BRIM: An Accurate Electricity Spot Price Prediction Scheme-Based Bidirectional Recurrent Neural Network and Integrated Market

Abstract

For the benefit from accurate electricity price forecasting, not only can various electricity market stakeholders make proper decisions to gain profit in a competitive environment, but also power system stability can be improved. Nevertheless, because of the high volatility and uncertainty, it is an essential challenge to accurately forecast the electricity price. Considering that recurrent neural networks (RNNs) are suitable for processing time series data, in this paper, we propose a bidirectional long short-term memory (LSTM)-based forecasting model, BRIM, which splits the state neurons of a regular RNN into two parts: the forward states (using the historical electricity price information) are designed for processing the data in positive time direction and backward states (using the future price information available at inter-connected markets) for the data in negative time direction. Moreover, due to the fact that inter-connected power exchange markets show a common trend for other neighboring markets and can provide signaling information for each other, it is sensible to incorporate and exploit the impact of the neighboring markets on forecasting accuracy of electricity price. Specifically, future electricity prices of the interconnected market are utilized both as input features for forward LSTM and backward LSTM. By testing on day-ahead electricity prices in the European Power Exchange (EPEX), the experimental results show the superiority of the proposed method BRIM in enhancing predictive accuracy in comparison with the various benchmarks, and moreover Diebold-Mariano (DM) shows that the forecast accuracy of BRIM is not equal to other forecasting models, and thus indirectly demonstrates that BRIM statistically significantly outperforms other schemes.

Forecasting day-ahead electricity prices: A review of state-of-the-art algorithms, best practices and an open-access benchmark

“…Similarly, as with most studies in 2018, the new studies were more limited than [12,59] as no comparisons with state-of-the-art statistical methods were made and long test datasets were seldom used. In this context, even though some studies [16,88] tried to compare the proposed methods with existing DL models [59], they either failed to re-estimated the benchmark models for the new case study [16] or they overfitted the DL benchmark models [88].…”

“…Conversely, studies proposing new ML methods only compare them with simple statistical methods [12][13][14][15][16] and show that ML models are more accurate.…”

Transfer Learning for Electricity Price Forecasting

“…Lago et al [39] integrates the French prices to forecast the Belgian prices. Chen et al [40] apply bidirectional integrated market based LSTM model to forecast the French electricity prices. Their integration part follows the framework of Ziel et al [38] and the forecasts are compared with the benchmark models [38; 39; 1].…”

“…For some specific problems, information which comes after time step t can be useful. Electricity prices are effected by past and future values [40]. We propose a bidirectional gated recurrent unit (BGRU) to overcome the limitation of basic GRU.…”

A comparison of modern deep neural network architectures for energy spot price forecasting

“…Di Persio and Honchar (2016a), Di Persio and Honchar (2016b), Di Persio and Honchar (2017a) and Di Persio and Honchar (2017b), but very few application of NN to energy markets are present in the literature, see, e.g. Bento et al (2018), Chen et al (2019), Di Persio and Honchar (2017b), Lago (2018a), Lago et al (2018b), Panapakidis and Dagoumas (2016), Zhang et al (2018) and Zhou et al (2019).…”