1. Forecasting of price signals using deep recurrent models.
- Author
-
Gundu, Venkateswarlu and Simon, Sishaj P.
- Abstract
This paper discussed a novel recurrent neural network architecture for electricity price forecasting in a distribution system. Uncertainty in price forecasts misleads utilities in making bidding plans, and investments, and being aware of the risks involved. Accurate price forecasting assists utilities in developing effective bidding strategies and making appropriate investment decisions. Hence, in this paper, a novel feature combination such as the day ahead, similar day, and the combination of both day ahead and similar day using PSO-based LSTM and GRU network models are presented. The proposed method involves the optimal selection of the recurrent network model for electricity price forecasting. Finally, an experimental study is carried out to select the layers and nodes of the network model. In this analysis, the performance of the proposed model is evaluated using the mean absolute percentage error. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF