The ability to forecast real-time electricity price for wind power is key to the operation of energy markets and hedging price risks. In new research published in the journal Applied Energy, Professor Kristen Schell, working with her PhD student Haolin Yang at Rensselaer Polytechnic Institute in the U.S., develop novel machine learning algorithms to boost the accuracy of these price forecasts.
The research group employed new deep neural network (DNN) architectures, which recent research suggests can capture temporal dependencies in historical price data, along with the ability to automatically extract important features of the dataset. Unfortunately, most existing price prediction DNN representations utilize basic architecture designs and either no pre-training, or simple training approaches. In this research, Schell and Yang studied the effect of transfer learning on three network representations and different source domains, as well as the mechanism of transfer learning. Their research showed that transfer learning improves accuracy across all network representations, and that the best performance is obtained with a novel, GRU-based architecture, termed GRU-TL, that has been pre-trained from a hybrid dataset of all wind farms in the same subzone. This model outperforms all statistical and deep learning benchmarks by an average of approximately 7% in the mean absolute percent error (MAPE) metric. The research group concludes that the underlying mechanism of transfer learning enables the pre-trained DNN representation to learn the features of the target dataset more accurately.