Deep solar radiation forecasting with convolutional neural network and long short-term memory network algorithms

This paper designs a hybridized deep learning framework that integrates the Convolutional Neural Network for pattern recognition with the Long Short-Term Memory Network for half-hourly global solar radiation (GSR) forecasting. The Convolution network is applied to robustly extract data input features from predictive variables (i.e., statistically significant antecedent inputs) while Long Short-Term Memory absorbs them for prediction. Half-hourly GSR for Alice Springs (Australia: 01 January 2006 to 31 August 2018) are extracted with stationarity checks applied via unit-root and mutual information test to capture antecedent GSR values required to forecast future GSR. The proposed hybrid model is benchmarked with standalone models as well as other Deep Learning, Single Hidden Layer and Tree based models. The results show that the benchmarked models are not able to generate satisfactory GSR predictions and the proposed hybrid model outperforms all other counterparts. The hybrid model registers superior results with over 70% of predictive errors lying below ±10 Wm−2 and outperforms the benchmark model for 1-Day half-hourly GSR prediction with low Relative Root Mean Square Error (≈1.515%), Mean Absolute Percentage Error (≈4.672%) and Absolute Percentage Bias (≈1.233%). This study ascertains that a proposed hybrid model based on a convolution network framework can accurately predict GSR and enable energy availability to be regularly monitored over multi-step horizons when coupled with a low latency Long Short-Term Memory network. Furthermore, it also concludes that the proposed model can have practical implications in forecasting GSR, capitalizing its versatility as a stratagem in monitoring solar powered systems by integrating freely available solar radiation into a real power grid system.

[1]  Wojciech Zaremba,et al.  An Empirical Exploration of Recurrent Network Architectures , 2015, ICML.

[2]  Llanos Mora-López,et al.  Modeling and forecasting hourly global solar radiation using clustering and classification techniques , 2016 .

[3]  Abinet Tesfaye Eseye,et al.  Short-term photovoltaic solar power forecasting using a hybrid Wavelet-PSO-SVM model based on SCADA and Meteorological information , 2018 .

[4]  Didier Mayer,et al.  Performance Prediction of Grid-Connected Photovoltaic Systems Using Remote Sensing , 2008 .

[5]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[6]  J. Nash,et al.  River flow forecasting through conceptual models part I — A discussion of principles☆ , 1970 .

[7]  Jun Yan,et al.  Multiscale Convolutional Neural Networks for Fault Diagnosis of Wind Turbine Gearbox , 2019, IEEE Transactions on Industrial Electronics.

[8]  Mohanad S. Al-Musaylh,et al.  Two-phase particle swarm optimized-support vector regression hybrid model integrated with improved empirical mode decomposition with adaptive noise for multiple-horizon electricity demand forecasting , 2018 .

[9]  Ravinesh C. Deo,et al.  Wavelet-based 3-phase hybrid SVR model trained with satellite-derived predictors, particle swarm optimization and maximum overlap discrete wavelet transform for solar radiation prediction , 2019, Renewable and Sustainable Energy Reviews.

[10]  Cihan H. Dagli,et al.  Solar Irradiance Forecasting Using Deep Neural Networks , 2017 .

[11]  Irena Koprinska,et al.  Deep Learning for Big Data Time Series Forecasting Applied to Solar Power , 2018, SOCO-CISIS-ICEUTE.

[12]  Adam R. Brandt,et al.  Solar PV output prediction from video streams using convolutional neural networks , 2018 .

[13]  Ravinesh C. Deo,et al.  Global solar radiation prediction by ANN integrated with European Centre for medium range weather forecast fields in solar rich cities of Queensland Australia , 2019, Journal of Cleaner Production.

[14]  D. Legates,et al.  Evaluating the use of “goodness‐of‐fit” Measures in hydrologic and hydroclimatic model validation , 1999 .

[15]  L. Mora-López,et al.  Multiplicative ARMA models to generate hourly series of global irradiation , 1998 .

[16]  K. Kaba,et al.  Estimation of daily global solar radiation using deep learning model , 2018, Energy.

