Day-ahead Electricity Market Prices Research Articles

Predicting Day-Ahead Electricity Market Prices through the Integration of Macroeconomic Factors and Machine Learning Techniques

Several events in the last years changed to some extent the common understanding of the electricity day-ahead market (DAM). The shape of the electricity price curve has been altered as some factors that underpinned the electricity price forecast (EPF) lost their importance and new influential factors emerged. In this paper, we aim to showcase the changes in EPF, understand the effects of uncertainties and propose a forecasting method using machine learning (ML) algorithms to cope with random events such as COVID-19 pandemic and the conflict in Black Sea region. By adjusting the training period according to the standard deviation that reflects the price volatility, feature engineering and by using two regressors for weighing the results, significant improvements in the performance of the EPF are achieved. One of the contributions of the proposed method consists in adjusting the training period considering the price variation. Thus, we introduce a rule-based approach given an empirical observation that for days with a higher growth in prices the training interval should be shortened, capturing the sharp variations of prices. The results of several cutting-edge ML algorithms represent the input for a predictive meta-model to obtain the best forecasting solution. The input dataset spans from Jan. 2019 to Aug. 2022, testing the proposed EPF method for both stable and more tumultuous intervals and proving its robustness. This analysis provides decision makers with an understanding of the price trends and suggests measures to combat spikes. Numerical findings indicate that on average mean absolute error (MAE) improved by 48% and root mean squared error (RMSE) improved by 44% compared to the baseline model (without feature engineering/adjusting training). When the output of the ML algorithms is weighted using the proposed meta-model, MAE further improved by 2.3% in 2020 and 5.14% in 2022. Less errors are recorded in stable years like 2019 and 2020 (MAE = 6.71, RMSE = 14.67) compared to 2021 and 2022 (MAE = 9.45, RMSE = 20.64).

Game-Theoretic Capacity Expansion Analysis Under Carbon Emission Constraints

In this paper, we develop a long-term capacity expansion analysis model that considers strategic bidding behavior by dominant generation companies together with new entry decisions, including renewables, to meet a long-term carbon reduction constraint. We deploy a centralized quadratic programming formulation of the Nash-Cournot game to analyze day-ahead electricity market prices. In our model, strategic producers simultaneously make investment and production decisions, deploying an open-loop Cournot game. We have implemented the model for Ukraine's day-ahead electricity market, which has been highly prone to market power issues. The case study results show significant inefficiencies in peak prices and explore scenarios of alternative mitigation options and their efficacies. Finally, the case study is extended to also explore implications of a low carbon future through emission-constrained runs, evaluation of the performance-based emission policy, and sensitivity on different portfolios of new renewable investors.

Day-Ahead Electricity Market Price Forecasting Considering the Components of the Electricity Market Price; Using Demand Decomposition, Fuel Cost, and the Kernel Density Estimation

This paper aims to improve the forecasting of electricity market prices by incorporating the characteristics of electricity market prices that are discretely affected by the fuel cost per unit, the unit generation cost of the large-scale generators, and the demand. In this paper, two new techniques are introduced. The first technique applies feature generation to the label and forecasts the transformed new variables, which are then post-processed by inverse transformation, considering the characteristic of the fuel types of marginal generators or prices through two variables: fuel cost per unit by the representative fuel type and argument of the maximum of Probability Density Function (PDF) calculated by Kernel Density Estimation (KDE) from the previous price. The second technique applies decomposition to the demand, followed by a feature selection process to apply the major decomposed feature. It is verified using gain or SHapley Additive exPlanations (SHAP) value in the feature selection process. In the case study, both showed improvement in all indicators. In the Korean Electricity Market, the unit generation cost for each generator is calculated monthly, resulting in a step-wise change in the electricity market price depending on the monthly fuel cost. Feature generation using the fuel cost per unit improved the forecasting by eliminating monthly volatility caused by the fuel costs and reducing the error that occurs at the beginning of the month. It improved the Mean Squared Percentage Error (MAPE) of 3.83[%]. Using the argument of the maximum PDF calculated by KDE improved the forecasting during the test period, where discrete monthly variations were not included. The resulting MAPE was 3.82[%]. Combining these two techniques resulted in the most accurate performance compared to the other techniques used, which had a MAPE of 3.49[%]. The MAPE of the forecasting with the decomposed data of the original price was 4.41[%].

