Comparison of machine learning methods for photovoltaic power forecasting based on numerical weather prediction

Renewable and Sustainable Energy Reviews, Journal Year: 2022, Volume and Issue: 161, P. 112364 - 112364

Published: March 23, 2022

The increase of the worldwide installed photovoltaic (PV) capacity and intermittent nature solar resource highlights importance power forecasting for grid integration technology. This study compares 24 machine learning models deterministic day-ahead based on numerical weather predictions (NWP), tested two-year-long 15-min resolution datasets 16 PV plants in Hungary. effects predictor selection benefits hyperparameter tuning are also evaluated. results show that two most accurate kernel ridge regression multilayer perceptron with an up to 44.6% forecast skill score over persistence. Supplementing basic NWP data Sun position angles statistically processed irradiance values as inputs a 13.1% decrease root mean square error (RMSE), which underlines selection. is essential exploit full potential models, especially less robust prone under or overfitting without proper tuning. overall best forecasts have 13.9% lower RMSE compared baseline scenario using linear regression. Moreover, only daily average 1.5% higher than scenario, demonstrates effectiveness even limited availability. this paper can support both researchers practitioners constructing data-driven techniques NWP-based forecasting.

Language: Английский

---

Multiple-Input Deep Convolutional Neural Network Model for Short-Term Photovoltaic Power Forecasting

IEEE Access, Journal Year: 2019, Volume and Issue: 7, P. 74822 - 74834

Published: Jan. 1, 2019

With the fast expansion of renewable energy system installed capacity in recent years, availability, stability, and quality smart grids have become increasingly important. The output forecasting applications also been developing rapidly such techniques particularly applied fields wind solar photovoltaic (PV). In case PV forecasting, many performed with machine learning hybrid techniques. this paper, we propose a high-precision deep neural network model named PVPNet to forecast power. methodology behind proposed is based on networks, able generate 24-h probabilistic deterministic power meteorological information, as temperature, radiation, historical data. accuracy determined by mean absolute error (MAE) root square (RMSE) values. results from experiments show that MAE RMSE algorithm are 109.4845 163.1513, respectively. prove prediction outperforms other benchmark models, effectively predicts complex time series high degree volatility irregularity.

Language: Английский

---

Solar photovoltaic power forecasting using optimized modified extreme learning machine technique

Engineering Science and Technology an International Journal, Journal Year: 2018, Volume and Issue: 21(3), P. 428 - 438

Published: May 8, 2018

Prediction of photovoltaic power is a significant research area using different forecasting techniques mitigating the effects uncertainty generation. Increasingly high penetration level (PV) generation arises in smart grid and microgrid concept. Solar source irregular nature as result PV intermittent highly dependent on irradiance, temperature other atmospheric parameters. Large scale to conventional system introduces challenges energy management. It very critical do exact solar power/irradiance order secure economic operation grid. In this paper an extreme learning machine (ELM) technique used for real time model whose location given Table 1. Here associated with incremental conductance (IC) maximum point tracking (MPPT) that based proportional integral (PI) controller which simulated MATLAB/SIMULINK software. To train single layer feed-forward network (SLFN), ELM algorithm implemented weights are updated by particle swarm optimization (PSO) their performance compared existing models like back propagation (BP) model.

Language: Английский

---

Day-Ahead Hourly Forecasting of Power Generation From Photovoltaic Plants

IEEE Transactions on Sustainable Energy, Journal Year: 2017, Volume and Issue: 9(2), P. 831 - 842

Published: Oct. 12, 2017

The ability to accurately forecast power generation from renewable sources is nowadays recognized as a fundamental skill improve the operation of systems. Despite general interest community in this topic, it not always simple compare different forecasting methodologies, and infer impact single components providing accurate predictions. In paper, we extensively methodologies with more sophisticated ones over 32 photovoltaic (PV) plants sizes technology whole year. Also, try evaluate weather conditions forecasts on prediction PV generation.

Language: Английский

---

The connotation of digital twin, and the construction and application method of shop-floor digital twin

Robotics and Computer-Integrated Manufacturing, Journal Year: 2020, Volume and Issue: 68, P. 102075 - 102075

Published: Nov. 2, 2020

Language: Английский

---

Photovoltaic power forecast based on satellite images considering effects of solar position

Applied Energy, Journal Year: 2021, Volume and Issue: 302, P. 117514 - 117514

Published: Aug. 13, 2021

Language: Английский

---

Solar photovoltaic power prediction using artificial neural network and multiple regression considering ambient and operating conditions

Energy Conversion and Management, Journal Year: 2023, Volume and Issue: 288, P. 117186 - 117186

Published: May 18, 2023

Language: Английский

---

Long term performance, losses and efficiency analysis of a 960 kW P photovoltaic system in the Mediterranean climate

Energy Conversion and Management, Journal Year: 2017, Volume and Issue: 145, P. 169 - 181

Published: May 5, 2017

Language: Английский

---

Short-term power prediction for photovoltaic power plants using a hybrid improved Kmeans-GRA-Elman model based on multivariate meteorological factors and historical power datasets

Energy Conversion and Management, Journal Year: 2018, Volume and Issue: 177, P. 704 - 717

Published: Oct. 11, 2018

Language: Английский

---

Long-term load forecasting: models based on MARS, ANN and LR methods

Central European Journal of Operations Research, Journal Year: 2018, Volume and Issue: 27(4), P. 1033 - 1049

Published: March 6, 2018

Language: Английский

---