[17]  Geoffrey E. Hinton,et al.  Deep Learning , 2015, Nature.

[18]  Yacine Rezgui,et al.  Predictive modelling for solar thermal energy systems: A comparison of support vector regression, random forest, extra trees and regression trees , 2018, Journal of Cleaner Production.

[19]  H. Hejase,et al.  Time-Series Regression Model for Prediction of Mean Daily Global Solar Radiation in Al-Ain, UAE , 2012 .

[20]  Stefan Lessmann,et al.  A comparative study of LSTM neural networks in forecasting day-ahead global horizontal irradiance with satellite data , 2018 .

[21]  R. Deo,et al.  Forecasting long-term global solar radiation with an ANN algorithm coupled with satellite-derived (MODIS) land surface temperature (LST) for regional locations in Queensland , 2017 .

[22]  Juan José Pantrigo,et al.  Convolutional Neural Networks and Long Short-Term Memory for skeleton-based human activity and hand gesture recognition , 2018, Pattern Recognit..

[23]  Xiaoxia Qi,et al.  Deep belief network based k-means cluster approach for short-term wind power forecasting , 2018, Energy.

[24]  J. A. Ruiz-Arias,et al.  Comparison of numerical weather prediction solar irradiance forecasts in the US, Canada and Europe , 2013 .

[25]  Ravinesh C. Deo,et al.  Self-adaptive differential evolutionary extreme learning machines for long-term solar radiation prediction with remotely-sensed MODIS satellite and Reanalysis atmospheric products in solar-rich cities , 2018, Remote Sensing of Environment.

[26]  C. W. Tong,et al.  A new hybrid support vector machine–wavelet transform approach for estimation of horizontal global solar radiation , 2015 .

[27]  B. Boruff,et al.  Assessing the potential for concentrated solar power development in rural Australia , 2010 .

[28]  T. Chai,et al.  Root mean square error (RMSE) or mean absolute error (MAE)? – Arguments against avoiding RMSE in the literature , 2014 .

[29]  R. Deo,et al.  Universally deployable extreme learning machines integrated with remotely sensed MODIS satellite predictors over Australia to forecast global solar radiation: A new approach , 2019, Renewable and Sustainable Energy Reviews.

[30]  Radha Krishna Prasad,et al.  Exergetic performance prediction of solar air heater using MLP, GRNN and RBF models of artificial neural network technique. , 2018, Journal of environmental management.

[31]  Nazri Mohd Nawi,et al.  The Effect of Data Pre-processing on Optimized Training of Artificial Neural Networks , 2013 .

[32]  Amimul Ahsan,et al.  Solar panels: Real efficiencies, potential productions and payback periods for major Australian cities , 2018 .

[33]  Digby Race,et al.  Power to change: Analysis of household participation in a renewable energy and energy efficiency programme in Central Australia , 2015 .

[34]  Chen Jie,et al.  Wind speed forecasting using nonlinear-learning ensemble of deep learning time series prediction and extremal optimization , 2018, Energy Conversion and Management.

[35]  Ravinesh C. Deo,et al.  Deep Learning Neural Networks Trained with MODIS Satellite-Derived Predictors for Long-Term Global Solar Radiation Prediction , 2019, Energies.

[36]  M F Sanner,et al.  Python: a programming language for software integration and development. , 1999, Journal of molecular graphics & modelling.

[37]  Yongqian Liu,et al.  Short-term forecasting and uncertainty analysis of wind turbine power based on long short-term memory network and Gaussian mixture model , 2019, Applied Energy.

[38]  Navdeep Jaitly,et al.  Towards End-To-End Speech Recognition with Recurrent Neural Networks , 2014, ICML.

[39]  John O. Carter,et al.  Using spatial interpolation to construct a comprehensive archive of Australian climate data , 2001, Environ. Model. Softw..

[40]  Yunpeng Wang,et al.  Long short-term memory neural network for traffic speed prediction using remote microwave sensor data , 2015 .

[41]  Thomas Reindl,et al.  Short term solar irradiance forecasting using a mixed wavelet neural network , 2016 .

