Forecasting Electricity Consumption Research Articles

Consumption prediction with privacy concern: Application and evaluation of Federated Learning

In this paper, we propose privacy-friendly electricity consumption prediction models based on Federated Learning (FL). Federated Learning provides a novel framework for Artificial Neural Network training. It decouples data storage from model training from decentralized data sources. A central server does not need to collect raw training data from individual clients but only the model parameters, like the gradients or weights updates. The sensitive raw data can be safely stored and used by end-users themselves. We improve the accuracy of the FL model by first clustering households, and training a personalized model for each cluster. We also analyze the Deep Leakage from Gradients (DLG) attack in our study case with three scenarios. Our simulations suggest that the DLG attack can barely succeed in consumption prediction. In this way, the FL model can guarantee clients’ privacy by design. We use a large-scale dataset that contains 3590 households’ 1.5 years of consumption to test the FL model’s performance. Several clustering algorithms are tested for the following experiments. To comprehensively test the FL model’s performance, we propose several popular neural network models: the simple Deep Neural Network, Long Short-Term Memory, Convolutional Neural Network, and WaveNet. These models are both trained under the centralized and federated framework. The federated trained models slightly sacrifice the model’s accuracy while guaranteeing the client’s data privacy. Meanwhile, the federated model shows remarkable scalability. Under the FL framework, new clients can obtain their prediction 6 times faster. Moreover, the federated model has strong robustness against missing or damaged training data. With a certain percentage of missing data in the training set, the centralized model’s accuracy gets 27% worse, while the federated model’s accuracy only gets 2% worse.

Deep Time Series Forecasting Models: A Comprehensive Survey

Deep learning, a crucial technique for achieving artificial intelligence (AI), has been successfully applied in many fields. The gradual application of the latest architectures of deep learning in the field of time series forecasting (TSF), such as Transformers, has shown excellent performance and results compared to traditional statistical methods. These applications are widely present in academia and in our daily lives, covering many areas including forecasting electricity consumption in power systems, meteorological rainfall, traffic flow, quantitative trading, risk control in finance, sales operations and price predictions for commercial companies, and pandemic prediction in the medical field. Deep learning-based TSF tasks stand out as one of the most valuable AI scenarios for research, playing an important role in explaining complex real-world phenomena. However, deep learning models still face challenges: they need to deal with the challenge of large-scale data in the information age, achieve longer forecasting ranges, reduce excessively high computational complexity, etc. Therefore, novel methods and more effective solutions are essential. In this paper, we review the latest developments in deep learning for TSF. We begin by introducing the recent development trends in the field of TSF and then propose a new taxonomy from the perspective of deep neural network models, comprehensively covering articles published over the past five years. We also organize commonly used experimental evaluation metrics and datasets. Finally, we point out current issues with the existing solutions and suggest promising future directions in the field of deep learning combined with TSF. This paper is the most comprehensive review related to TSF in recent years and will provide a detailed index for researchers in this field and those who are just starting out.

Applicability of data-driven methods in modeling electricity demand-climate nexus: A tale of Singapore and Hong Kong

Predicting electricity demand at city level is vital for stakeholders. There are various data-driven methods in electricity consumption prediction, but their applicability to forecast monthly electricity demand has yet to be systematically explored. This study examines several widely used data-driven methods, namely multiple linear regression (MLR), machine learning (ML) method including support vector machine (SVM) and random forest (RF), and deep learning (DL) method including long short-term memory network (LSTM) and LSTM-gated recurrent unit (GRU), for long-term energy prediction using climatic and historical electricity datasets in Singapore (2005–2019) and Hong Kong (1975–2019). In Singapore, ML outperforms other methods in terms of statistical criteria and time series plots, and SVM provides the best accuracy with mean absolute percentile error (MAPE) of 2.55 %, while MLR exhibits the worst accuracy with MAPE of 3.1 %. In Hong Kong, DL surpasses ML in terms of statistical criteria, time series stability, and generalization ability. MLR achieves the best overall prediction accuracy with R2 up to 0.95 but shows poor ability for predicting peak and low electricity consumption, while LSTM exhibits the smallest bias in these months. RF shows strong overfitting issues in both cities. Overall, SVM and LSTM are recommended for small and large datasets, respectively.