Multi-Criteria Optimization of Vehicle-to-Grid Service to Minimize Battery Degradation and Electricity Costs

Increased use of renewable energy sources in energy sector as well as improvements and electrification in transportation sector significantly contribute to reduction of green-house gasses emissions and mitigation of problems with fossil fuel dependency. Optimal integration of electric vehicles (EVs) into the grids and their charging/discharging schedules have to be realized in accordance with electricity demand, day-ahead electricity market prices and intermittency of photovoltaic and wind generators electricity production. A microgrid that includes non-deferrable loads, renewable energy sources, EV fleet and its charging station is analyzed in this paper. Its Vehicle-to-Grid (V2G) service is optimized with the aim of minimizing the operational costs and obtaining peak load shaving and valley filling of the load curve, which is especially effective in the case of EVs fleet with occupational time intervals known in advance. Optimized schedule of EVs charging and discharging is obtained as a result of the procedure that uses multi-criteria optimization function. These criteria include minimization of microgrid electricity costs as the local aggregator’s benefit, maximization of the flattening of total microgrid demand curve as main grid operator’s benefit, and minimization of battery degradation (due to a number of charging/discharging cycles) as EVs owner’s benefit which is the novelty of this paper. Experimental analysis is performed on several scenarios and program Lingo is used to solve the optimization problem.

ARIMA Model for Day-Ahead Electricity Market Price Forecasting

Electricity price forecasting becomes a significant challenge on a day-to-day basis and price variations are even more volatile on an hourly basis. Therefore, this paper is used several approaches to analyze the Bulgarian hourly electricity price dynamics in the day-ahead market. Proper analysis crucially depends on the choice of an adequate model. Reviewed are the factors which may influence the electricity spot prices and characteristics of the time series of prices. Methods include and variety of modeling approaches that are applied and evaluated for forecasting electricity prices such as time-series models and regression models. The forecasting technique is to model day-ahead spot prices using well known ARIMA/SARIMA model including stationarity checks, seasonal decompose, differencing, autoregressive modeling, and autocorrelation to analyze and forecast time series hourly data. For each approach, model estimates and forecasts are developed using hourly price data, reshaped, and aggregated data on a daily and monthly basis for the Bulgarian day-ahead market.

Impact of the COVID-19 Pandemic Crisis on the Efficiency of European Intraday Electricity Markets

Our goal is to examine the efficiency of different intraday electricity markets and if any of their price prediction models are more accurate than others. This paper includes a comprehensive review of Germany, France, and Norway’s (NOR1) day-ahead and intraday electricity market prices. These markets represent different energy mixes which would allow us to analyze the impact of the energy mix on the efficiencies of these markets. To draw conclusions about extreme market conditions, (i) we reviewed the market data linked to COVID-19. We expected higher volatility in the lockdowns than before and therefore decrease in the efficiency of the prediction models. With our analysis, (ii) we want to draw conclusions as to whether a mix based mainly on renewable energies such as that in Norway implies lower volatilities even in times of crisis. This would answer (iii) whether a market with an energy mix like Norway is more efficient in highly volatile phases. For the analysis, we use data visualization and statistical models as well as sample and out-of-sample data. Our finding was that while the different price and volatility levels occurred, the direction of the market was similar. We could find evidence that our expectations (i–iii) were met.

