Electricity Consumption Dataset Research Articles

Mining Customers’ Changeable Electricity Consumption for Effective Load Forecasting

Most existing approaches for electricity load forecasting perform the task based on overall electricity consumption. However, using such a global methodology can affect load forecasting accuracy, as it does not consider the possibility that customers’ consumption behavior may change at any time. Predicting customers’ electricity consumption in the presence of unstable behaviors poses challenges to existing models. In this article, we propose a principled approach capable of handling customers’ changeable electricity consumption. We devise a network-based method that first builds and tracks clusters of customer consumption patterns over time. Then, on the evolving clusters, we develop a framework that exploits long short-term memory recurrent neural network and survival analysis techniques to forecast electricity consumption. Our experiments on real electricity consumption datasets illustrate the suitability of the proposed approach.

Scenarios of future Indian electricity demand accounting for space cooling and electric vehicle adoption

India is expected to witness rapid growth in electricity use over the next two decades. Here, we introduce a custom regression model to project electricity consumption in India over the coming decades, which includes a bottom-up estimate of electricity consumption for two major growth drivers, air conditioning, and vehicle electrification. The model projections are available at a customizable level of spatial aggregation at an hourly temporal resolution, which makes them useful as inputs to long-term electricity infrastructure planning studies. The approach is used to develop electricity consumption data sets spanning various technology adoption and growth scenarios up to the year 2050 in five-year increments. The aim of the data is to provide a range of scenarios for India’s demand growth given new technology adoption. With long-term hourly demand projections serving as an essential input for electricity infrastructure modeling, this data publication enables further work on energy efficiency, generation, and transmission expansion planning for a fast-growing and increasingly important region from a global climate mitigation perspective.

Equipping Seasonal Exponential Smoothing Models with Particle Swarm Optimization Algorithm for Electricity Consumption Forecasting

Electricity consumption forecasting plays an important role in investment planning of electricity infrastructure, and in electricity production/generation and distribution. Accurate electricity consumption prediction over the mid/long term is of great interest to both practitioners and academics. Considering that monthly electricity consumption series usually show an obvious seasonal variation due to their inherent nature subject to temperature during the year, in this paper, seasonal exponential smoothing (SES) models were employed as the modeling technique, and the particle swarm optimization (PSO) algorithm was applied to find a set of near-optimal smoothing parameters. Quantitative and comprehensive assessments were performed with two real-world electricity consumption datasets on the basis of prediction accuracy and computational cost. The experimental results indicated that (1) whether the accuracy measure or the elapsed time was considered, the PSO performed better than grid search (GS) or genetic algorithm (GA); (2) the proposed PSO-based SES model with a non-trend component and additive seasonality term significantly outperformed other competitors for the majority of prediction horizons, which indicates that the model could be a promising alternative for electricity consumption forecasting.

An Improved Residential Electricity Load Forecasting Using a Machine-Learning-Based Feature Selection Approach and a Proposed Integration Strategy

Load forecasting (LF) has become the main concern in decentralized power generation systems with the smart grid revolution in the 21st century. As an intriguing research topic, it facilitates generation systems by providing essential information for load scheduling, demand-side integration, and energy market pricing and reducing cost. An intelligent LF model of residential loads using a novel machine learning (ML)-based approach, achieved by assembling an integration strategy model in a smart grid context, is proposed. The proposed model improves the LF by optimizing the mean absolute percentage error (MAPE). The time-series-based autoregression schemes were carried out to collect historical data and set the objective functions of the proposed model. An algorithm consisting of seven different autoregression models was also developed and validated through a feedforward adaptive-network-based fuzzy inference system (ANFIS) model, based on the ML approach. Moreover, a binary genetic algorithm (BGA) was deployed for the best feature selection, and the best fitness score of the features was obtained with principal component analysis (PCA). A unique decision integration strategy is presented that led to a remarkably improved transformation in reducing MAPE. The model was tested using a one-year Pakistan Residential Electricity Consumption (PRECON) dataset, and the attained results verify that the proposed model obtained the best feature selection and achieved very promising values of MAPE of 1.70%, 1.77%, 1.80%, and 1.67% for summer, fall, winter, and spring seasons, respectively. The overall improvement percentage is 17%, which represents a substantial increase for small-scale decentralized generation units.