A federated and transfer learning based approach for households load forecasting

The escalating global population and increased smart-devices demand have made electricity load estimation a significant challenge for relevant authorities. Even with advancements in Artificial Intelligence, pinpointing both immediate and long-term electricity requirements remain a challenging and complex task. Predicting electricity usage at the household level can offer insights into this trend, aiding in the development of efficient load management strategies. However, creating individual predictive models for every household is a resource-intensive task. The current research proposes a streamlined approach that merges clustering analysis with federated and transfer learning to craft household-level predictive models. Firstly, clustering analysis is employed to discern electricity usage patterns, forming cluster-specific datasets. Subsequently, a federated learning model is introduced to each identified cluster, ensuring that the foundational model is finely calibrated to the unique consumption patterns inherent to each group. To evaluate the effectiveness of our approach, we employed Root Mean Square Error and Mean Absolute Error as evaluation metrics. During the evaluation phase, the transfer learning approach is employed, fine-tuning our root model on 70% of the data and testing its performance on the remaining 30%. The robustness of the proposed methodology is validated using a comprehensive dataset detailing electricity usage across Australian households. When bench-marked against prevailing methodologies, our proposed model consistently demonstrates either closer or superior predictive accuracy, emphasizing its potential as a formidable tool in real-time electricity consumption forecasting.

Day-Ahead Electricity Consumption Prediction of Individual Household–Capturing Peak Consumption Pattern

Day-Ahead Electricity Consumption Prediction of Individual Household–Capturing Peak Consumption Pattern

Enhancing hourly electricity forecasting using fuzzy cognitive maps with sample entropy

Accurate electricity consumption forecasting is the crucial to ensure the stable operation of the power system and optimize the smart grid dispatch. However, existing methods may not effectively capture complex potential features and change patterns, and are susceptible to noise, leading to deficiencies such as suboptimal electricity consumption forecasting accuracy, poor stability, and weak generalization capabilities. To fill this gap, a high order fuzzy cognitive map (FCM) forecasting method based on sample entropy and minimax concave penalty (MCP) regularization is proposed. Firstly, two mode decomposition methods are innovatively fused by utilizing sample entropy to extract the complex features of electricity consumption data, then a high order FCM model with powerful system state prediction capability and interpretable knowledge representation is constructed, and finally a FCM learning method based on MCP regularization is proposed. The forecasting method has the merits of low forecasting error, high stability and strong generalization capabilities. Comprehensive experimental results on multiple electricity consumption datasets show that the proposed method achieves optimal performance compared to 10 state-of-the-art baseline methods, demonstrating effectiveness in the task of electricity consumption forecasting, which contributes to more efficient energy management and cost savings, and provides a reliable tool for smart grid planning and management.

An advanced framework for net electricity consumption prediction: Incorporating novel machine learning models and optimization algorithms

In recent years, the escalating demand for electric energy has underscored the need for robust prediction models capable of accurately anticipating consumption patterns. The imperative lies in enabling utilities and policymakers to optimize resource allocation, strategically plan infrastructure development, and ensure the stability and efficiency of the power grid. This study undertakes a comprehensive comparative analysis of machine learning techniques employed in predicting net electricity consumption in Turkey. The primary goal is to augment the accuracy and performance of electricity load forecasting, thereby contributing to effective energy management and fostering sustainable development within the power sector. Two machine learning models, including CatBoost and Extreme Gradient Boosting (XGBoost), are strategically integrated with optimization algorithms such as Sparrow Search Algorithm (SSA), Phasor Particle Swarm Optimization (PPSO), and Hybrid Grey Wolf Optimization (GWO). The core analysis centers on evaluating the performance of these integrated models based on key accuracy metrics and runtime efficiency. Notably, the results underscore that the XGBoost-SSA model emerges as the superior performer, exhibiting heightened accuracy and superior performance in predicting electricity consumption. This model showcases the highest coefficient of determination (R2) value and demonstrates lower errors during the testing phase, thereby presenting a promising and effective approach for electricity consumption prediction in the specific context of Turkey.

Prediction of industrial electricity consumption based on grey cluster weighted Markov model

Accurate prediction of industrial electricity consumption is not only beneficial to maintaining the steady development of the economy but also to conserving energy. To improve the prediction accuracy of industrial electricity consumption, a grey cluster weighted Markov model is proposed. It is applied to predict the industrial electricity consumption in four different regions in China. The prediction results are compared with the traditional discrete grey prediction model, which shows that the present model is more effective in these aspects of prediction accuracy, stability and extensibility. The research can provide theoretical references for the “West-East electricity transmission project” in China.

