Electric Forecasting Research Articles

Two-Stage Combined Model for Short-Term Electricity Forecasting in Ports

Two-Stage Combined Model for Short-Term Electricity Forecasting in Ports

Enhancing Electricity Demand Forecasting Accuracy Through Hybrid Models and Deep Learning Techniques: A Systematic Literature Review

Abstract: This reviewed literature on electricity forecasting covers its history, terminology, and techniques. A systematic review of existing studies highlighted key findings and future research opportunities. Conventional statistical techniques and MLA can predict electricity demand over time with various techniques and forecasting windows tailored to data and problem specifics. Most studies focused on STLF, often without testing techniques on MTLF and LTLF. The key findings include: Many studies (26%) used conventional statistical methods like ARIMA, ARIMAX, and SARIMAX for electricity forecasting, often without benchmarking algorithms. Various factors, such as time, weather, electricity price, population, and economy, influence ELF. Weather parameters were the most commonly used predictors, though performance varied across studies. A global increase in electricity demand has driven numerous studies, though less research has been done in low- and middle-income countries. Deep neural networks like LSTM have been underutilised in electricity forecasting. LSTM's ability to store memory and address the vanishing gradient problem makes it promising for future research, particularly in hybrid models combining CNN and LSTM for forecasting peak load demand based on economic and environmental factors.

Novel Custom Loss Functions and Metrics for Reinforced Forecasting of High and Low Day-Ahead Electricity Prices Using Convolutional Neural Network–Long Short-Term Memory (CNN-LSTM) and Ensemble Learning

Day-ahead electricity price forecasting (DAEPF) is vital for participants in energy markets, particularly in regions with high integration of renewable energy sources (RESs), where price volatility poses significant challenges. The accurate forecasting of high and low electricity prices is particularly essential, as market participants seek to optimize their strategies by selling electricity when prices are high and purchasing when prices are low to maximize profits and minimize costs. In Japan, the increasing integration of RES has caused day-ahead electricity prices to frequently fall to almost zero JPY/kWh during periods of high RES output, creating significant profitability challenges for electricity retailers. This paper introduces novel custom loss functions and metrics specifically designed to improve the forecasting accuracy of extreme prices (high and low prices) in DAEPF, with a focus on the Japanese wholesale electricity market, addressing the unique challenges posed by the volatility of RES. To implement this, we integrate these custom loss functions into a Convolutional Neural Network–Long Short-Term Memory (CNN-LSTM) model, augmented by an ensemble learning approach and multimodal features. The proposed custom loss functions and metrics were rigorously validated, demonstrating their effectiveness in accurately predicting high and low electricity prices, thereby indicating their practical application in enhancing the economic strategies of market participants.

Textual data for electricity load forecasting

AbstractTraditional mid‐term electricity forecasting models rely on calendar and meteorological information such as temperature and wind speed to achieve high performance. However depending on such variables has drawbacks, as they may not be informative enough during extreme weather. While ubiquitous, textual sources of information are hardly included in prediction algorithms for time series, despite the relevant information they may contain. In this work, we propose to leverage openly accessible weather reports for electricity demand and meteorological time series prediction problems. Our experiments on French and British load data show that the considered textual sources allow to improve overall accuracy of the reference model, particularly during extreme weather events such as storms or abnormal temperatures. Additionally, we apply our approach to the problem of imputation of missing values in meteorological time series, and we show that our text‐based approach beats standard methods. Furthermore, the influence of words on the time series' predictions can be interpreted for the considered encoding schemes of the text, leading to a greater confidence in our results.

