actual appliance start-up waveforms. Additionally, appliance switching is achieved through precise control of relay switches, thus mitigating interference caused by mechanical switches. By furnishing such a dataset, we aim not only to enhance the recognition accuracy of existing NILM algorithms but also to facilitate the application of NILM algorithms in regions sharing similar electricity consumption characteristics to those of China.

Classification model of electricity consumption behavior based on sparse denoising autoencoder feature dimensionality reduction and spectral clustering

The development of electrical measurement technology has brought high latitude residential electricity consumption data to power companies, which contains the characteristics of users' electricity consumption behavior and provides data support for the behavior classification. In order to improve the efficiency of data feature extraction and the accuracy of electricity consumption behavior identification, a classification model based on sparse denoising autoencoder feature dimensionality reduction and spectral clustering is proposed in this paper. Firstly, the sparse denoising autoencoder (SDAE) and the manually defining electricity consumption characteristic indicators are deployed to extract features from the residential daily electricity consumption data, and then the spectral clustering is employed to classify the extracted electricity consumption characteristics. Secondly, the t-distributed stochastic neighbor embedding (t-SNE) is applied to visualize and analyze the classification results, and on this basis, the secondary classification is implemented to fix the issue of the confused electricity consumption behaviors. Finally, the typical consumption behavior curves are calculated by Gaussian distance weighting method, and the characteristics of power consumption behavior are analyzed and summarized. The proposed approach is evaluated and verified by using the electricity dataset in Fujian, China.

Research on load curve fitting method based on K-means algorithm

Some areas of China are affected by the fluctuation of new energy output and load, and the power supply as a whole presents an unbalanced phenomenon of abundant electricity supply but shortage of power supply. To understand the characteristics of the load electricity consumption curve more accurately, this paper proposes a method of fitting the load electricity consumption curve based on K-mea ns algorithm. First, analyze the characteristics of the load electricity consumption based on the industry subdivision, and build the load electricity consumption characteristic index system. Then, the k-mea ns cluster analysis is carried out based on the characteristics of the load power consumption curve of the subdivided industries to determine the number of clusters, and the curve fitting and deviation test are carried out for each cluster group. This paper carries out practical analysis and theoretical research based on the data of typical industry load power consumption curve in Ningxia.

Line–Household Relationship Identification Method for a Low-Voltage Distribution Network Based on Voltage Clustering and Electricity Consumption Characteristics

To address the issue of inconspicuous electricity consumption characteristics among vacant users in low-voltage distribution networks (LVDNs), which hinders effective line–household relationship identification (LHRI), a method for identifying line–household relationship based on voltage clustering and electricity consumption characteristics is proposed. Initially, the paper employs Dynamic Time Warping (DTW) to analyze the similarity of user voltage profiles and utilizes the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm to cluster users. This approach identifies the topological relationship between vacant users and regular users to obtain multiple user categories. Subsequently, by analyzing the electricity consumption characteristic, the connection relationships between different user categories and phase lines are clarified based on the correlation between the electricity consumption characteristic vector of phase lines and the electricity consumption characteristic vector of user categories, thereby revealing the line–household relationship for all users. On the test dataset, the LHRI algorithm proposed in this article achieved 100% accuracy, within an allowable error range of 0.2%, and improved the accuracy by 20% compared to the traditional identification method. Finally, the LVDN simulation model established by OpenDSS 9.4.0.3 was used to verify the effectiveness of the proposed method, confirming its potential and advantages in practical applications.

Comparison Analysis for Electricity Consumption Prediction of Multiple Campus Buildings Using Deep Recurrent Neural Networks

As the scale of electricity consumption grows, the peak electricity consumption prediction of campus buildings is essential for effective building energy system management. The selection of an appropriate model is of paramount importance to accurately predict peak electricity consumption of campus buildings due to the substantial variations in electricity consumption trends and characteristics among campus buildings. In this paper, we proposed eight deep recurrent neural networks and compared their performance in predicting peak electricity consumption for each campus building to select the best model. Furthermore, we applied an attention approach capable of capturing long sequence patterns and controlling the importance level of input states. The test cases involve three campus buildings in Incheon City, South Korea: an office building, a nature science building, and a general education building, each with different scales and trends of electricity consumption. The experiment results demonstrate the importance of accurate model selection to enhance building energy efficiency, as no single model’s performance dominates across all buildings. Moreover, we observe that the attention approach effectively improves the prediction performance of peak electricity consumption.