Internet of Vehicles (IoV) Based Framework for electricity Demand Forecasting in V2G

The integration of smart grids with Advanced Metering Infrastructure (AMI) has bridged the realms of the Internet of Vehicles (IoV) and Electric Vehicles (EVs), yet challenges persist in managing EV Battery Range and charging infrastructure effectively. This paper presents an innovative IoV-based framework tailored for EVs, with a specific focus on forecasting electricity consumption in a Vehicle-to-Grid (V2G) scenario. By exploring the hurdles surrounding electric vehicle usage, the research lays the foundation for Electric Vehicles (EVs). The proposed IoV model optimizes IoT sensor utilization through a fog layer and employs a Back Propagation (BP) neural network for battery State of Charge (SoC) estimation, integrating Principal Component Analysis (PCA) for data dimensionality reduction. Leveraging substantial computing capabilities, the cloud layer predicts Electricity Consumption Data (ECD) associated with EVs in V2G scenarios. Performance evaluation metrics like Akaike Information Criteria (AIC), Mean Error (ME), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), and Root Mean Squared Error (RMSE) are assessed across state-of-the-art forecasting algorithms. Incorporating EV potential into the optimal model reveals a significant 10% reduction in electricity demand. This research advances IoV-based frameworks, offering insights to enhance EV efficiency within the broader energy infrastructure.

Machine learning for electric energy consumption forecasting: Application to the Paraguayan system

Abstract In this paper we address the problem of short-term electric energy prediction using a time series forecasting approach applied to data generated by a Paraguayan electricity distribution provider. The dataset used in this work contains data collected over a three-year period. This is the first time that these data have been used; therefore, a preprocessing phase of the data was also performed. In particular, we propose a comparative study of various machine learning and statistical strategies with the objective of predicting the electric energy consumption for a given prediction horizon, in our case seven days, using historical data. In this paper we have tested the effectiveness of the techniques with different historical window sizes. Specifically, we considered two ensemble strategies, a neural network, a deep learning technique and linear regression. Moreover, in this study, we tested whether the inclusion of meteorological data can help achieve better predictions. In particular, we considered data regarding temperature, humidity, wind speed and atmospheric pressure registered during the three-year period of data collection. The results show that, in general, the deep learning approach obtains the best results and that such results are obtained when meteorological data are also considered. Moreover, when meteorological data is used, a smaller historical window size is required to obtain precise predictions.

School Electricity Consumption in a Small Island Country: The Case of Fiji

Electricity consumption in buildings is one of the major causes of energy usage and knowledge of this can help building owners and users increase energy efficiency and conservation efforts. For Pacific Island countries, building electricity demand data is not readily accessible or available for constructing models to predict electricity demand. This paper starts to fill this gap by studying the case of schools in Fiji. The aim of the paper is to assess the factors affecting electricity demand for grid-connected Fijian schools and use this assessment to build mathematical models (multiple linear regression (MLR) and artificial neural network (ANN)) to predict electricity consumption. The average grid-connected electricity demand in kWh/year was 1411 for early childhood education schools, 5403 for primary schools, and 23,895 for secondary schools. For predicting electricity demand (ED) for all grid-connected schools, the stepwise MLR model shows that taking logarithm transformations on both the dependent variable and independent variables (number of students, lights, and air conditioning systems) yields statistically significant independent variables with an R2 value of 73.3% and RMSE of 0.2248. To improve the predicting performance, ANN models were constructed on both the natural form of variables and transformed variables. The optimum ANN model had an R2 value of 95.3% and an RMSE of 59.4 kWh/year. The findings of this study can assist schools in putting measures in place to reduce their electricity demand, associated costs, and carbon footprint, as well as help government ministries make better-informed policies.

A Novel Deep Learning-Based Data Analysis Model for Solar Photovoltaic Power Generation and Electrical Consumption Forecasting in the Smart Power Grid

With the installation of solar panels around the world and the permanent fluctuation of climatic factors, it is, therefore, important to provide the necessary energy in the electrical network in order to satisfy the electrical demand at all times for smart grid applications. This study first presents a comprehensive and comparative review of existing deep learning methods used for smart grid applications such as solar photovoltaic (PV) generation forecasting and power consumption forecasting. In this work, electrical consumption forecasting is long term and will consider smart meter data and socioeconomic and demographic data. Photovoltaic power generation forecasting is short term by considering climatic data such as solar irradiance, temperature, and humidity. Moreover, we have proposed a novel hybrid deep learning method based on multilayer perceptron (MLP), long short-term memory (LSTM), and genetic algorithm (GA). We then simulated all the deep learning methods on a climate and electricity consumption dataset for the city of Douala. Electrical consumption data are collected from smart meters installed at consumers in Douala. Climate data are collected at the climate management center in the city of Douala. The results obtained show the outperformance of the proposed optimized method based on deep learning in the both electrical consumption and PV power generation forecasting and its superiority compared to basic methods of deep learning such as support vector machine (SVM), MLP, recurrent neural network (RNN), and random forest algorithm (RFA).