Performance evaluation of multivariate deep-time convolution neural architectures for short-term electricity forecasting: Findings and failures

Deep learning (DL) models hold great promise in enhancing the decision-making abilities of electricity market participants and system operators in the short term, as they excel at capturing the intricate interdependencies and uncertainties in electricity consumption. Nevertheless, currently, the adoption of DL models for electricity prediction is not keeping up with the advancements in DL research. Within this framework, the aims of this study are to examine the probabilistic, multivariate, and multihorizon time series prediction of deep time convolution neural (DTCN) models. The effectiveness of DTCN in forecasting electricity consumption over daily, weekly and monthly time horizons is emperically assessed. We analyze both probabilistic and deterministic predictions. The evaluation data comprises real residential electricity consumption, average temperature and humidity with hourly granularity, collected over one year from a residential building. For evaluation, we considered accuracy and uncertainties, sensitivity to hyperparameters, the impact of exogeneous variables, as well as computational efficiency. Results show that the probabilistic models offer greater advantages when it comes to forecasting on multihorizons, especially for diverse datasets with non-uniform time series models. Unlike traditional models, the DTCN models demonstrated superior performance in capturing complex patterns and reducing forecast errors in short-term scenarios. However, their performance can be influenced by the sensitivity of hyperparameters. These findings can act as a benchmark for quantitatively assessing the performance of novel multivariate DL models in predicting electricity demand.

Enhancing smart grid electricity prediction with the fusion of intelligent modeling and XAI integration

This study examines the vital role of accurate load forecasting in the energy planning of smart cities. It introduces a hybrid approach that uses machine learning (ML) to forecast electricity usage in homes, improving accuracy through the extraction of correlated features. The accuracy of predictions is assessed using loss functions and the root mean square error (RMSE). In response to increasing interest in explainable artificial intelligence (XAI), this paper proposes a framework for predicting energy consumption in smart homes. This user-friendly approach helps users understand their energy consumption patterns by employing shapley additive explanations (SHAP) techniques to provide clear explanations. The research uses gradient boosting and long short-term memory neural networks to forecast energy usage. In the context of sustainable urban development, it emphasizes the importance of conserving energy in homes. The paper explores AI and ML methods for predicting residential energy use, aiming to make socially meaningful impacts. It highlights the need to understand the factors affecting predictions to improve the accountability, reliability, and justification of decisions in energy optimization. Explainable AI techniques are used to gain insights into the prediction models and identify factors influencing household energy consumption. This research aids in decision-making processes related to electricity forecasting, advancing discussions on intelligent decision-making in power management, especially in smart grids and sustainable urban development.

Adaptive Probabilistic Forecasting of Electricity (Net-)Load

Adaptive Probabilistic Forecasting of Electricity (Net-)Load

Resilient Electricity Load Forecasting Network with Collective Intelligence Predictor for Smart Cities

Accurate electricity forecasting is essential for smart cities to maintain grid stability by allocating resources in advance, ensuring better integration with renewable energies, and lowering operation costs. However, most forecasting models that use machine learning cannot handle the missing values and possess a single point of failure. With rapid technological advancement, smart cities are becoming lucrative targets for cyberattacks to induce packet loss or take down servers offline via distributed denial-of-service attacks, disrupting the forecasting system and inducing missing values in the electricity load data. This paper proposes a collective intelligence predictor, which uses modular three-level forecasting networks to decentralize and strengthen against missing values. Compared to the existing forecasting models, it achieves a coefficient of determination score of 0.98831 with no missing values using the base model in the Level 0 network. As the missing values in the forecasted zone rise to 90% and a single-model forecasting method is no longer effective, it achieves a score of 0.89345 with a meta-model in the Level 1 network to aggregate the results from the base models in Level 0. Finally, as missing values reach 100%, it achieves a score of 0.81445 by reconstructing the forecast from other zones using the meta-model in the Level 2 network.

Hourly Photovoltaic Production Prediction Using Numerical Weather Data and Neural Networks for Solar Energy Decision Support

The day-ahead photovoltaic electricity forecast is increasingly necessary for grid operators and for energy communities. In the present work, the hourly PV production is estimated using two models based on feedforward neural networks (FFNNs). Most existing models use solar radiation as an input. Instead, the models proposed here use numerical weather prediction (NWP) data: ambient temperature, relative humidity, and wind speed, which are easily accessible to anyone. The first proposed model uses multiple inputs, while the second one uses only the necessary information. A sensitivity analysis allows for the identification of the variables that are most influential on the estimation accuracy. This study concludes that the hourly temperature trend is the most important variable for prediction. The models’ accuracy was tested using experimental and NWP data, with the second model having almost the same accuracy as the first despite using fewer input data. The results obtained using experimental data as inputs show a coefficient of determination (R2) of 0.95 for the hourly PV energy produced. The RMSE is about 6.4% of the panel peak power. When NWP data are used as inputs, R2 is 0.879 and the RMSE is 10.5%. These models can have a significant impact by enabling individual energy communities to make their forecasts, resulting in energy savings and increased self-consumed energy.