Data augmentation for time series regression: Applying transformations, autoencoders and adversarial networks to electricity price forecasting

A model’s expected generalisation error is inversely proportional to its training set size. This relationship can pose a problem when modelling multivariate time series, because structural breaks, low sampling rates, and high data gathering costs can severely restrict training set sizes, increasing a model’s expected generalisation error by spurring regression model overfitting. Artificially expanding the training set size, using data augmentation methods, can, however, counteract the restrictions imposed by small sample sizes: increasing a model’s robustness to overfitting and boosting out-of-sample prediction accuracies. While existing time series augmentation methods have predominantly utilised feature space transformations to artificially expand training set sizes and boost prediction accuracies, we propose using autoencoders (AEs), variational autoencoders (VAEs) and Wasserstein generative adversarial networks with a gradient penalty (WGAN-GPs) for time series augmentation. To evaluate our proposed augmentors, as a case study we forecast Belgian and Dutch day-ahead electricity market prices using both autoregressive models and artificial neural networks. Overall, our results demonstrate that AEs, VAEs, and WGAN-GPs can significantly boost regression accuracies; on average decreasing benchmark model mean absolute errors by 2.23%, 2.73% and 2.97% respectively. Moreover, our results demonstrate that combining AE, VAE, and WGAN-GP generated time series can further boost regression accuracies; on average decreasing benchmark errors by 3.44%. As our proposed augmentors outperform existing augmentation methods, we strongly believe that both practitioners and researchers aiming to generate time series or reduce time series regression errors will find utility in our study.

Fundamental pricing laws and long memory effects in the day-ahead power market

We apply linear and nonlinear panel models to investigate fundamental pricing laws in the Greek day-ahead electricity market. We link the persistence behaviour and other dynamic properties of electricity price trajectories to an extended array of fundamental variables reflecting market conditions and the price course of energy commodities in global markets. Assuming that the dynamics of the intraday power pricing curve is adequately specified within a panel modelling framework with long memory and common unobserved stochastic effects, we are able to provide consistent estimates of price responsiveness to the main fundamental drivers and thus avoid biases induced by other popular modelling approaches. We also device a panel variant of the standard smooth-transition regression model to investigate regime switching effects in the price responsiveness mechanism. Empirical results suggest that Greek day-ahead electricity market prices respond significantly to hour-specific (load, marginal generation capacity) and global (Brent price, emission allowance rights cost) fundamental variables. In some intraday power delivery zones, these covariates are in a fractional cointegrating relationship with electricity prices being responsible for their drifting behaviour. The strength and type of cointegration changes across intraday trading periods in response to market conditions. The results of the nonlinear model demonstrate the ineffectiveness of linear model designs (either univariate or panel) to accurately estimate the levels of price risk and associate it with changes in the market climate.

Incorporating Production Task Scheduling in Energy Management of an Industrial Microgrid: A Regret-Based Stochastic Programming Approach

In this paper, a regret-based risk-averse stochastic production task and energy management (PTEM) model for industrial microgrid is proposed, which is applied to a battery manufacturing plant (BMP) microgrid. We incorporate production task scheduling with production constraints in the energy management of industrial microgrid. The general PTEM of industrial microgrids is analyzed and the production processes and energy requirements of a BMP multi-energy microgrid prosumer are presented, in which thermostatically controlled loads (TCLs) in buildings are considered. Then, the regret-based risk-averse PTEM model is formulated as a mixed-integer linear programming (MILP) problem based on stochastic programming considering uncertainties of day-ahead electricity market price and photovoltaic (PV) power output. Conditional value-at-risk (CVaR) is introduced for risk control and utilized for minimizing the expected regret. Numerical studies are carried out to validate the proposed model by comparing with risk-neutral approach and deterministic model based on real market prices from PJM electricity market. The results show that the proposed PTEM approach is effectively risk-averse. In addition, the impacts of total target production task and risk-control parameter on expected cost and regret are analyzed.