Behavioural patterns in aggregated demand response developments for communities targeting renewables

Encouraging consumers to embrace renewable energies and energy-efficient technologies is at stake, and so the energy players such as utilities and policy-makers are opening up a range of new value propositions towards more sustainable communities. For instance, developments of turn-key demand response aggregation and optimisation of distributed loads are rapidly emerging across the globe in a variety of business models focused on maximising the inherent flexibility and diversity of the behind-the-meter assets. However, even though these developments’ added value is understood and of wide interest, measurement of the desired levels of consumer engagement is still on demonstration stages and assessment of technology readiness. In this paper, we analyse the characteristics of the loads, the behaviour of parameters, and in a final extent, the behaviour of each kind of consumer participating in aggregated demand scheduling. We apply both non-automatic and machine learning methods to extract the relevant factors and to recognise the potential consumer behaviour on a series of scenarios that are drawn using both synthetic data and living labs datasets. Our experimentation showcases a number of three patterns in which factors like the community’s demand volume and the consumer’s flexibility dominate and impact the performance of the tested development. The experimentation also makes current limitations arise within the existing electricity consumption datasets and their potential for inference and forecasting demand flexibility analytics.

A Stochastic Estimation Framework for Yearly Evolution of Worldwide Electricity Consumption

The determination of electric energy consumption is remarked as one of the most vital objectives for electrical engineers as it is highly essential in determining the actual energy demand made on the existing electricity supply. Therefore, it is important to find out about the increasing trend in electric energy demands and use all over the world. In this work, we present a prediction scheme for the progression of worldwide aggregates of cumulative electricity consumption using the time series of the records released annually for the net electricity use throughout the world. Consequently, we make use of an autoregressive (AR) model by retaining the best possible autoregression order recording the highest regression accuracy and the lowest standardized regression error. The resultant regression scheme was proficiently employed to regress and forecast the evolution of next-decade data for the net consumption of electricity worldwide from 1980 to 2019 (in billion kilowatt-hours). The experimental outcomes exhibited that the highest accuracy in regressing and forecasting the global consumption of electricity is 95.7%. The prediction results disclose a linearly growing trend in the amount of electricity issued annually over the past four decades’ observation for the global net electricity consumption dataset.

Research on Deep Generative Model Application for Shortterm Load Forecasting of Enterprise Electricity

This paper mainly applies deep generative models for short-term load forecasting on the enterprise electricity consumption dataset. After data cleaning on the electricity use dataset with the help of related weather data, we complement missing data and improve the data quality to better implement neural network generative prediction models. We build DeepAR and Wavenet as the representative of deep generative models. The main result is that deep generative models perform better compared with other baseline models, such as ARIMA, machine learning and baseline neural networks, no matter what accuracy metric and prediction horizon. Further improvement is to test in higher frequency electricity dataset with better quality.

A short-term hybrid forecasting model for time series electrical-load data using random forest and bidirectional long short-term memory

In the presence of the deregulated electric industry, load forecasting is more demanded than ever to ensure the execution of applications such as energy generation, pricing decisions, resource procurement, and infrastructure development. This paper presents a hybrid machine learning model for short-term load forecasting (STLF) by applying random forest and bidirectional long short-term memory to acquire the benefits of both methods. In the experimental evaluation, we used a Bangladeshi electricity consumption dataset of 36 months. The paper provides a comparative study between the proposed hybrid model and state-of-art models using performance metrics, loss analysis, and prediction plotting. Empirical results demonstrate that the hybrid model shows better performance than the standard long short-term memory and the bidirectional long short-term memory models by exhibiting more accurate forecast results.