[42]  Hsu-Yung Cheng,et al.  Predicting solar irradiance with all-sky image features via regression , 2013 .

[43]  Yugang Niu,et al.  Hourly day-ahead solar irradiance prediction using weather forecasts by LSTM , 2018 .

[44]  Shiwen Mao,et al.  LASSO and LSTM Integrated Temporal Model for Short-Term Solar Intensity Forecasting , 2019, IEEE Internet of Things Journal.

[45]  M. Abd Elaziz,et al.  Modeling of solar energy systems using artificial neural network: A comprehensive review , 2019, Solar Energy.

[46]  Jikai Liu,et al.  Bioactive polyketides and 8,14-seco-ergosterol from fruiting bodies of the ascomycete Daldinia childiae. , 2017, Phytochemistry.

[47]  Xiaoxia Qi,et al.  A comparison of day-ahead photovoltaic power forecasting models based on deep learning neural network , 2019, Applied Energy.

[48]  Stéphanie Monjoly,et al.  Hourly forecasting of global solar radiation based on multiscale decomposition methods: A hybrid approach , 2017 .

[49]  Junwei Wang,et al.  Gated recurrent units based neural network for time heterogeneous feedback recommendation , 2018, Inf. Sci..

[50]  Shanlin Yang,et al.  Optimal load dispatch of community microgrid with deep learning based solar power and load forecasting , 2019, Energy.

[51]  M. Collares-Pereira,et al.  TAG: A time-dependent, autoregressive, Gaussian model for generating synthetic hourly radiation , 1992 .

[52]  Cyril Voyant,et al.  Multi-horizon solar radiation forecasting for Mediterranean locations using time series models , 2013, ArXiv.

[53]  Hoshin Vijai Gupta,et al.  Decomposition of the mean squared error and NSE performance criteria: Implications for improving hydrological modelling , 2009 .

[54]  Jürgen Schmidhuber,et al.  LSTM: A Search Space Odyssey , 2015, IEEE Transactions on Neural Networks and Learning Systems.

[55]  Yasuhiro Hayashi,et al.  Feature Extraction of NWP Data for Wind Power Forecasting Using 3D-Convolutional Neural Networks , 2018, Energy Procedia.

[56]  Gaël Varoquaux,et al.  Scikit-learn: Machine Learning in Python , 2011, J. Mach. Learn. Res..

[57]  Yi Qin,et al.  The Optimized Deep Belief Networks With Improved Logistic Sigmoid Units and Their Application in Fault Diagnosis for Planetary Gearboxes of Wind Turbines , 2019, IEEE Transactions on Industrial Electronics.

[58]  Oliver Kramer,et al.  Input quality aware convolutional LSTM networks for virtual marine sensors , 2018, Neurocomputing.

[59]  Frederico G. Guimarães,et al.  Short-term load forecasting by using a combined method of convolutional neural networks and fuzzy time series , 2019, Energy.

[60]  Michelangelo Ceci,et al.  Predictive Modeling of PV Energy Production: How to Set Up the Learning Task for a Better Prediction? , 2017, IEEE Transactions on Industrial Informatics.

[61]  Hui Liu,et al.  Smart deep learning based wind speed prediction model using wavelet packet decomposition, convolutional neural network and convolutional long short term memory network , 2018 .

[62]  Khubaib Amjad Alam,et al.  Support vector regression based prediction of global solar radiation on a horizontal surface , 2015 .

[63]  Ali Mostafaeipour,et al.  Prediction of wind speed using a new Grey-extreme learning machine hybrid algorithm: A case study , 2019 .

[64]  M. Collares-Pereira,et al.  Simple procedure for generating sequences of daily radiation values using a library of Markov transition matrices , 1988 .

[65]  Jie Zhang,et al.  LSTM-EFG for wind power forecasting based on sequential correlation features , 2019, Future Gener. Comput. Syst..

[66]  Yitao Liu,et al.  Deep belief network based deterministic and probabilistic wind speed forecasting approach , 2016 .

[67]  Gordon Reikard.  Predicting solar radiation at high resolutions: A comparison of time series forecasts , 2009 .