Quantifying the effect of renewable generation on day–ahead electricity market prices: The Spanish case

Abstract The penetration of renewable generation has grown since the electricity sector has been deregulated. To account for that, this paper proposes a methodology to estimate the downward effect of renewable generation participation upon the day-ahead electricity market prices, since such an effect is quite intuitive observing the merit order of the generating units. The European Electricity Market Matching Algorithm (EMMA) is currently based on Euphemia (Price Coupling of Regions), though there are several differences among countries across Europe. The new algorithm proposed uses market orders, which include aggregate hourly orders such as aggregate supply and demand curves. These orders are simple orders and the marginal price is affected by complex orders, especially by the minimum income condition (MIC) used in the Iberian Electricity Market and considered in our proposed algorithm. A case study of the Spanish day-ahead electricity market is evaluated for 2015, for which a daily generation sample is composed of 16 days in 2015. The sample is created following the characteristics of thermal production, renewable production and inframarginal production. The conclusions are drawn comparing the simulations of the real marginal prices and the new marginal prices after incorporating renewable generation participation into the aggregate demand curve at the maximum price.

Introducing Technical Indicators to Electricity Price Forecasting: A Feature Engineering Study for Linear, Ensemble, and Deep Machine Learning Models

Day-ahead electricity market (DAM) volatility and price forecast errors have grown in recent years. Changing market conditions, epitomised by increasing renewable energy production and rising intraday market trading, have spurred this growth. If forecast accuracies of DAM prices are to improve, new features capable of capturing the effects of technical or fundamental price drivers must be identified. In this paper, we focus on identifying/engineering technical features capable of capturing the behavioural biases of DAM traders. Technical indicators (TIs), such as Bollinger Bands, Momentum indicators, or exponential moving averages, are widely used across financial markets to identify behavioural biases. To date, TIs have never been applied to the forecasting of DAM prices. We demonstrate how the simple inclusion of TI features in DAM forecasting can significantly boost the regression accuracies of machine learning models; reducing the root mean squared errors of linear, ensemble, and deep model forecasts by up to 4.50%, 5.42%, and 4.09%, respectively. Moreover, tailored TIs are identified for each of these models, highlighting the added explanatory power offered by technical features.

Optimizing Day-Ahead Electricity Market Prices: Increasing the Total Surplus for Energy Exchange Istanbul

Problem definition: We design a combinatorial auction to clear the Turkish day-ahead electricity market, and we develop effective tabu search and genetic algorithms to solve the problem of matching bidders and maximizing social welfare within a reasonable amount of time for practical purposes. Academic/practical relevance: A double-sided blind combinatorial auction is used to determine electricity prices for day-ahead markets in Europe. Considering the integer requirements associated with market participants’ bids and the nonlinear social welfare objective, a complicated problem arises. In Turkey, the total number of bids reaches 15,000, and this large problem needs to be solved within minutes every day. Given the practical time limit, solving this problem with standard optimization packages is not guaranteed, and therefore, heuristic algorithms are needed to quickly obtain a high-quality solution. Methodology: We use nonlinear mixed-integer programming and tabu search and genetic algorithms. We analyze the performance of our algorithms by comparing them with solutions commercially available to the market operator. Results: We provide structural results to reduce the problem size and then develop customized heuristics by exploiting the problem structure in the day-ahead market. Our algorithms are guaranteed to generate a feasible solution, and Energy Exchange Istanbul has been using them since June 2016, increasing its surplus by 448,418 Turkish liras (US$128,119) per day and 163,672,570 Turkish liras (US$46,763,591) per year, on average. We also establish that genetic algorithms work better than tabu search for the Turkish day-ahead market. Managerial implications: We deliver a practical tool using innovative optimization techniques to clear the Turkish day-ahead electricity market. We also modify our model to handle similar European day-ahead markets and show that performances of our heuristics are robust under different auction designs.

