Deep Learning Neural Networks Trained with MODIS Satellite-Derived Predictors for Long-Term Global Solar Radiation Prediction
Solar energy predictive models designed to emulate the long-term (e.g., monthly) global solar radiation (<i>GSR</i>) trained with satellite-derived predictors can be employed as decision tenets in the exploration, installation and management of solar energy production systems in remote a...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2019-06-01
|
| Series: | Energies |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1996-1073/12/12/2407 |
| _version_ | 1811263211191664640 |
|---|---|
| author | Sujan Ghimire Ravinesh C Deo Nawin Raj Jianchun Mi |
| author_facet | Sujan Ghimire Ravinesh C Deo Nawin Raj Jianchun Mi |
| author_sort | Sujan Ghimire |
| collection | DOAJ |
| description | Solar energy predictive models designed to emulate the long-term (e.g., monthly) global solar radiation (<i>GSR</i>) trained with satellite-derived predictors can be employed as decision tenets in the exploration, installation and management of solar energy production systems in remote and inaccessible solar-powered sites. In spite of a plethora of models designed for <i>GSR</i> prediction, deep learning, representing a state-of-the-art intelligent tool, remains an attractive approach for renewable energy exploration, monitoring and forecasting. In this paper, algorithms based on deep belief networks and deep neural networks are designed to predict long-term <i>GSR</i>. Deep learning algorithms trained with publicly-accessible Moderate Resolution Imaging Spectroradiometer (MODIS) satellite data are tested in Australia’s solar cities to predict the monthly <i>GSR:</i> single hidden layer and ensemble models. The monthly-scale MODIS-derived predictors (2003−2018) are adopted, with 15 diverse feature selection approaches including a Gaussian Emulation Machine for sensitivity analysis used to select optimal MODIS-predictor variables to simulate <i>GSR</i> against ground-truth values. Several statistical score metrics are adopted to comprehensively verify surface <i>GSR</i> simulations to ascertain the practicality of deep belief and deep neural networks. In the testing phase, deep learning models generate significantly lower absolute percentage bias (≤3%) and high Kling−Gupta efficiency (≥97.5%) values compared to the single hidden layer and ensemble model. This study ascertains that the optimal MODIS input variables employed in <i>GSR</i> prediction for solar energy applications can be relatively different for diverse sites, advocating a need for feature selection prior to the modelling of <i>GSR</i>. The proposed deep learning approach can be adopted to identify solar energy potential proactively in locations where it is impossible to install an environmental monitoring data acquisition instrument. Hence, MODIS and other related satellite-derived predictors can be incorporated for solar energy prediction as a strategy for long-term renewable energy exploration. |
| first_indexed | 2024-04-12T19:40:40Z |
| format | Article |
| id | doaj.art-38309380943247ccb962402d397dd3bc |
| institution | Directory Open Access Journal |
| issn | 1996-1073 |
| language | English |
| last_indexed | 2024-04-12T19:40:40Z |
| publishDate | 2019-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Energies |
| spelling | doaj.art-38309380943247ccb962402d397dd3bc2022-12-22T03:19:05ZengMDPI AGEnergies1996-10732019-06-011212240710.3390/en12122407en12122407Deep Learning Neural Networks Trained with MODIS Satellite-Derived Predictors for Long-Term Global Solar Radiation PredictionSujan Ghimire0Ravinesh C Deo1Nawin Raj2Jianchun Mi3School of Agricultural Computational and Environmental Sciences, Centre for Sustainable Agricultural Systems & Centre for Applied Climate Sciences, University of Southern Queensland, Springfield, QLD 4300, AustraliaSchool of Agricultural Computational and Environmental Sciences, Centre for Sustainable Agricultural Systems & Centre for Applied Climate Sciences, University of Southern Queensland, Springfield, QLD 4300, AustraliaSchool of Agricultural Computational and Environmental Sciences, Centre for Sustainable Agricultural Systems & Centre for Applied Climate Sciences, University of Southern Queensland, Springfield, QLD 4300, AustraliaSchool of Agricultural Computational and Environmental Sciences, Centre