Electricity Price Forecasting Research Articles

A historical weather forecast dataset from the European Centre for Medium-Range Weather Forecasts (ECMWF) for energy forecasting

Weather is often found to be a key driving factor for power generation and energy consumption. To capture the effects of weather, many energy forecasting practices, such as load forecasting, renewable power generation forecasting, gas and electricity price forecasting, and power distribution systems outage forecasting, would rely on numerical weather prediction (NWP). In the academic literature, however, energy forecasting models have often been developed based on settings of ex post forecasting, where the actual observations of weather variables during the forecasted period are being used. Such gap between academic research and field practices is partly due to the shortage of historical weather forecasts. To that end, an NWP forecast dataset from the European Centre for Medium-Range Weather Forecasts (ECMWF) High Resolution (HRES) model, as available in the ECMWF’s Archive Catalogue, is offered to the energy forecasting community under the Creative Commons Attribution 4.0 International (CC BY 4.0) license. Since the raw data is massive in size, a subset which is thought sufficient for energy forecasting research purposes is provided through this article. Four years (2017–2020) of HRES forecasts of 14 frequently used weather variables, over the geographical region bounded by 63° N, −126°W, 21° S, and 36° E (most of Europe and North America), on a 0.5°by 0.5°longitude/latitude grid, are released in the form of NetCDF files. This dataset is able to support a variety of aforementioned energy forecasting tasks. In addition to introducing various means to utilize the dataset, this article provides a set of case studies on post-processing of day-ahead solar forecasts. The R code being used to produce the results shown in this article is also made available, so that the readers can reproduce this case study as well as adopt the code for other relevant studies.

Optimization of Interval Type-2 Fuzzy Logic System Using Grasshopper Optimization Algorithm

The estimation of the fuzzy membership function parameters for interval type 2 fuzzy logic system (IT2-FLS) is a challenging task in the presence of uncertainty and imprecision. Grasshopper optimization algorithm (GOA) is a fresh population based meta-heuristic algorithm that mimics the swarming behavior of grasshoppers in nature, which has good convergence ability towards optima. The main objective of this paper is to apply GOA to estimate the optimal parameters of the Gaussian membership function in an IT2-FLS. The antecedent part parameters (Gaussian membership function parameters) are encoded as a population of artificial swarm of grasshoppers and optimized using its algorithm. Tuning of the consequent part parameters are accomplished using extreme learning machine. The optimized IT2-FLS (GOAIT2FELM) obtained the optimal premise parameters based on tuned consequent part parameters and is then applied on the Australian national electricity market data for the forecasting of electricity loads and prices. The forecasting performance of the proposed model is compared with other population-based optimized IT2-FLS including genetic algorithm and artificial bee colony optimization algorithm. Analysis of the performance, on the same data-sets, reveals that the proposed GOAIT2FELM could be a better approach for improving the accuracy of the IT2-FLS as compared to other variants of the optimized IT2-FLS.

A Survey of Preprocessing Methods Used for Analysis of Big Data Originated From Smart Grids

In this paper, a brief survey of data preprocessing methods is presented. Specifically, the data preprocessing methods used in the smart grid (SG) domain are surveyed. Also, with the advent of SG, data collection on a large scale became possible. The data is essential for electricity demand, generation and price forecasting, which plays an important role in making energy efficient decisions, and long and short term predictions regarding energy generation, consumption and storage. However, the forecasting accuracy decreases when data is used in raw form. Hence, data preprocessing is considered essential. This paper provides an overview of the data preprocessing methods and a detailed discussion of the methods used in the existing literature. A comparison of the methods is also given. A survey of closely related survey papers is also presented and the papers are compared based on their contributions. Moreover, based on the discussion of the data preprocessing methods, a narrative is built with a critical analysis. Finally, future research directions are discussed to guide the readers.