Uncertainty-Based Models for Optimal Management of Energy Hubs Considering Demand Response

Energy hub (EH) is a concept that is commonly used to describe multi-carrier energy systems. New advances in the area of energy conversion and storage have resulted in the development of EHs. The efficiency and capability of power systems can be improved by using EHs. This paper proposes an Information Gap Decision Theory (IGDT)-based model for EH management, taking into account the demand response (DR). The proposed model is applied to a semi-realistic case study with large consumers within a day ahead of the scheduling time horizon. The EH has some inputs including real-time (RT) and day-ahead (DA) electricity market prices, wind turbine generation, and natural gas network data. It also has electricity and heat demands as part of the output. The management of the EH is investigated considering the uncertainty in RT electricity market prices and wind turbine generation. The decisions are robust against uncertainties using the IGDT method. DR is added to the decision-making process in order to increase the flexibility of the decisions made. The numerical results demonstrate that considering DR in the IGDT-based EH management system changes the decision-making process. The results of the IGDT and stochastic programming model have been shown for more comprehension.

Hybrid model using three-stage algorithm for simultaneous load and price forecasting

Short-term load and price forecasting is an important issue in the optimal operation of restructured electric utilities. This paper presents a new intelligent hybrid three-stage model for simultaneous load and price forecasting. The proposed algorithm uses wavelet and Kalman machines for the first stage load and price forecasting. Each of the load and price data is decomposed into different frequency components, and Kalman machine is used to forecast each frequency components of load and price data. Then a Kohonen Self Organizing Map (SOM) finds similar days of load frequency components and feeds them into the second stage forecasting machine. In addition, mutual information based feature selection is used to find the relevant price data and rank them based on their relevance. The second stage uses Multi-Layer Perceptron Artificial Neural Network (MLP-ANN) and Adaptive Neuro-Fuzzy Inference System (ANFIS) for forecasting of load and price frequency components, respectively. The third stage machine uses the second stage outputs and feeds them into its MLP-ANN and ANFIS machines to improve the load and price forecasting accuracy. The proposed three-stage algorithm is applied to Nordpool and mainland Spain power markets. The obtained results are compared with the recent load and price forecast algorithms, and showed that the three-stage algorithm presents a better performance for day-ahead electricity market load and price forecasting.

Event-based scheduling of industrial technical virtual power plant considering wind and market prices stochastic behaviors - A case study in Iran

The Virtual Power Plants containing Distributed Energy Resources are classified into two main categories of Commercial Virtual Power Plant and Technical Virtual Power Plant as a suitable way to manage industrial environments. Here, Industrial Technical Virtual Power Plant is defined as a scheduling unit containing loads and generations located in an industrial grid. A comprehensive framework is proposed here for normal and contingency conditions for various Virtual Power Plants participating in a short-term market. This framework performs a day-ahead and intra-day generation scheduling by selecting the best Demand Response programs. In this framework, the wind generations and the day-ahead and intra-day electricity market prices are considered as the stochastic parameters. A risk-management aspect is noticed in the proposed stages for contingency conditions. Then, some element changes such as seasonal load change and single-line outage are trained in the system to accredit the proposed solution in the contingency condition. Hereof, an appropriate technique is defined to represent the proposed model and solution. To specify the effectiveness and efficiency of the proposed methodology, the modified Isfahan Regional Electric Power Company network in Iran is experimented to test the method and to assess some encouraging aspects as well. By the proposed approach, attractiveness of Demand Response programs is revealed in industrial grids and the lower cost will be imposed. Also the improvement percentage of load shedding can be gained by performing the proposed scheduling that is so important for industrial processes.

Multi-market demand response using economic model predictive control of space heating in residential buildings

Several studies have evaluated the potential for residential buildings participating in demand response programs based on the day-ahead electricity market prices. However, little is known about the benefits of residential buildings providing demand response by engaging in trading on the intraday market. This paper presents a simulation-based study of the performance of an economic model predictive control scheme used to enable demand response through parallel utilization of day-ahead market prices and intraday market trading. The performance of the control scheme was evaluated by simulating ten apartments in a residential building located in Denmark through a heating season (four months) using historical market data. The results showed that the addition of intraday trading to the more conventional day-ahead market price-based control problem increased the total cost savings from 2.9% to 5.6% in the existing buildings, and 13%–19% in retrofitted buildings with higher energy-efficiency. In the existing building the proposed control scheme traded on average 12.7kWh/m2 on the intraday market throughout the simulation corresponding to 21% of the reference consumption. For a retrofitted building the traded volume was 9.6kWh/m2 which corresponds to 52% of the reference consumption. These results suggest that the benefits of considering intraday market trading as a demand response incentive mechanism apply to a wide range of buildings.