Seasonal characteristics of outdoor thermal comfort and residential electricity consumption: A Snapshot in Bangkok Metropolitan Area

Outdoor thermal comfort negatively influences urban residents’ health and increases residential electricity consumption (REC) for cooling demand. This study adopted the remote sensing-based modified temperature-humidity index (MTHI) to monitor and assess seasonal and spatial characteristics of urban thermal comfort and its association with REC. Thermal comfort and REC have a strong correlation (p < 0.05) and the same seasonal patterns as regional climate patterns. Thermal comfort is mainly controlled by land use, land cover (LULC), and physical characteristics of the built environments. Among the physical features, the building height is the most prominent element stimulating thermal discomfort as it regulates other elements linking to urban microclimates, such as sky view factor and ventilation. The thermal comfort in a particular region will be remarkably improved when urban density and building height remain under 65.43% and 17.88 meters, respectively. A green space proportion above 18.9% is also a reference value to optimize thermal comfort in built environments. The most vulnerable regions because of thermal discomfort in the Bangkok Metropolitan Area (BMA) account for 21.4% of the dense urban centers, with a building density of 72.5% and a population density of 17,173 persons/km2. The research findings enrich the current knowledge of thermal comfort characteristics and REC, which are significantly helpful for regional planning to mitigate thermal discomfort in this megacity.

A novel hybrid grey system forecasting model based on seasonal fluctuation characteristics for electricity consumption in primary industry

As a critical economic index, the annual electricity consumption plays an essential role in reflecting the domestic economic environment. For the sustainable development of the electricity power industry, an accurate electricity consumption forecasting model can not only guide the power generation work in advance but also help to determine the regulation law of electricity economic development. To accurately describe the seasonal fluctuation characteristics of electricity consumption in China's primary industry, Fourier modified grey forecasting model based on seasonal fluctuation characteristics and initial condition optimization (FDSGM(1,1,x(β),γ)) was put forward. Based on the empirical data of seasonal electricity consumption in China's primary industry, comparing with SGM(1,1), RSGM(1,1) and DSGM(1,1), the minimum mean absolute percentage errors (MAPEs) in the training set was reduced to 0.2 %, while that in the testing set was also reduced to 6.02 %, which was significantly better than those of the previous three forecasting models and was also markedly better than the prediction accuracy of reference [15]. The results indicated that FDSGM(1,1,x(β),γ) could identify seasonal fluctuations and variations and was a reliable seasonal forecasting tool that could be used to ensure sustainable development of electricity markets.

Characterizing load profile-based enterprise profiling under COVID-19 lockdown policy: A provincial case in China

The COVID-19 epidemic has led to devastating consequences worldwide due to strict social distancing measures, travel bans, city lockdowns, and other activity restrictions. Numerous lessons have been accumulated by various industries or sectors in the process of coping with COVID-19. In our paper, the actual electricity consumption of 1.145 million enterprises in a province of China is investigated in the normal period, the breakout period, and the recovery period. An artificial intelligence based enterprise profiling model is established to characterize power consumption profile geographically, temporally, and industrially. The recovery rate of enterprise power consumption is proposed as a key indicator representing the status of production resumption. Under lockdown measures, enterprise in different regions or sectors exhibits diversified responses in power consumption, the unsupervised learning model and the correlated coefficient extract the abstract characteristics and labels from various enterprise power profile. Ultimately, enterprises with similar electricity consumption characteristics are grouped and labeled to provide customized power services. Accurate enterprise profiling can effectively aid in facing with the challenges of insufficient orders, tight capital chains, and rising costs caused by the epidemic.