Long Term Forecasting of Peak Demand and Annual Electricity Consumption of the West African Power Pool Interconnected Network by 2032

The uneven distribution of primary sources of electric power generation in Economic Community of West African States (ECOWAS) compelled the heads of states to create the West African Power Pool (WAPP). The vision of this system is to set up a common electrical energy market to satisfy the balance between supply and demand at an affordable price using the interconnected network. Forecasting maximum power demand and energy consumption is essential for planning and the coordination of new power plant and transmission lines building. This work consists of predicting maximum power demand and total energy that must transit through the WAPP interconnected network by the year 2032. We compare the performances of three time series models namely the Long Short-Term Memory (LSTM), Auto-Regressive Integrated Moving Average (ARIMA) and Fb Facebook Prophet. Electric power and energy data used for training the systems comes from the WAPP authorties. The results show that, for monthly peaks, the Facebook (Fb) Prophet model is the best, with a MAPE (mean absolute error percentage) of 3.1% and a low RMSE (root mean square error) of 1.225 GW. For energy prediction, ARIMA performances are the best compared to others with (RMSE 1.20 TWh, MAPE 1.00%). Thus, the forecast for total annual energy consumption and annual peak demand will be, respectively, 96.85TWh and 13.6 GW in 2032.

Empowering Sustainability: A Consumer-Centric Analysis Based on Advanced Electricity Consumption Predictions

This study addresses the critical challenge of accurately forecasting electricity consumption by utilizing Exponential Smoothing and Seasonal Autoregressive Integrated Moving Average (SARIMA) models. The research aims to enhance the precision of forecasting in the dynamic energy landscape and reveals promising outcomes by employing a robust methodology involving model application to a large amount of consumption data. Exponential Smoothing demonstrates accurate predictions, as evidenced by a low Sum of Squared Errors (SSE) of 0.469. SARIMA, with its seasonal ARIMA structure, outperforms Exponential Smoothing, achieving lower Mean Absolute Percentage Error (MAPE) values on both training (2.21%) and test (2.44%) datasets. This study recommends the adoption of SARIMA models, supported by lower MAPE values, to influence technology adoption and future-proof decision-making. This study highlights the societal implications of informed energy planning, including enhanced sustainability, cost savings, and improved resource allocation for communities and industries. The synthesis of model analysis, technological integration, and consumer-centric approaches marks a significant stride toward a resilient and efficient energy ecosystem. Decision-makers, stakeholders, and researchers may leverage findings for sustainable, adaptive, and consumer-centric energy planning, positioning the sector to address evolving challenges effectively and empowering consumers while maintaining energy efficiency.

Time-Stratified Analysis of Electricity Consumption: A Regression and Neural Network Approach in the Context of Turkey

This study aims to apply seasonality and temporal effects in the analysis of electricity consumption in Turkey as a case mixed with regression and neural network methodologies. The study goal is to increase knowledge about the features and trending forces behind electricity usage which provide informed recommendations for smart energy planning and regulation. Comparing and contrasting the regression and neural network models makes it possible to carry out a thorough analysis of the merits and demerits of each model. Moreover, the examination of the limits of the models and their performance in forecasting electricity consumption patterns over the long term is done. The results of this study have a significant impact on power forecasting techniques, and they have meaningful effects on the policymakers, planners and utilities in Turkey. Understanding the story of the use of electricity around the world is very important for the development of sustainable energy policies, resource provision, and the maintenance of reliable and smart energy networks in the country.

Time series forecasting of electricity consumption using hybrid model of recurrent neural networks and genetic algorithms

The forceful energy efficiency to manage the demand is essential to meet development goals. Palestine has suffered from an electricity deficit, whereas the city of Tulkarm suffers from a chronic one. The dataset was collected from Tulkarm city in Palestine; this city is considered one of the cities that suffers the most from frequent power outages. It's difficult to determine the most powerful Artificial intelligence (AI) approaches that can accurately forecast electricity consumption. This paper presents a hybrid model that combines Recurrence Neural Networks (RNNs) and Genetic Algorithms (GAs) [RNN-GAs] to forecast electricity consumption and optimize demand. In the proposed model the K-means clustering technique produces specific initial population seeding and optimization crossover operators to enhance the efficiency and find the optimal solution. The results showed that the proposed Nonlinear Autoregressive with External (Exogenous) (NARX) (NARX-GAs) with the K-means clustering technique outperforms the hybrid model NARX-GAs. The NARX-GAs-K Mean Clustering recorded an RMSE value of 0.08759, which performs a good balance with the lowest RMSE, especially in long-term forecasting, and also outperforms the other hybrid forecasting models that depend on RNN-GAs. Finally, the forecasting results of the hybrid NARX-GAs-K Mean Clustering can predict accurately the energy consumption in a city, which leads to the use of the model in similar cities to forecast and manage the demand for electricity consumption.