Cloud Electricity Storage for Multi-Service Battery Operation: Case of RTE

Problem definition: We study a cloud storage operator that provides shared storage service for electricity end-users using the residual part of a multi-service grid-scale battery primarily used for high-priority grid services. We design an optimal product with regard to pricing and customer portfolio. Academic/practical relevance: A framework and solution approach for assessing and operating such multi-service battery operations with stochastic services and different priority levels is an open problem. Methodology: We model the problem as a two-stage stochastic optimization between high-priority stochastic grid services and low-priority cloud storage for stochastic end-users. We also propose operational metrics of multiplexing gain and probability of blocking to assess the operation of the multi-service multi-user battery. To address the computational challenge of solving the stochastic optimization with a large number of end-users, we propose effective capacity as a convex approximation that allows an analytical solution. We then provide an empirical analysis based on real grid congestion data from RTE France and a large dataset of end-users' electricity consumption in California. Results: Our empirical analysis shows (i) our proposed effective capacity is a close approximation, (ii) battery operation and profit are sensitive to the cost of external resources, number of end-users, and the leasing price of the battery, and (iii) with only a slight discount of the leasing price (~1%), the profit of the third party from a stochastic residual battery can be the same as that from a deterministic one. Managerial implications: Cloud storage as a low-priority service can profitably exist alongside other high-priority battery services, making integration of more storage in the grid economically viable and allowing larger intermittent renewables, a key path towards reduced carbon emissions.

Effect of weather on monthly electricity consumption in three coastal cities in West Africa

In several regions worldwide, demand for electricity can be highly dependent on weather conditions. This study investigates the relationships between weather and electricity consumption in three West African cities. Monthly electricity consumption datasets for the cities of Abidjan (Ivory Coast), Cotonou (Benin) and Lome (Togo) for the 1990–2015, 2000–2015 and 2008–2014 periods respectively were collected from national electricity companies, and meteorological data of the synoptic stations were used to compute Cooling Degree-Days in the three cities. The Cooling Degree-Days indices were estimated using air temperature and two temperature indices (the Humidex and the Heat Index). For the statistical analysis, classical multiplicative decomposition was applied to consumption data for subperiods for which consumption was considered to show relatively homogeneous evolutionary behavior (Abidjan and Lome from 2011 to 2014 and Cotonou from 2009 to 2014). Regardless of the temperature indices considered in the three cities, the Cooling Degree-Days indices are well correlated with the seasonal variability of power consumption and particularly, the peak consumption observed in March and the lower consumption in August. Slightly better correlations are obtained for Cotonou and Abidjan when the heat index (combining both temperature and relative humidity) are used to calculate the Cooling Degree-Days.

Electricity Theft Detection With Automatic Labeling and Enhanced RUSBoost Classification Using Differential Evolution and Jaya Algorithm

Electricity theft has emerged as one of the major reasons of Non-Technical Losses (NTLs) in the power distribution systems and has become a global issue. Therefore, power utilities are concerned about resolving the issue of electricity theft. In this regard, the data collected by Advanced Metering Infrastructure (AMI) can be used to devise data-driven machine learning-based Electricity Theft Detection (ETD) methods. In this paper, a novel data-driven ETD method is proposed that firstly labels the electricity consumers as fair or malicious based on three analyses: intra-consumer, inter-consumer, and temperature-electricity consumption relation. After assigning labels to the data, significant features are extraction from data by introducing a new feature extractor that is based on Reconstruction Independent Component Analysis (RICA) and sparse auto-encoder. Finally, classification is performed using two newly proposed enhanced classifiers, named as Differential Evolution (DE) Random Undersampling Boosting (DE-RUSBoost) and Jaya-RUSBoost. The performance of RUSBoost is enhanced using two nature-inspired swarm intelligence-based optimization algorithms, namely DE and Jaya optimization. The performance evaluation of the proposed classifiers is performed by conducting comprehensive simulations on real-world data taken from the UMass* smart homes electricity consumption dataset and the State Grid Corporation of China (SGCC) electricity theft dataset. DE-RUSBoost achieves an Area Under the Curve (AUC) of 0.89 and Jaya-RUSBoost achieves AUC of 0.95. The proposed classifiers have superior performance compared to two state-of-the-art benchmarks, i.e., Wide And Deep Convolution Neural Network (WADCNN) and grid search-based RUSBoost.

MORED: A Moroccan Buildings’ Electricity Consumption Dataset

This paper consists of two parts: an overview of existing open datasets of electricity consumption and a description of the Moroccan Buildings’ Electricity Consumption Dataset, a first of its kind, coined as MORED. The new dataset comprises electricity consumption data of various Moroccan premises. Unlike existing datasets, MORED provides three main data components: whole premises (WP) electricity consumption, individual load (IL) ground-truth consumption, and fully labeled IL signatures, from affluent and disadvantaged neighborhoods. The WP consumption data were acquired at low rates (1/5 or 1/10 samples/s) from 12 households; the IL ground-truth data were acquired at similar rates from five households for extended durations; and IL signature data were acquired at high and low rates (50 k and 4 samples/s) from 37 different residential and industrial loads. In addition, the dataset encompasses non-intrusive load monitoring (NILM) metadata.