Modeling and Forecasting Electricity Demand and Prices: A Comparison of Alternative Approaches

Electricity demand and price forecasting are key components for the market participants and system operators as precise forecasts are necessary to manage power systems effectively. However, forecasting electricity demand and prices are challenging due to their specific features, such as high frequency, volatility, long trend, nonconstant mean and variance, mean reversion, multiple seasonalities, calendar effects, and spikes/jumps. Thus, the main aim of this study is to propose models that can efficiently forecast electricity demand and prices. To this end, the time series (demand/price) is divided into two components. The first component is considered a deterministic component that includes a trend, yearly, seasonal, and weekly periodicities, calendar effects, and lagged exogenous information and is modeled by parametric and nonparametric approaches. The second component is known as a stochastic (residual) component that is estimated using univariate autoregressive (AR) and multivariate vector autoregressive (VAR) models. The estimation of these models is carried out by four different estimation methods, including ordinary least squares (O), Lasso (L), Ridge (R), and Elastic‐net (E). The proposed modeling scheme is applied to Nordic electricity demand and price time series, and one‐day‐ahead out‐of‐sample forecasts are obtained for a whole year. Besides descriptive statistics, a statistical significance test is also used to evaluate the models’ forecasting accuracy. The results suggest that the proposed methodology effectively forecasts the price and demand for electricity. In addition, the choice of the estimation procedure used for both deterministic and stochastic components has a significant effect on the forecasting results. Furthermore, multivariate vector autoregressive gives superior performance compared to univariate autoregressive models.

Week Ahead Electricity Power and Price Forecasting Using Improved DenseNet-121 Method

In the Smart Grid (SG) residential environment, consumers change their power consumption routine according to the price and incentives announced by the utility, which causes the prices to deviate from the initial pattern. Thereby, electricity demand and price forecasting play a significant role and can help in terms of reliability and sustainability. Due to the massive amount of data, big data analytics for forecasting becomes a hot topic in the SG domain. In this paper, the changing and non-linearity of consumer consumption pattern complex data is taken as input. To minimize the computational cost and complexity of the data, the average of the feature engineering approaches includes: Recursive Feature Eliminator (RFE), Extreme Gradient Boosting (XGboost), Random Forest (RF), and are upgraded to extract the most relevant and significant features. To this end, we have proposed the DensetNet-121 network and Support Vector Machine (SVM) ensemble with Aquila Optimizer (AO) to ensure adaptability and handle the complexity of data in the classification. Further, the AO method helps to tune the parameters of DensNet (121 layers) and SVM, which achieves less training loss, computational time, minimized overfitting problems and more training/test accuracy. Performance evaluation metrics and statistical analysis validate the proposed model results are better than the benchmark schemes. Our proposed method has achieved a minimal value of the Mean Average Percentage Error (MAPE) rate i.e., 8% by DenseNet-AO and 6% by SVM-AO and the maximum accurateness rate of 92% and 95%, respectively.

Day-Ahead Probabilistic Electricity Price Forecast on the German Epex

Day-Ahead Probabilistic Electricity Price Forecast on the German Epex

Studies on Day-Ahead Electricity Price Forecasting: Sarimax, Lstm and Cnn-Lstm with the Electrical Consumption as an Exogenous Input

Studies on Day-Ahead Electricity Price Forecasting: Sarimax, Lstm and Cnn-Lstm with the Electrical Consumption as an Exogenous Input

Automated Electricity Price Forecast Using Combined Models

Automated Electricity Price Forecast Using Combined Models

Day-Ahead Electricity Price Forecasting Employing a Novel Hybrid Frame of Deep Learning Methods: A Case Study in Nsw, Australia

Day-Ahead Electricity Price Forecasting Employing a Novel Hybrid Frame of Deep Learning Methods: A Case Study in Nsw, Australia

Comparative Study of the Identification Methods of the Management System of the Day-Ahead Market of Polish Energy Market S.A.