for Sustainable Agricultural Systems & Centre for Applied Climate Sciences, University of Southern Queensland, Springfield, QLD 4300, AustraliaSolar energy predictive models designed to emulate the long-term (e.g., monthly) global solar radiation (<i>GSR</i>) trained with satellite-derived predictors can be employed as decision tenets in the exploration, installation and management of solar energy production systems in remote and inaccessible solar-powered sites. In spite of a plethora of models designed for <i>GSR</i> prediction, deep learning, representing a state-of-the-art intelligent tool, remains an attractive approach for renewable energy exploration, monitoring and forecasting. In this paper, algorithms based on deep belief networks and deep neural networks are designed to predict long-term <i>GSR</i>. Deep learning algorithms trained with publicly-accessible Moderate Resolution Imaging Spectroradiometer (MODIS) satellite data are tested in Australia’s solar cities to predict the monthly <i>GSR:</i> single hidden layer and ensemble models. The monthly-scale MODIS-derived predictors (2003−2018) are adopted, with 15 diverse feature selection approaches including a Gaussian Emulation Machine for sensitivity analysis used to select optimal MODIS-predictor variables to simulate <i>GSR</i> against ground-truth values. Several statistical score metrics are adopted to comprehensively verify surface <i>GSR</i> simulations to ascertain the practicality of deep belief and deep neural networks. In the testing phase, deep learning models generate significantly lower absolute percentage bias (≤3%) and high Kling−Gupta efficiency (≥97.5%) values compared to the single hidden layer and ensemble model. This study ascertains that the optimal MODIS input variables employed in <i>GSR</i> prediction for solar energy applications can be relatively different for diverse sites, advocating a need for feature selection prior to the modelling of <i>GSR</i>. The proposed deep learning approach can be adopted to identify solar energy potential proactively in locations where it is impossible to install an environmental monitoring data acquisition instrument. Hence, MODIS and other related satellite-derived predictors can be incorporated for solar energy prediction as a strategy for long-term renewable energy exploration.https://www.mdpi.com/1996-1073/12/12/2407global solar radiationenergy securitydeep learningdeep belief networkdeep neural networksolar cities in Australia |
| spellingShingle | Sujan Ghimire Ravinesh C Deo Nawin Raj Jianchun Mi Deep Learning Neural Networks Trained with MODIS Satellite-Derived Predictors for Long-Term Global Solar Radiation Prediction Energies global solar radiation energy security deep learning deep belief network deep neural network solar cities in Australia |
| title | Deep Learning Neural Networks Trained with MODIS Satellite-Derived Predictors for Long-Term Global Solar Radiation Prediction |
| title_full | Deep Learning Neural Networks Trained with MODIS Satellite-Derived Predictors for Long-Term Global Solar Radiation Prediction |
| title_fullStr | Deep Learning Neural Networks Trained with MODIS Satellite-Derived Predictors for Long-Term Global Solar Radiation Prediction |
| title_full_unstemmed | Deep Learning Neural Networks Trained with MODIS Satellite-Derived Predictors for Long-Term Global Solar Radiation Prediction |
| title_short | Deep Learning Neural Networks Trained with MODIS Satellite-Derived Predictors for Long-Term Global Solar Radiation Prediction |
| title_sort | deep learning neural networks trained with modis satellite derived predictors for long term global solar radiation prediction |
| topic | global solar radiation energy security deep learning deep belief network deep neural network solar cities in Australia |
| url | https://www.mdpi.com/1996-1073/12/12/2407 |
| work_keys_str_mv | AT sujanghimire deeplearningneuralnetworkstrainedwithmodissatellitederivedpredictorsforlongtermglobalsolarradiationprediction AT ravineshcdeo deeplearningneuralnetworkstrainedwithmodissatellitederivedpredictorsforlongtermglobalsolarradiationprediction AT nawinraj deeplearningneuralnetworkstrainedwithmodissatellitederivedpredictorsforlongtermglobalsolarradiationprediction AT jianchunmi deeplearningneuralnetworkstrainedwithmodissatellitederivedpredictorsforlongtermglobalsolarradiationprediction |