Short-Term Load Forecasting Using Encoder-Decoder WaveNet: Application to the French Grid
The prediction of time series data applied to the energy sector (prediction of renewable energy production, forecasting prosumers’ consumption/generation, forecast of country-level consumption, etc.) has numerous useful applications. Nevertheless, the complexity and non-linear behaviour associated w...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-04-01
|
| Series: | Energies |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1996-1073/14/9/2524 |
| _version_ | 1797535979924357120 |
|---|---|
| author | Fernando Dorado Rueda Jaime Durán Suárez Alejandro del Real Torres |
| author_facet | Fernando Dorado Rueda Jaime Durán Suárez Alejandro del Real Torres |
| author_sort | Fernando Dorado Rueda |
| collection | DOAJ |
| description | The prediction of time series data applied to the energy sector (prediction of renewable energy production, forecasting prosumers’ consumption/generation, forecast of country-level consumption, etc.) has numerous useful applications. Nevertheless, the complexity and non-linear behaviour associated with such kind of energy systems hinder the development of accurate algorithms. In such a context, this paper investigates the use of a state-of-art deep learning architecture in order to perform precise load demand forecasting 24-h-ahead in the whole country of France using RTE data. To this end, the authors propose an encoder-decoder architecture inspired by WaveNet, a deep generative model initially designed by Google DeepMind for raw audio waveforms. WaveNet uses dilated causal convolutions and skip-connection to utilise long-term information. This kind of novel ML architecture presents different advantages regarding other statistical algorithms. On the one hand, the proposed deep learning model’s training process can be parallelized in GPUs, which is an advantage in terms of training times compared to recurrent networks. On the other hand, the model prevents degradations problems (explosions and vanishing gradients) due to the residual connections. In addition, this model can learn from an input sequence to produce a forecast sequence in a one-shot manner. For comparison purposes, a comparative analysis between the most performing state-of-art deep learning models and traditional statistical approaches is presented: Autoregressive-Integrated Moving Average (ARIMA), Long-Short-Term-Memory, Gated-Recurrent-Unit (GRU), Multi-Layer Perceptron (MLP), causal 1D-Convolutional Neural Networks (1D-CNN) and ConvLSTM (Encoder-Decoder). The values of the evaluation indicators reveal that WaveNet exhibits superior performance in both forecasting accuracy and robustness. |
| first_indexed | 2024-03-10T11:53:04Z |
| format | Article |
| id | doaj.art-6e56210c54a14cd9ace359788452e398 |
| institution | Directory Open Access Journal |
| issn | 1996-1073 |
| language | English |
| last_indexed | 2024-03-10T11:53:04Z |
| publishDate | 2021-04-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Energies |
| spelling | doaj.art-6e56210c54a14cd9ace359788452e3982023-11-21T17:32:30ZengMDPI AGEnergies1996-10732021-04-01149252410.3390/en14092524Short-Term Load Forecasting Using Encoder-Decoder WaveNet: Application to the French GridFernando Dorado Rueda0Jaime Durán Suárez1Alejandro del Real Torres2IDENER, 41300 Seville, SpainIDENER, 41300 Seville, SpainDepartment of Systems and Automation, University of Seville, 41092 Seville, SpainThe prediction of time series data applied to the energy sector (prediction of renewable energy production, forecasting prosumers’ consumption/generation, forecast of country-level consumption, etc.) has numerous useful applications. Nevertheless, the complexity and non-linear behaviour associated with such kind of energy systems hinder the development of accurate algorithms. In such a context, this paper investigates the use of a state-of-art deep learning architecture in order to perform precise load demand forecasting 24-h-ahead in the whole country of France using RTE data. To this end, the authors propose an encoder-decoder architecture inspired by WaveNet, a deep generative model initially designed by Google DeepMind for raw audio waveforms. WaveNet uses dilated causal convolutions and skip-connection to utilise long-term information. This kind of novel ML architecture presents different advantages regarding other statistical algorithms. On the one hand, the proposed deep learning model’s training process can be parallelized in GPUs, which is an advantage in terms of training times compared to recurrent networks. On the other hand, the model prevents degradations problems (explosions and vanishing gradients) due to the residual connections. In addition, this model can learn from an input sequence to produce a forecast sequence in a one-shot manner. For comparison purposes, a comparative analysis between the most performing state-of-art deep learning models and traditional statistical approaches is presented: Autoregressive-Integrated Moving Average (ARIMA), Long-Short-Term-Memory, Gated-Recurrent-Unit (GRU), Multi-Layer Perceptron (MLP), causal 1D-Convolutional Neural Networks (1D-CNN) and ConvLSTM (Encoder-Decoder). The values of the evaluation indicators reveal that WaveNet exhibits superior performance in both forecasting accuracy and robustness.https://www.mdpi.com/1996-1073/14/9/2524time series forecastingenergy consumption forecastingdeep learningmachine learningconvolutional neural networksartificial neural networks |
| spellingShingle | Fernando Dorado Rueda Jaime Durán Suárez Alejandro del Real Torres Short-Term Load Forecasting Using Encoder-Decoder WaveNet: Application to the French Grid Energies time series forecasting energy consumption forecasting deep learning machine learning convolutional neural networks artificial neural networks |
| title | Short-Term Load Forecasting Using Encoder-Decoder WaveNet: Application to the French Grid |
| title_full | Short-Term Load Forecasting Using Encoder-Decoder WaveNet: Application to the French Grid |
| title_fullStr | Short-Term Load Forecasting Using Encoder-Decoder WaveNet: Application to the French Grid |
| title_full_unstemmed | Short-Term Load Forecasting Using Encoder-Decoder WaveNet: Application to the French Grid |
| title_short | Short-Term Load Forecasting Using Encoder-Decoder WaveNet: Application to the French Grid |
| title_sort | short term load forecasting using encoder decoder wavenet application to the french grid |
| topic | time series forecasting energy consumption forecasting deep learning machine learning convolutional neural networks artificial neural networks |
| url | https://www.mdpi.com/1996-1073/14/9/2524 |
| work_keys_str_mv | AT fernandodoradorueda shorttermloadforecastingusingencoderdecoderwavenetapplicationtothefrenchgrid AT jaimeduransuarez shorttermloadforecastingusingencoderdecoderwavenetapplicationtothefrenchgrid AT alejandrodelrealtorres shorttermloadforecastingusingencoderdecoderwavenetapplicationtothefrenchgrid |