Nowadays, identification and neural methods are used more and more often in modeling IT forecasting systems in addition to analytical methods. Six characteristic models used to forecast the Day-Ahead Market system functioning as a transaction management system at the Polish Power Exchange (POLPX) and the Nord Pool Spot market have been selected for comparative analysis. The research was preceded by a detailed discussion of modern criteria used to assess the quality of model fitting to the system, namely: effectiveness, efficiency, and robustness. In the literature, there are two main groups of system modeling methods, namely time series modeling methods and identification modeling methods, including neural modeling methods. Modeling usually results in such models as parametric models and artificial neural networks learned neural models of the Day-Ahead Market, as well as time series models, among others. In the comparative analysis, special attention was paid to the accuracy of the obtained models concerning the system. It has been pointed out that the studied solutions used to measure the accuracy of modeling criteria such as accuracy of fit or efficiency, and did not use the modeling efficiency, which is very important in IT forecasting systems for such large markets as the Day-Ahead Market of POLPX. The search for the best market models, including identification models of the Day-Ahead Market operation that can be used in electricity price forecasting is a very important issue both from the point of view of algorithmic solutions and economical solutions.

Forecasting day-ahead electricity prices: A comparison of time series and neural network models taking external regressors into account

The amount of renewable energies in electricity production has increased significantly in the last decade, resulting in more variability of the day-ahead electricity spot price. The Electricity Price Forecast (EPF) has to adapt to the new situation by applying flexible models. However, the numerous available forecasting methods differ widely, with no distinct candidate offering the best solution. Against this background, we conduct a comparative study of four different approaches to forecasting the German day-ahead electricity spot price. In addition to the prominent Seasonal Integrated Auto-Regressive Moving Average model ((S)ARIMA(X)) and the Long-Short Term Memory (LSTM) neural network models, we employ a Convolutional Neural Network LSTM (CNN-LSTM) and an extended two-stage multivariate Vector Auto-Regressive model (VAR) approach as hybrid models. For better performance, we include common external influences such as the consumer load, fuel and CO2 emission prices, average solar radiation and wind speed in our analysis. We analyse hourly data for twelve samples from October 2017 to September 2018. Each model is implemented to deliver price forecasts at three horizons, i.e., one day, seven days and thirty days ahead. While the LSTM model achieves the best forecasting performance on average, the two-stage VAR follows closely behind and performs exceedingly well for shorter prediction horizons. Further, we provide evidence that a combination of both forecasting methods outperforms each of the single models. This indicates that combining advanced methods could lead to further improvements in electricity spot price forecasts.

EPFG: Electricity Price Forecasting with Enhanced GANS Neural Network

Power load forecasting in Data Analytics is an emerging technology. In this paper, we have proposed the Generative Adversarial Networks (GANS) neural network model as the classifier for probabilistic electricity price forecasting. To assess the performance of these frameworks, we apply our models on the dataset cater by (IESO) in Ontario, Canada. We have compared our proposed model with Random Forest, Support vector machine (SVM), and XG-Boost. MSE, RMSE, MAE metrices are considered for the evaluation of the model’s performance. The outcome indicates that the mean squared error (MSE) of our proposed model is 687.513 whereas the MSE of existing methodologies is 830.15, 746.812, and 776.201 which is more than our proposed methodology. Mean absolute error (MAE) of SVM and our proposed GANS Neural Network (EPFEG) have the lowest MAE that is 8%. Furthermore, EPFEG achieved almost 7% better accuracy than existing schemes.

A Novel Hybrid Feature Selection Method for Day-Ahead Electricity Price Forecasting

The paper proposes a novel hybrid feature selection (FS) method for day-ahead electricity price forecasting. The work presents a novel hybrid FS algorithm for obtaining optimal feature set to gain optimal forecast accuracy. The performance of the proposed forecaster is compared with forecasters based on classification tree and regression tree. A hybrid FS method based on the elitist genetic algorithm (GA) and a tree-based method is applied for FS. Making use of selected features, aperformance test of the forecaster was carried out to establish the usefulness of the proposed approach. By way of analyzing and forecasts for day-ahead electricity prices in the Australian electricity markets, the proposed approach is evaluated and it has been established that, with the selected feature, the proposed forecaster consistently outperforms the forecaster with a larger feature set. The proposed method is simulated in MATLAB and WEKA software.