Modelling and forecasting India's electricity consumption using artificial neural networks

AbstractPrecise electricity forecasting is a pertinent challenge in effectively controlling the supply and demand of power. This is due to the inherent volatility of electricity, which cannot be stored and must be utilised promptly. Thus, this study develops a framework integrating canonical cointegrating regressions (CCR), time series artificial neural network (ANN) and a multilayer perceptron ANN model for analysing and projecting India's gross electricity consumption to 2030. Annual data for the years 1961–2020 have been collected for variables like gross domestic product (GDP), population, inflation GDP deflator (annual %), annual average temperature and electricity consumption. The study was conducted in three phases. In the first phase of the study, the CCR method was used to check the significance of the selected variables. In the second phase, the projected values of independent variables (GDP, population, inflation GDP deflator [annual %] and annual average temperature) were predicted using the time series ANN model. Finally, a multilayer perceptron ANN model with independent variables was used to forecast the gross electricity consumption in India by 2030. The result shows that the electricity consumption in India will increase by around 50% in the next 10 years, reaching over 1800 TWh in 2030. The proposed approach can be utilised to effectively implement energy policies, as an accurate prediction of energy consumption can help capture future demand.

Community stochastic domestic electricity forecasting

The domestic sector is a significant energy consumer – accounting for around 40% of global electricity demand – due to household demand diversity and complexity. An accurate and robust estimation of domestic electrical loads, environmental impacts, and energy-efficiency potential is crucial for optimal planning and management of energy systems and applications. However, uncertainties resulting from simplistic socio-technical attributes, microclimatic variations, and oversimplification of the effects of interdependent variables make domestic energy modelling challenging. In this research, a hybrid bottom-up community energy forecasting framework is developed to estimate sub-hourly domestic electricity demand using a combination of statistical and engineering modelling approaches by considering key factors influencing household consumption, including demographic characteristics, occupancy patterns, and the features, ownership, and utilisation patterns of electric appliances. The framework is tested on a community in Wales, UK and validated on an annual, daily, and sub-hourly basis with monitored electricity usage averages derived from the UK Energy Follow-Up Survey and the sub-national electricity consumption datasets. Results closely reflect annual and daily household demand at individual dwellings and aggregated levels, with an estimation accuracy of up to 90%. Moreover, the framework facilitates more reliable sub-hourly demand profiles compared to conventional simulation practices that overestimate daily electricity demand and sub-hourly peaks by up to 15% and 50%, respectively.

Short-Term Forecasting of Daily Electricity of Different Campus Building Clusters Based on a Combined Forecasting Model

In the building field, campus buildings are a building group with great energy-saving potential due to a lack of reasonable energy management policies. The accurate prediction of power energy usage is the basis for energy management. To address this issue, this study proposes a novel combined forecasting model based on clustering results, which can achieve a short-time prediction of daily electricity based on a campus building's electricity data over the past 15 days. Considering the diversity of campus buildings in energy consumption and functional aspects, the selected campus buildings are firstly classified into three categories using K-Means clustering in terms of their daily power consumption. Compared with the mainstream building energy consumption prediction models, i.e., LSTM and SVR, the results show that the combined forecast model is superior to other models. Furthermore, an average percentage fluctuation (APF) index is found to be close to the MAPE, which can reflect the prediction accuracy in advance.