Portfolio Decision of Short-Term Electricity Forecasted Prices through Stochastic Programming

Deregulated electricity markets encourage firms to compete, making the development of renewable energy easier. An ordinary parameter of electricity markets is the electricity market price, mainly the day-ahead electricity market price. This paper describes a new approach to forecast day-ahead electricity market prices, whose methodology is divided into two parts as: (i) forecasting of the electricity price through autoregressive integrated moving average (ARIMA) models; and (ii) construction of a portfolio of ARIMA models per hour using stochastic programming. A stochastic programming model is used to forecast, allowing many input data, where filtering is needed. A case study to evaluate forecasts for the next 24 h and the portfolio generated by way of stochastic programming are presented for a specific day-ahead electricity market. The case study spans four weeks of each one of the years 2014, 2015 and 2016 using a specific pre-treatment of input data of the stochastic programming (SP) model. In addition, the results are discussed, and the conclusions are drawn.

Zonal merit-order effects of wind generation development on day-ahead and real-time electricity market prices in Texas

This paper uses a regression-based approach to explore the impact of wind generation development on wholesale electricity prices in the Electric Reliability Council of Texas (ERCOT) market. We find that wind generation development has a greater effect on real-time market (RTM) prices than day-ahead market (DAM) prices. Higher wind generation forecast errors tend to reduce the RTM prices, chiefly because unanticipated increases in wind generation reduce the real-time net loads to be served by fossil fuel power plants. Improving ERCOT’s load and wind generation forecast accuracy tends to make the DAM and RTM prices converge, thus enhancing ERCOT’s market trading efficiency. Finally, the estimated merit-order effects are greatest in the ERCOT zones where the wind generation capacity locally resides.

Predictive Densities for Day-Ahead Electricity Prices Using Time-Adaptive Quantile Regression

A large part of the decision-making problems actors of the power system are facing on a daily basis requires scenarios for day-ahead electricity market prices. These scenarios are most likely to be generated based on marginal predictive densities for such prices, then enhanced with a temporal dependence structure. A semi-parametric methodology for generating such densities is presented: it includes: (i) a time-adaptive quantile regression model for the 5%–95% quantiles; and (ii) a description of the distribution tails with exponential distributions. The forecasting skill of the proposed model is compared to that of four benchmark approaches and the well-known the generalist autoregressive conditional heteroskedasticity (GARCH) model over a three-year evaluation period. While all benchmarks are outperformed in terms of forecasting skill overall, the superiority of the semi-parametric model over the GARCH model lies in the former’s ability to generate reliable quantile estimates.

Comparative analysis of features of Polish and Lithuanian Day-ahead electricity market prices

The goal of this article is to better understand the processes of electricity market price formation in Poland and Lithuania through an analysis of the features (volatility and spikes) of Lithuanian and Polish day-ahead electricity market prices and to assess how acquired electricity price features could affect the achievement of the main goals of the national energy policy. The following indicators have been calculated to determine electricity market price volatility: the oscillation coefficient, the coefficient of variation, an adjusted coefficient of variation, the standard deviation indicator, the daily velocity indicator (based on the overall average price) and the daily velocity indicator (based on the daily average price). Critical values for electricity market price have been calculated to evaluate price spikes. This analysis reveals that electricity market-price volatility is moderate in Poland and high in Lithuania. Electricity price spikes have been an observable phenomenon both in Lithuanian and in Polish day-ahead electricity markets, but they are more common in Lithuania, encompassing 3.15% of the time period analysed in Poland and 4.68% of the time period analysed in Lithuania. Volatile, spiking and increasing electricity prices in day-ahead electricity markets in Lithuania and Poland create preconditions and substantiate the relevance of implementation of the national energy policies and measures.