Abstract
With the deregulation of the electric energy industry, accurate electricity price forecasting (EPF) is increasingly significant to market participants' bidding strategies and uncertainty risk control. However, it remains a challenging task owing to the high volatility and complicated nonlinearity of electricity prices. Aimed at this, a novel hybrid deep-learning framework is proposed for day-ahead EPF, which includes four modules: the feature preprocessing module, the deep learning-based point prediction module, the error compensation module, and the probabilistic prediction module. The feature preprocessing module is based on isolation forest (IF), and least absolute shrinkage and selection operator (Lasso), which is used to detect outliers and select the correlated features of electricity price series. The point prediction module combines the deep belief network (DBN), long-short-term memory (LSTM) neural network (RNN), and convolutional neural network (CNN), and is employed to extract complicated nonlinear features. The residual error between forecasting price and actual price can be reduced based on the error compensation module. The probabilistic prediction module based on quantile regression (QR) is used to estimate the uncertainty under various confidence levels. The PJM market data is employed in case studies to evaluate the proposed framework, and the results revealed that it has a competitive advantage compared with all of the considered comparison methods.
Highlights
Distributed energy resources (DERs), such as wind power and photovoltaic power, play a vital role in the electricity market [1]
Compared with contrast models without error compensation module (ECM), it is found that the average index of convolutional neural network (CNN) with ECM, deep belief network (DBN) with ECM, and long-short-term memory (LSTM) recurrent neural network (RNN) with ECM are improved by 2.44%, 2.96%, and 5.26%, respectively, and their variances are increased by 7.68%, 38.10%, and 60.42%, respectively
The results show that the Average coverage percentage (ACP) of CNN+quantile regression (QR) are averagely improved by 0.12%, 0.09%, 0.27%, 0.27%, 3.7%, and 2.7%, respectively, compared with DBN+QR, LSTM+QR, light gradient boosting machine (LGBM)+QR, BPNN+QR, support vector regression (SVR)+QR, and k-nearest neighbor (KNN)+QR
Summary
Distributed energy resources (DERs), such as wind power and photovoltaic power, play a vital role in the electricity market [1]. Shallow learning models are based on the principle of error minimization, and usually have better performance than physical methods and statistical methods Due to their notable capabilities in extracting features, they has been one of the most common methods for electricity prices forecasting. We need to combining feature preprocessing techniques, deep learning-based point prediction models, error compensation module, and probabilistic prediction model to rethink and design a hybrid deep learning framework for day-ahead EPF in this article. (2) The deep learning-based point prediction module combining three types of deep-learning models (DBN, LSTM, and CNN) is developed and contrasted to extract complicated nonlinear features of electricity price, and their forecasting performance is compared with shallow learning models.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