An operational bidding framework for aggregated electric vehicles on the electricity spot market

Fluctuating electricity prices offer potential economic savings for the consumption of electricity by flexible assets such as Electric Vehicles (EVs). This study proposes an operational bidding framework that minimizes the charging costs of an EV fleet by submitting an optimized bid to the day-ahead electricity market. The framework consists of a bidding module that determines the most cost-effective bid by considering an electricity price and an EV charging demand forecast module. In this study we develop and evaluate several regression and machine learning models that forecast the electricity price and EV charging demand. Furthermore, we examine the composition of a most optimal operational bidding framework by comparing the outcome of the bidding module when fed with each of the forecast models. This is determined by considering the day-ahead electricity price and imbalance costs due to forecast errors. The study demonstrates that the best performing self-contained forecast models with the objective of electricity price and EV charging demand forecasting, do not deliver the best overall results when included in the bidding framework. Additionally, the results show that the best performing framework obtains a 26% cost savings compared to a reference case where EVs are charged inflexibly. This corresponds to an achieved savings potential of 92%. Consequently, along with the developed bidding framework, these results provide a fundamental basis for effective electricity trading on the day-ahead market.

Scenarios modelling for forecasting day-ahead electricity prices: Case studies in Australia

Electricity prices in spot markets are volatile and can be affected by various factors, such as generation and demand, system contingencies, local weather patterns, bidding strategies of market participants, and uncertain renewable energy outputs. Because of these factors, electricity price forecasting is challenging. This paper proposes a scenario modeling approach to improve forecasting accuracy, conditioning time series generative adversarial networks on external factors. After data pre-processing and condition selection, a conditional TSGAN or CTSGAN is designed to forecast electricity prices. Wasserstein Distance, weights limitation, and RMSProp optimizer are used to ensure that the CTGAN training process is stable. By changing the dimensionality of random noise input, the point forecasting model can be transformed into a probabilistic forecasting model. For electricity price point forecasting, the proposed CTSGAN model has better accuracy and has better generalization ability than the TSGAN and other deep learning methods. For probabilistic forecasting, the proposed CTSGAN model can significantly improve the continuously ranked probability score and Winkler score. The effectiveness and superiority of the proposed CTSGAN forecasting model are verified by case studies.

Performance of a Hybrid Neural-Based Framework for Alternative Electricity Price Forecasting in the Smart Grid

Electricity forecasting is an essential task for energy management systemsof microgrids deployed in smart grids. Accurate price forecasting will eventuallyenhance the economic operation of microgrids. In this regard, theliterature is rich with studies focused on predicting electricity price datausing artificial neural networks. However, most of them consider a singlemodel such as multi-layer perceptron (MLP) and radial basis function (RBF)to perform electricity price forecasting. In this paper, a hybrid frameworkbased on simultaneously utilizing MLP-RBF neural networks is presented topredict the Iranian electricity market price. In addition, few works in literatureconsidered Iran’s electricity market as their case of analysis and investigation.Forecasting results indicate that MLP neural networks outperform theRBF neural networks. The values for the coefficient of determination (R)
 corresponding to MLP and RBF neural networks are obtained 0.55 and 0.44,respectively. However, the proposed hybrid framework performed better thanboth MLP and RBF models with R-value equal to 0.71. In addition to this,the MSE and RMSE values show the superiority of the proposed method tothe single methods.