Using Deep Neural Network Methods for Forecasting Energy Productivity Based on Comparison of Simulation and DNN Results for Central Poland—Swietokrzyskie Voivodeship

Forecasting electricity demand is of utmost importance for ensuring the stability of the entire energy sector. However, predicting the future electricity demand and its value poses a formidable challenge due to the intricate nature of the processes influenced by renewable energy sources. Within this piece, we have meticulously explored the efficacy of fundamental deep learning models designed for electricity forecasting. Among the deep learning models, we have innovatively crafted recursive neural networks (RNNs) predominantly based on LSTM and combined architectures. The dataset employed was procured from a SolarEdge designer. The dataset encompasses daily records spanning the past year, encompassing an exhaustive collection of parameters extracted from solar farm (based on location in Central Europe (Poland Swietokrzyskie Voivodeship)). The experimental findings unequivocally demonstrated the exceptional superiority of the LSTM models over other counterparts concerning forecasting accuracy. Consequently, we compared multilayer DNN architectures with results provided by the simulator. The measurable results of both DNN models are multi-layer LSTM-only accuracy based on R2—0.885 and EncoderDecoderLSTM R2—0.812.

Bayesian-Optimization-Based Long Short-Term Memory (LSTM) Super Learner Approach for Modeling Long-Term Electricity Consumption

This study utilized different methods, namely classical multiple linear regression (MLR), statistical approach exponential smoothing (EXPS), and deep learning algorithm long short-term memory (LSTM) to forecast long-term electricity consumption in the Kingdom of Saudi Arabia. The originality of this research lies in (1) specifying exogenous variables that significantly affect electrical consumption; (2) utilizing the Bayesian optimization algorithm (BOA) to develop individual super learner BOA-LSTM models for forecasting the residential and total long-term electric energy consumption; (3) measuring forecasting performances of the proposed super learner models with classical and statistical models, viz. MLR and EXPS, by employing the broadly used evaluation measures regarding the computational efficiency, model accuracy, and generalizability; and finally (4) estimating forthcoming yearly electric energy consumption and validation. Population, gross domestic products, imports, and refined oil products significantly impact residential and total annual electricity consumption. The coefficient of determination (R2) for all the proposed models is greater than 0.93, representing an outstanding fitting of the models with historical data. Moreover, the developed BOA-LSTM models have the best performance with R2>0.99, enhancing the predicting accuracy (Mean Absolute Percentage Error (MAPE)) by 59.6% and 54.8% compared to the MLR and EXPS models, respectively, of total annual electricity consumption. This forecasting accuracy in residential electricity consumption for the BOA-LSTM model is improved by 62.7% and 68.9% compared to the MLR and EXPS models. This study achieved a higher accuracy and consistency of the proposed super learner model in long-term electricity forecasting, which can be utilized in energy strategy management to secure the sustainability of electric energy.

Multiple Feature Extraction Long Short-Term Memory Using Skip Connections for Ship Electricity Forecasting

The power load data of electric-powered ships vary with the ships’ operational status and external environmental factors such as sea conditions. Therefore, a model is required to accurately predict a ship’s power load, which depends on changes in the marine environment, weather environment, and the ship’s situation. This study used the power data of an actual ship to predict the power load of the ship. The research on forecasting a ship’s power load fluctuations has been quite limited, and the existing models have inherent limitations in predicting these fluctuations accurately. In this paper, A multiple feature extraction (MFE)-long short-term memory (LSTM) model with skip connections is introduced to address the limitations of existing deep learning models. This novel approach enables the analysis and forecasting of the intricate load variations in ships, thereby facilitating the prediction of complex load fluctuations. The performance of the model was compared with that of a previous convolutional neural network-LSTM network with a squeeze and excitation (SE) model and deep feed-forward (DFF) model. The metrics used for comparison were the mean absolute error, root mean squared error, mean absolute percentage error, and R-squared, wherein the best, average, and worst performances were evaluated for both models. The proposed model exhibited a superior predictive performance for the ship’s power load compared to that of existing models, as evidenced by the performance metrics: mean absolute error (MAE) of 55.52, root mean squared error of (RMSE) 125.62, mean absolute percentage error (MAPE) of 3.56, and R-squared (R2) of 0.86. Therefore, the proposed model is expected to be used for power load prediction during electric-powered ship operations.

