How Does Neural Network Model Capacity Affect Photovoltaic Power Prediction? A Study Case
The use of models capable of forecasting the production of photovoltaic (PV) energy is essential to guarantee the best possible integration of this energy source into traditional distribution grids. Long Short-Term Memory networks (LSTMs) are commonly used for this purpose, but their use may not be...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-01-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/3/1357 |
| _version_ | 1797623270152863744 |
|---|---|
| author | Carlos Henrique Torres de Andrade Gustavo Costa Gomes de Melo Tiago Figueiredo Vieira Ícaro Bezzera Queiroz de Araújo Allan de Medeiros Martins Igor Cavalcante Torres Davi Bibiano Brito Alana Kelly Xavier Santos |
| author_facet | Carlos Henrique Torres de Andrade Gustavo Costa Gomes de Melo Tiago Figueiredo Vieira Ícaro Bezzera Queiroz de Araújo Allan de Medeiros Martins Igor Cavalcante Torres Davi Bibiano Brito Alana Kelly Xavier Santos |
| author_sort | Carlos Henrique Torres de Andrade |
| collection | DOAJ |
| description | The use of models capable of forecasting the production of photovoltaic (PV) energy is essential to guarantee the best possible integration of this energy source into traditional distribution grids. Long Short-Term Memory networks (LSTMs) are commonly used for this purpose, but their use may not be the better option due to their great computational complexity and slower inference and training time. Thus, in this work, we seek to evaluate the use of neural networks MLPs (Multilayer Perceptron), Recurrent Neural Networks (RNNs), and LSTMs, for the forecast of 5 min of photovoltaic energy production. Each iteration of the predictions uses the last 120 min of data collected from the PV system (power, irradiation, and PV cell temperature), measured from 2019 to mid-2022 in Maceió (Brazil). In addition, Bayesian hyperparameters optimization was used to obtain the best of each model and compare them on an equal footing. Results showed that the MLP performs satisfactorily, requiring much less time to train and forecast, indicating that they can be a better option when dealing with a very short-term forecast in specific contexts, for example, in systems with little computational resources. |
| first_indexed | 2024-03-11T09:26:17Z |
| format | Article |
| id | doaj.art-972dcbee106c48b9986134368f61d3a2 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-11T09:26:17Z |
| publishDate | 2023-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-972dcbee106c48b9986134368f61d3a22023-11-16T17:59:54ZengMDPI AGSensors1424-82202023-01-01233135710.3390/s23031357How Does Neural Network Model Capacity Affect Photovoltaic Power Prediction? A Study CaseCarlos Henrique Torres de Andrade0Gustavo Costa Gomes de Melo1Tiago Figueiredo Vieira2Ícaro Bezzera Queiroz de Araújo3Allan de Medeiros Martins4Igor Cavalcante Torres5Davi Bibiano Brito6Alana Kelly Xavier Santos7Computing Institute, A. C. Simões Campus, Federal University of Alagoas—UFAL, Maceió 57072-970, BrazilComputing Institute, A. C. Simões Campus, Federal University of Alagoas—UFAL, Maceió 57072-970, BrazilCenter of Agrarian Sciences, Engineering and Agricultural Sciences Campus, Federal University of Alagoas—UFAL, Rio Largo 57100-000, BrazilComputing Institute, A. C. Simões Campus, Federal University of Alagoas—UFAL, Maceió 57072-970, BrazilElectrical Engineering Department, Center of Technology, Federal University of Rio Grande do Norte—UFRN, Natal 59072-970, BrazilCenter of Agrarian Sciences, Engineering and Agricultural Sciences Campus, Federal University of Alagoas—UFAL, Rio Largo 57100-000, BrazilComputing Institute, A. C. Simões Campus, Federal University of Alagoas—UFAL, Maceió 57072-970, BrazilCenter of Agrarian Sciences, Engineering and Agricultural Sciences Campus, Federal University of Alagoas—UFAL, Rio Largo 57100-000, BrazilThe use of models capable of forecasting the production of photovoltaic (PV) energy is essential to guarantee the best possible integration of this energy source into traditional distribution grids. Long Short-Term Memory networks (LSTMs) are commonly used for this purpose, but their use may not be the better option due to their great computational complexity and slower inference and training time. Thus, in this work, we seek to evaluate the use of neural networks MLPs (Multilayer Perceptron), Recurrent Neural Networks (RNNs), and LSTMs, for the forecast of 5 min of photovoltaic energy production. Each iteration of the predictions uses the last 120 min of data collected from the PV system (power, irradiation, and PV cell temperature), measured from 2019 to mid-2022 in Maceió (Brazil). In addition, Bayesian hyperparameters optimization was used to obtain the best of each model and compare them on an equal footing. Results showed that the MLP performs satisfactorily, requiring much less time to train and forecast, indicating that they can be a better option when dealing with a very short-term forecast in specific contexts, for example, in systems with little computational resources.https://www.mdpi.com/1424-8220/23/3/1357photovoltaic solar energyshort term energy forecastmultilayer perceptronrecurrent neural networklong short-term memory |
| spellingShingle | Carlos Henrique Torres de Andrade Gustavo Costa Gomes de Melo Tiago Figueiredo Vieira Ícaro Bezzera Queiroz de Araújo Allan de Medeiros Martins Igor Cavalcante Torres Davi Bibiano Brito Alana Kelly Xavier Santos How Does Neural Network Model Capacity Affect Photovoltaic Power Prediction? A Study Case Sensors photovoltaic solar energy short term energy forecast multilayer perceptron recurrent neural network long short-term memory |
| title | How Does Neural Network Model Capacity Affect Photovoltaic Power Prediction? A Study Case |
| title_full | How Does Neural Network Model Capacity Affect Photovoltaic Power Prediction? A Study Case |
| title_fullStr | How Does Neural Network Model Capacity Affect Photovoltaic Power Prediction? A Study Case |
| title_full_unstemmed | How Does Neural Network Model Capacity Affect Photovoltaic Power Prediction? A Study Case |
| title_short | How Does Neural Network Model Capacity Affect Photovoltaic Power Prediction? A Study Case |
| title_sort | how does neural network model capacity affect photovoltaic power prediction a study case |
| topic | photovoltaic solar energy short term energy forecast multilayer perceptron recurrent neural network long short-term memory |
| url | https://www.mdpi.com/1424-8220/23/3/1357 |
| work_keys_str_mv | AT carloshenriquetorresdeandrade howdoesneuralnetworkmodelcapacityaffectphotovoltaicpowerpredictionastudycase AT gustavocostagomesdemelo howdoesneuralnetworkmodelcapacityaffectphotovoltaicpowerpredictionastudycase AT tiagofigueiredovieira howdoesneuralnetworkmodelcapacityaffectphotovoltaicpowerpredictionastudycase AT icarobezzeraqueirozdearaujo howdoesneuralnetworkmodelcapacityaffectphotovoltaicpowerpredictionastudycase AT allandemedeirosmartins howdoesneuralnetworkmodelcapacityaffectphotovoltaicpowerpredictionastudycase AT igorcavalcantetorres howdoesneuralnetworkmodelcapacityaffectphotovoltaicpowerpredictionastudycase AT davibibianobrito howdoesneuralnetworkmodelcapacityaffectphotovoltaicpowerpredictionastudycase AT alanakellyxaviersantos howdoesneuralnetworkmodelcapacityaffectphotovoltaicpowerpredictionastudycase |