A Novel Machine Learning-Based Price Forecasting for Energy Management Systems

Price forecasting (PF) is the primary concern in distributed power generation. This paper presents a novel and improved technique to forecast electricity prices. The data of various power producers, Capacity Purchase Price (CPP), Power Purchase Price (PPP), Tariff rates, and load demand from National Electric Power Regulatory Authority (NEPRA) are considered for MAPE reduction in PF. Eight time-series and auto-regression algorithms are developed for data fetching and setting the objective function. The feed-forward ANFIS based on the ML approach and space vector regression (SVR) is introduced to PF by taking the input from time series and auto-regression (AR) algorithms. Best-feature selection is conducted by adopting the Binary Genetic Algorithm (BGA)-Principal Component Analysis (PCA) approach that ultimately minimizes the complexity and computational time of the model. The proposed integration strategy computes the mean absolute percentage error (MAPE), and the overall improvement percentage is 9.24%, which is valuable in price forecasting of the energy management system (EMS). In the end, EMS based on the Firefly algorithm (FA) has been presented, and by implementing FA, the cost of electricity has been reduced by 21%, 19%, and 20% for building 1, 2, and 3, respectively.

Mid-term energy cost-oriented flow shop scheduling: Integration of electricity price forecasts, modeling, and solution procedures

The paper deals with a flow shop scheduling problem, where energy costs are to be minimized with respect to time-varying electricity prices coming from the day-ahead market EPEX Spot Germany/Austria. To make efficient use of the corresponding price volatility in the production planning phase, i.e., weeks prior to the start of production, price forecasts have to be incorporated into decision making. Since the suitability of mid-term electricity price forecasts for scheduling applications has not been clarified, we analyze forecasts from the literature for their usability by means of extensive scenarios as well as an appropriate evaluation scheme. Furthermore, solution methods are provided for the considered flow shop scheduling problem. We study a Fix-and-Optimize and a priority rule-based sampling procedure in order to solve large-scale instances within a reasonable time. The performance analysis shows that Fix-and-Optimize achieves very good results in terms of solution quality. If production planners have to make quick decisions, the sampling procedure provides excellent run times.

Experimental Analysis of GBM to Expand the Time Horizon of Irish Electricity Price Forecasts

In response to the inherent challenges of generating cost-effective electricity consumption schedules for dynamic systems, this paper espouses the use of GBM or Gradient Boosting Machine-based models for electricity price forecasting. These models are applied to data streams from the Irish electricity market and achieve favorable results, relative to the current state-of-the-art. Presently, electricity prices are published 10 h in advance of the trade day of interest. Using the forecasting methodology outlined in this paper, an estimation of these prices can be made available one day in advance of the official price publication, thus extending the time available to plan electricity utilization from the grid to be as cost effectively as possible. Extreme Gradient Boosting Machine (XGBM) models achieved a Mean Absolute Error (MAE) of 9.93 for data from 30 September 2018 to 12 December 2019 which is an 11.4% improvement on the avant-garde. LGBM models achieve a MAE score 9.58 on more recent data: the full year of 2020.

Integrated day-ahead and intraday self-schedule bidding for energy storage systems using approximate dynamic programming

Most modern energy markets trade electricity in advance for technical reasons. Thus, market participants must commit to delivering or consuming a certain amount of energy before the actual delivery. In Germany, two markets with daily auctions coexist. In the day-ahead auction market, the energy is traded in 60-minute time slots, which are further partitioned into 15-minute time slots for the intraday auction market. Because of the slow ramp-ups of nuclear and fossil power plants, these price-makers trade mostly in the day-ahead market. Only the residual energy is traded in the intraday market, where the market prices fluctuate substantially more. These fluctuations as well as the expected price difference between these markets can be exploited by fast ramping energy storage systems. We address the decision problem of an owner of an energy storage who trades on both markets, taking ramping times into account. Because the state variable of our dynamic programming formulation includes all features of our high-dimensional electricity price forecast, this problem cannot be solved to optimality. Instead, we use approximate dynamic programming. In a numerical study based on real-world data, we benchmark the algorithm against an adapted state-of-the-art approach from literature and an expectation model with a receding horizon. Furthermore, we investigate the influence of the price forecast on expected profit and demonstrate that it is essential for the dynamic program to capture the high dimensionality of the price forecast to compete with the expectation model, which does not suffer from the curses of dimensionality.