Detection of electricity theft using data processing and LSTM method in distribution systems

Electricity theft is a big problem faced by all energy distribution services and continues to rising. Therefore, studies on electricity theft detection techniques have increased in recent years. Unsuitable calibration and illegal calibration of energy meters during production may cause non-technical losses. Non-technical losses have been a major concern for the resulting security risks and the immeasurable loss of income. In most of the meter tampered locations, damaged meter terminals and/or illegal applications cannot be distinguishable during checking. In fact, electric distribution companies will never be able to eliminate electricity theft. But it is possible to take measure to detect, prevent and reduce it. In this paper, we developed by using deep learning methods on real daily electricity consumption data (Electricity consumption dataset of State Grid Corporation of China). Data reduction has been made by developing a new method to make the dataset more usable and to extract meaningful results. A Long Short-Term Memory (LSTM) based deep learning method has been developed for the dataset to be able to recognize the actual daily electricity consumption data of 2016. In order to evaluate the performance of the proposed method, the accuracy, prediction and recall metric was used by considering the five cross-fold technique. Performance of the proposed methods were found to be better than previously reported results.

COMPARISON OF LINEAR REGRESSIONS AND NEURAL NETWORKS FOR FORECASTING ELECTRICITY CONSUMPTION

Electricity has a major role in humans that is very necessary for daily life. Forecasting of electricity consumption can guide the government's strategy for the use and development of energy in the future. But the complex and non-linear electricity consumption dataset is a challenge. Traditional time series models in such as linear regression are unable to solve nonlinear and complex data-related problems. While neural networks can overcome the problems of nonlinear and complex data relationships. This was proven in the experiments in this study. Experiments carried out with linear regressions and neural networks on the electricity consumption dataset A and the electricity consumption dataset B. Then the RMSE results are compared on the linear regressions and neural networks of the two datasets. On the electricity consumption dataset, A obtained by RMSE of 0.032 used the linear regression, and RMSE of 0.015 used the neural network. On the electricity consumption, dataset B obtained by RMSE of 0.488 used the linear regression, and RMSE of 0.466 used the neural network. The use of neural networks shows a smaller RMSE value compared to the use of linear regressions. This shows that neural networks can overcome nonlinear problems in the electricity consumption dataset A and the electricity consumption dataset B. So that the neural networks are afforded to improve performance better than linear regressions. This study to prove that there is a nonlinear relationship in the electricity consumption dataset used in this study, and compare which performance is better between using linear regression and neural networks.

Energy load time-series forecast using decomposition and autoencoder integrated memory network

With the increasing population and rising living standard, the demand for energy and materials have increased to a greater extent. The accurate estimation of increasing electricity demand is prerequisite for strategies planning, improving revenue, reducing power wastage and stable operation of the energy demand management system. Recent advancements in the field of electricity load forecasting provide powerful tools to capture non-linear energy demand trends and outperform conventional load prediction models. However, the existing demand prediction models suffer from some significant shortcomings that need to be addressed for improved prediction accuracy. In this context, the current research work proposes a deep learning based hybrid approach which firstly implements Variational Mode Decomposition (VMD) and Autoencoder models to extract meaningful sub-signals/features from the data. Subsequently, a Long Short-term Memory (LSTM) network model is trained for each sub-signal to forecast electricity demand by utilizing historical, seasonal and timestamp data dependencies. The support for incorporating seasonal and timestamp information to LSTM model is provided through the agglomerative clustering algorithm. Furthermore, an error variance modelling strategy is also employed to enhance the prediction accuracy of the proposed approach. The experiments are conducted on electricity consumption dataset of Himachal Pradesh, India. Performance assessment of the proposed approach is made by comparing prediction results with Support Vector Regression (SVR), Recurrent Neural Network (RNN), Deep Belief Network (DBN) and EMD+LSTM models. The experimental results demonstrate that the proposed model outperforms other state-of-the-art demand forecasting models and has the lowest MAPE (3.04%).