Understanding the characteristics of residential electricity consumption in Korea: Applying machine learning techniques to household-level energy survey data

Understanding the characteristics of residential electricity consumption in Korea: Applying machine learning techniques to household-level energy survey data

The potential of load identification and regulation in the textile industry

Firstly, this paper introduces the daily load characteristics and annual load characteristics of the textile industry and points out that the electricity load of the textile industry shows obvious seasonal changes. There is a certain degree of randomness. Then, this paper analyzes the behavior characteristics of electricity consumption in the textile industry, including the behavior mode of electricity consumption and the characteristics of electric equipment. It points out that the textile industry has a certain potential and space in energy utilization. Finally, this paper focuses on the load regulation situation of the textile industry and summarizes the load proportion, regulation mode, regulation time, and adjustable load proportion of the textile industry. This paper studies and discusses the energy utilization and load control of the textile industry, which has certain theoretical and practical significance.

Non-technical losses detection with Gramian angular field and deep residual network

Non-technical losses (NTL) refer to unrecorded power consumption generated by dishonest customers, which is a substantial issue affecting the power system stability and economic efficiency of the power grid. The detection of dishonest customers is hindered by the complexity of NTL such as data feature selection, retention time feature, and power consumption pattern judgment. This work addresses these issues using meter recording data, proposing an NTL detection approach with Gram’s angle field (GAF) and deep residual network (ResNet). Principal component analysis (PCA) method is applied to compress multiple electricity detection indexes, which aims to obtain multi-dimensional power consumption data characteristics without changing its timing characteristics. The GAF method is used to convert the time-series power features of individual users into a two-dimensional image, achieving the purpose of maintaining the user’s time-series features and user-based units. The images generated by the GAF method, which contain information about the electricity consumption characteristics of many customers, are classified by ResNet to highlight customers with NTL. The claimed algorithm was tested on a dataset consisting of both fraudulent and non-fraudulent subscriber data. The results demonstrated that the NTL detection method based on GAF and ResNet is superior to the traditional NTL detection method and has high accuracy.

T-LGBKS: An Interpretable Machine Learning Framework for Electricity Consumption Forecasting

Electricity is an essential resource that plays a vital role in modern society, and its demand has increased rapidly alongside industrialization. The accurate forecasting of a country’s electricity demand is crucial for economic development. A high-precision electricity forecasting framework can assist electricity system managers in predicting future demand and production more accurately, thereby effectively planning and scheduling electricity resources and improving the operational efficiency and reliability of the electricity system. To address this issue, this study proposed a hybrid forecasting framework called T-LGBKS, which incorporates TPE-LightGBM, k-nearest neighbor (KNN), and the Shapley additive explanation (SHAP) methods. The T-LGBKS framework was tested using Chinese provincial panel data from 2005 to 2021 and compared with seven other mainstream machine learning models. Our testing demonstrated that the proposed framework outperforms other models, with the highest accuracy (R2=0.9732). This study also analyzed the interpretability of this framework by introducing the SHAP method to reveal the relationship between municipal electricity consumption and socioeconomic characteristics (such as how changes in economic strength, traffic levels, and energy structure affect urban electricity demand). The findings of this study provide guidance for policymakers and assist decision makers in designing and implementing electricity management systems in China.

Projecting the Electricity Demand of the Residential Sector in Thailand During and After the COVID-19 Pandemic

The residential sector in Thailand has been a fast-growing energy consumption sector since 1995 at a rate of 6% per year. This sector makes a significant contribution to Thailand’s rising electricity demand especially during the COVID-19 pandemic. This study projects Thailand’s residential electricity consumption characteristics and the factors affecting the growth of electricity consumption using a system dynamics (SD) modeling approach to forecast long-term electricity consumption in Thailand. Furthermore, the COVID-19 pandemic and the lockdown can be seen as a forced social experiment, with the findings demonstrating how to use resources under particular circumstances. Four key factors affecting the electricity demand used in the SD model development include (1) work and study from home, (2) socio-demographic, (3) temperature changing, and (4) rise of GDP. Secondary and primary data, through questionnaire survey method, were used as data input for the model. The simulation results reveal that changing behavior on higher-wattage appliances has huge impacts on overall electricity consumption. The pressure to work and study at home contributes to rises of electricity consumption in the residential sector during and after COVID-19 pandemic. The government and related agencies may use the study results to plan for the electricity supply in the long term.

