Electricity price prediction based on hybrid model of adam optimized LSTM neural network and wavelet transform
By Zihan Chang, Yang Zhang, Wenbo Chen. Energy, 2019
342 citations recorded by the source corpus.
To a large extent, electricity price prediction is a daunting task because it depends on factors, such as weather, fuel, load and bidding strategies etc. Those features generate a lot of fluctuations to electricity price. As a type of RNN, LSTM has a good performance on processing time series data as well as some nonlinear and complex problems. To explore more accurate electricity price forecasting approach, in this paper, a new hybrid model based on wavelet transform and Adam optimized LSTM neural network, denoted as WT-Adam-LSTM, is proposed. After the wavelet transform, nonlinear sequence of electricity price can be decomposed and processed data will have a more stable variance, and the combination of Adam, one of efficient stochastic gradient-based optimizers, and LSTM can capture appropriate behaviors precisely for electricity price. This study presented four cases to verify the performance of the hybrid model, and the dataset from New South Wales of Australia and French were adopted to illustrate the excellence of the hybrid model. The results show that the proposed model can significantly improve the prediction accuracy.
Subfields: Electrical and Electronic Engineering, Management Science and Operations Research
View Paper: https://www.semanticscholar.org/paper/20a1eeb5e67c20c46e7cc3af844ef8c82b07d196
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Electrical and Electronic Engineering
- Grzegorz Marcjasz
- B. Schutter
- Jesus Lago
- Rafal Weron
- Florian Ziel
- Jakub Nowotarski
- Bartosz Uniejewski
- Stefan Trück
In Management Science and Operations Research
Citations in the corpus
- Forecasting day-ahead electricity prices: A review of state-of-the-art algorithms, best practices and an open-access benchmark (2020, methodology): “• In many of the existing studies [17–23] the testing periods are too short to yield conclusive results.”