A disaggregated approach to energy-growth nexus: Micro-regional view

We explore the relationship between economic growth and energy consumption using regional level data for 2004–2014 period. Data on electricity consumption for regional distributional areas cannot be used with the economic data for county/regional level since the borders of regional electricity distributional areas do not correspond to administrative NUTS-3 borders. We use a proprietary dataset of actual electricity consumption for Croatian counties which allows us to revisit the energy-growth nexus dilemma from a disaggregated viewpoint. We find significant policy implications stemming from comovement between counties and long term positive influence of energy, capital and labour on industrial output.

Modeling and Analysis of Residential Electricity Consumption Statistics: A Tracy-Widom Mixture Density Approximation

Residential electricity loads highly fluctuate which makes it challenging to predict and estimate changes and anomalies. This article studies the statistical distribution of the maximum load of residential settlements and proposes a new model that is more reliable than conventional methods to predict extreme events in residential electricity consumption. Specially, a multimodal Tracy-Widom distribution is proposed to characterize the maximum residential electricity consumption data. We also propose a numerical method to approximate the density function of the multimodal Tracy-Widom distribution as opposed to the conventional approach in the literature, where a multimodal normal distribution is utilized. A simulated electricity consumption data set is used to test and validate the proposed methods. The results demonstrate that the multimodal Tracy-Widom distribution is more accurate than conventional methods in modeling residential electricity consumption data. In addition, the numerical results show that residential electricity consumption behavior is determined by certain socioeconomic factors for which correlations with household electricity consumption exist.

Cultural Algorithm Initializes Weights of Neural Network Model for Annual Electricity Consumption Prediction

The accurate prediction of annual electricity consumption is crucial in managing energy operations. The neural network (NN) has achieved a lot of achievements in yearly electricity consumption prediction due to its universal approximation property. However, the well-known back-propagation (BP) algorithms for training NN has easily got stuck in local optima. In this paper, we study the weights initialization of NN for the prediction of annual electricity consumption using the Cultural algorithm (CA), and the proposed algorithm is named as NN-CA. The NN-CA was compared to the weights initialization using the other six metaheuristic algorithms as well as the BP. The experiments were conducted on the annual electricity consumption datasets taken from 21 countries. The experimental results showed that the proposed NN-CA achieved more productive and better prediction accuracy than other competitors. This result indicates the possible consequences of the proposed NN-CA in the application of annual electricity consumption prediction.

Evolutionary model construction for electricity consumption prediction

Nowadays the ever-increasing energy consumption in buildings has caused supply shortages and adverse environmental impacts. The accurate prediction of energy consumption in smart buildings may help to monitor and control energy usage. As energy consumption is inevitably affected by exogenous factors such as temperature and wind speed, it is fundamentally important to select the informative channels of the factors, to extract the valuable features from the selected channels applied to the optimal-configured model, to improve prediction accuracy. However, existing work considers these parts in an almost disjoint way and lacks a model taking them into account, which may decrease prediction performance. Motivated by this challenge, an end-to-end prediction framework, called evolutionary model construction (EMC), is proposed to focus on performing these parts jointly. To implement EMC, a two-step evolutionary algorithm (EA) is designed, where one EA is firstly used to focus on exploiting the informative channels, while a new algorithm is proposed to concentrate on selecting the suitable feature extraction methods and respective time window sizes applied to the selected channels, and selecting the parameters in the predictor. The implementation of EMC chooses neural network with random weights as the predictor due to its highly recognized efficacy. We evaluate EMC in comparison with the existing approaches on a real-world electricity consumption dataset with various auxiliary factors. The superiority of EMC is further proved by analyzing and discussing the result according to the days in 1 week, time stamps in 1 day and month information on test samples.

Nonnegative Matrix Factorization with Side Information for Time Series Recovery and Prediction

Motivated by the reconstruction and the prediction of electricity consumption, we extend Nonnegative Matrix Factorization~(NMF) to take into account side information (column or row features). We consider general linear measurement settings, and propose a framework which models non-linear relationships between features and the response variables. We extend previous theoretical results to obtain a sufficient condition on the identifiability of the NMF in this setting. Based the classical Hierarchical Alternating Least Squares~(HALS) algorithm, we propose a new algorithm (HALSX, or Hierarchical Alternating Least Squares with eXogeneous variables) which estimates the factorization model. The algorithm is validated on both simulated and real electricity consumption datasets as well as a recommendation dataset, to show its performance in matrix recovery and prediction for new rows and columns.