Load Balance Forecasting Based on Hybrid Deep Neural Network

Load forecasting is the foundation of utility design, and it is a fundamental business problem in the utility industry. Load forecasting, mainly referring to forecasting electricity demand and energy, is being used throughout all segments of the electric power industry, including generation, transmission, distribution, and retail. In this paper, a long short-term memory network with a hybrid approach is improved with a dense algorithm and proposed for electricity load forecasting. A long short-term memory network is designed to effectively exhibit the dynamic behavior of load time series. The proposed model is tested for Panama study including historical data and weather variables. The prediction accuracy is validated by performance metrics, and the best of the metrics are attained when mean absolute error is 5.262, mean absolute percentage error 0.0000376, and root mean square error 18.243. The experimental results show a high prediction rate for load balance forecasting of electric power consumption.

A discrete time-varying grey Fourier model with fractional order terms for electricity consumption forecast

Accurate electricity consumption forecast plays an important role in managing and operating a country’s power system and is a challenging task due to the data exhibiting the mixture characteristics of trend, periodicity and randomness. To accurately describe the dynamic changes of seasonal fluctuations of electricity consumption, a novel discrete time-varying grey Fourier model with fractional order terms is proposed by coupling Fourier functions with fractional order time-varying terms, as the grey action, to fit the amplitude variations hidden in seasonal pattern and improve the fitting and forecasting accuracy. Moreover, the mechanism and properties of the proposed model are discussed. Then the optimal fractional order is determined by the Dingo Optimization Algorithm (DOA) and the optimal Fourier order is selected by the hold-out method to enhance the robustness of the model. Subsequently, the results of designed experiments based on Monte-Carlo method testify the validity of the Fourier order selection method and the proposed model. Finally, this model is applied to predict the monthly residential electricity consumption of China from 2015 to 2022. The results indicate that the proposed model captures the dynamic amplitude variations of the time series and has better prediction performances than other benchmark grey prediction models and non-grey prediction models.

Deep learning-based forecasting of electricity consumption

Building energy management systems (BEMS) are integrated computerized systems that track and manage the energy use of many pieces of building-related machinery and equipment, including lighting, power systems, and HVAC systems. Modern buildings must have BEMSs in order to reduce energy usage while maintaining comfort. Not only for energy-saving purposes, BEMS is essential in enhancing the quality of the energy supply, which helps to gain a better understanding of how energy is used and the building's energy usage. When the dynamics of a building's energy usage are known, it is possible to determine which changes are most likely to reduce consumption. Numerous connected devices, operating modes, energy usage, and environmental factors can all be monitored and controlled in real-time using BEMS. Changing operating times and setting points to maximize comfort and efficiency is made simple by this. In this paper, we have primarily addressed the two significant issues of model optimization and electricity consumption forecasts. Future forecasting has been done using the LSTM based time series approach. We generated data on the amount of electricity consumed by a hospital facility and tested our suggested methodologies on actual data. The findings gained demonstrated that the strategies were successful with both types of data. On actual data, the trend in electricity consumption can be accurately predicted. Several model optimizers enhanced the suggested methods' performance as well. Our objective function gain accuracy result of 95%.

Forecasting electricity consumption in China's Pearl River Delta urban agglomeration under the optimal economic growth path with low-carbon goals: Based on data of NPP-VIIRS-like nighttime light

To reflect the trend of electricity consumption (EC) in China's Pearl River Delta (PRD) under a balanced environment and economy, this paper proposes an EC forecasting framework from the perspective of optimal economic growth. The historical GDP statistics of the PRD are calibrated with data from the new "NPP-VIIRS-like" nighttime light dataset from 2000 to 2020, and two simulation scenarios of optimal economic growth are constructed according to the carbon emission reduction rate. Finally, based on the electricity data from 2000 to 2017, an error correction model is constructed to predict and compare the trend of EC in the PRD under different carbon emission scenarios. The results demonstrate the following. (1) The economic growth path under low-carbon constraints is more closely aligned with the actual economic development of the PRD. (2) The electricity demand required to sustain optimal economic growth in the PRD under low-carbon constraints is projected to reach saturation around 2037, approximately a decade earlier than the scenario without carbon constraints. The results of the projections are expected to guide future work on power system planning and economic development assessment in the context of carbon reduction.