An Improved Informer Network for Short-Term Electric Load Forecasting

Background: Electric load forecasting plays an essential role in the dispatching operation of power systems. It can be divided into long-term, medium-term, and short-term according to the forecast time. Accurate short-term electric forecasting helps the system operate safely and reliably, reduces resource waste, and improves economic efficiency. Objective: To fully use the time-series characteristics in load data and improve the accuracy of short-term electric load forecasting, we propose an improved Informer model called Nysformer. Methods: Firstly, the input of data is improved, and the information is input into the model in the form of difference. Then, the Nystrom self-attention mechanism was proposed, approximating the standard self-attention mechanism using an approximation with O(n) time complexity and memory utilization. Results: We conducted experiments on a publicly available dataset, and the results show that the proposed Nysformer model has lower time complexity and higher accuracy than the standard Informer model. Conclusion: An improved informer network is proposed for short-term electric load forecasting, and the experimental results demonstrate the proposed model Nysformer can improve the accuracy of short-term electric load forecasting.

A Reinforcement Learning Approach for Ensemble Machine Learning Models in Peak Electricity Forecasting

Electricity peak load forecasting plays an important role in electricity generation capacity planning to ensure reliable power supplies. To achieve high forecast accuracy, multiple machine learning models have been implemented to forecast the monthly peak load in Thailand over the past few years, yielding promising results. One approach to further improve forecast accuracy is to effectively select the most accurate forecast value for each period from among the forecast values generated by these models. This article presents a novel reinforcement learning approach using the double deep Q-network (Double DQN), which acts as a model selector from a pool of available models. The monthly electricity peak load data of Thailand from 2004 to 2017 are used to demonstrate the effectiveness of the proposed method. A hyperparameter tuning methodology using a fractional factorial design is implemented to significantly reduce the number of required experimental runs. The results indicate that the proposed selection model using Double DQN outperforms all tested individual machine learning models in terms of mean square error.

Annual Contribution Electricity Forecasting Model Based on Logistic Regression Analysis

Due to the complex characteristics of the annual contribution time series, it is difficult to achieve the ideal prediction effect by a single prediction. Therefore, the annual contribution electricity prediction model based on Logistic regression analysis is studied. The statistical method of time series analysis combined with the fuzzy correlation feature analysis method is used to obtain the high-voltage power transmission data of business expansion and the power consumption data after power transmission. We use the fuzzy clustering theory to complete the customer segmentation and accurately locate the same type of user groups. On this basis, we preprocess the characteristic data of the annual contribution electricity forecast and build the annual contribution electricity forecast model based on Logistic regression analysis to realize the annual contribution electricity forecast. The experimental results show that the proposed method has a good prediction effect of annual contribution power, and can effectively shorten the prediction time of annual contribution power.

Application of Sarima Model in Load Forecasting in Hanoi City

Many countries and researchers are interested in load forecasting due to its significance. The results of load forecasting are an indispensable basis for electricity planning and investment plan for the power system in the future. In addition, short-term load forecasting is also a mandatory requirement in proactively developing business and system operation plan of units in power sector. When the SARIMA(p,d,q)(P,D,Q)S model is used for load forecasting in Hanoi, the results ensure forecast accuracy, feasibility of available data as well as the software's ease of use. SARIMA model can also be used in load forecasting in other provinces and cities in Vietnam or used for electricity forecasting. With the research sample from 2019 to 2021, the model SARIMA(0,1,6)(0.1,1)24 and (0,1,7)(0.1,1) 24 are used for short-term load forecasting in term of typical working day and day-off in Hanoi.

Short-term power forecasting model based on GWO-LSTM network

In view of the time-series characteristics of the grid load data, this paper proposes a method to predict electricity demand by optimizing a long-and short-term memory (LSTM) neural network model using the grey wolf optimization algorithm, taking into account the effects of time, weather conditions and holiday conditions on electricity loads. The model overcomes the disadvantage that the backpropagation through the time algorithm tends to converge to a local optimum. The experimental results show that the prediction results outperform those of the traditional LSTM for short-term electricity loads, providing a reference direction for future electricity forecasting models.