Customer characteristics analysis method based on the selection of electricity consumption characteristics and behavioral portraits of different groups of people

Because the traditional methods do not select the best feature collection in feature analysis, the accuracy and effectiveness of user feature clustering are not high, and the accuracy of user feature classification is not high. Therefore, this paper proposes a customer feature analysis method based on power consumption feature selection and behavior portrait of different people. The optimal feature set is obtained according to the maximum correlation and minimum redundancy criterion, and the user portrait task is described. The spatial feature domain classification method is used to classify the user portrait information, and the user label database is constructed according to the classification results. The AP clustering algorithm is used to cluster the power user portrait information and complete the customer feature analysis. Experimental results show that this method effectively improves the accuracy and effectiveness of user feature clustering, and the accuracy of user feature classification is high, indicating that the application effect is good.

Daily electrical energy consumption characteristics and design implications for off-grid homes on the Navajo Nation

Access to electricity remains elusive for over 700 million people worldwide. Although much attention has been given to solving this vexing problem in the context of communities in Sub-Saharan Africa and South Asia, often overlooked is that in even within so-called developed countries access to electricity might not be universal. It is estimated that more than 200,000 people living in North America do not have a connection to the electric grid. Many of these people live on Native American reservations or other Tribal Lands. Recently, there has been a concerted effort in the public, private, and non-profit sectors to electrify these homes with off-grid solar systems. This paper analyzes the electrical energy usage characteristics of 127 homes on the Navajo reservation in the southwestern United States. The homes have identical 3.8 kW solar arrays, 35.1 kWh 48 V gel lead acid battery banks, and 8 kW inverters. High-resolution inverter data was collected from these systems for a two-year period. Several statistical analyses were conducted on the data, including computing the statistical moments, creating empirical distributions of daily consumption, and analyzing consumption variation across different timescales. The results show average consumption of 2.78 kWh per day, but with a wide range of variation among homes. The implications of the analyses on pre-implementation system design, post-implementation interventions, and comparisons to incipient electricity consumption characteristics elsewhere in the world are discussed. The use of the average daily consumption in data-driven load estimation is evaluated and found to offer promising improvements over survey-based methods.

Day-ahead bidding strategy of cloud energy storage serving multiple heterogeneous microgrids in the electricity market

Cloud energy storage (CES) receives increasing attention as an efficient and viable paradigm for the provision of distributed energy storage services. This paper exploits CES’s service modes to both energy storage and electricity trading for its users, e.g., microgrid (MG). The optimal day-ahead bidding strategy is investigated for CES as an independent entity in the electricity market. Besides, two energy service modes are introduced considering MG's requirements and preferences. Each mode consists of a set of schemes for energy storage system (ESS) rent (power-based and capacity-based) and electricity trading (internal price with an improved pricing method and market clearing price). Finally, a stochastic programming (SP)-based optimization model is formulated to maximize the CES’s expected profits fully considering the electricity market’s settlement mechanism and uncertainty as well as the participation interest of MG. The proposed solution is extensively assessed through a case study of CES’s energy services to five heterogeneous MGs with distinct electricity generation and consumption characteristics and two energy service modes are assessed through comparative experiments. The numerical results confirm the effectiveness and benefits of the proposed optimal day-ahead bidding solution.

Data-driven intelligent method for detection of electricity theft

Anti-electricity leakage or stealing plays a crucial role in the energy market as accurate detection is beneficial for electricity safety, unit commitment, reduction of corporate losses, and so on. In a contemporary competitive energy market, it can significantly improve the efficiency of electric power enterprises and reduce the operational cost of the power system. In this study, through collecting historical data and analyzing the characteristics of electricity consumption, we established the index system for the detection of electricity leakage or stealing. Based on this, a focal loss-based 1d densely connected convolutional network, which we term the FLB-DCNNs, is proposed for the detection of electricity leakage or stealing. The electricity theft samples obtained in the actual on-site inspection are researched in the empirical analysis, and the results prove the effectiveness of the proposed approach. It attains an average precision of 98.51% for detecting electricity leakage or stealing users, and the average recall rate also reaches 98.17%. The recommended procedure provides a new idea for electricity theft detection and is easily transplanted to other related fields. Our data and code are available at https://github.com/xtu502/electricity-theft-detection.