Improving Solar Radiation Forecasting in Cloudy Conditions by Integrating Satellite Observations

Solar radiation forecasting is the basis of building a robust solar power system. Most ground-based forecasting methods are unable to consider the impact of cloud changes on future solar radiation. To alleviate this limitation, this study develops a hybrid network which relies on a convolutional neu...

Full description

Saved in:
Bibliographic Details
Main Authors: Qiangsheng Bu, Shuyi Zhuang, Fei Luo, Zhigang Ye, Yubo Yuan, Tianrui Ma, Tao Da
Format: Article
Language:English
Published: MDPI AG 2024-12-01
Series:Energies
Subjects:
Online Access:https://www.mdpi.com/1996-1073/17/24/6222
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850241438088953856
author Qiangsheng Bu
Shuyi Zhuang
Fei Luo
Zhigang Ye
Yubo Yuan
Tianrui Ma
Tao Da
author_facet Qiangsheng Bu
Shuyi Zhuang
Fei Luo
Zhigang Ye
Yubo Yuan
Tianrui Ma
Tao Da
author_sort Qiangsheng Bu
collection DOAJ
description Solar radiation forecasting is the basis of building a robust solar power system. Most ground-based forecasting methods are unable to consider the impact of cloud changes on future solar radiation. To alleviate this limitation, this study develops a hybrid network which relies on a convolutional neural network to extract cloud motion patterns from time series of satellite observations and a long short-term memory neural network to establish the relationship between future solar radiation and cloud information, as well as antecedent measurements. We carefully select the optimal scales to consider the spatial and temporal correlations of solar radiation and design test experiments at ten stations to check the model performance in various climate zones. The results demonstrate that the solar radiation forecasting accuracy is considerably improved, particularly in cloudy conditions, compared with purely ground-based models. The maximum magnitude of improvements reaches up to 50 W/m<sup>2</sup> (15%) in terms of the (relative) root mean squared error (RMSE) for 1 h ahead forecasts. The network achieves superior forecasts with correlation coefficients varying from 0.96 at 1 h ahead to 0.85 at 6 h ahead. Forecast errors are related to cloud regimes, of which the cloud amount leads to a maximum relative RMSE difference of about 50% with an additional 5% from cloud variability. This study ascertains that multi-source data fusion contributes to a better simulation of cloud impacts and a combination of different deep learning techniques enables more reliable forecasts of solar radiation. In addition, multi-step forecasts with a low latency make the advance planning and management of solar energy possible in practical applications.
format Article
id doaj-art-9147971dcca645c8a44d74cd015def98
institution OA Journals
issn 1996-1073
language English
publishDate 2024-12-01
publisher MDPI AG
record_format Article
series Energies
spelling doaj-art-9147971dcca645c8a44d74cd015def982025-08-20T02:00:37ZengMDPI AGEnergies1996-10732024-12-011724622210.3390/en17246222Improving Solar Radiation Forecasting in Cloudy Conditions by Integrating Satellite ObservationsQiangsheng Bu0Shuyi Zhuang1Fei Luo2Zhigang Ye3Yubo Yuan4Tianrui Ma5Tao Da6State Grid Jiangsu Electric Power Company Ltd. Research Institute, Nanjing 211103, ChinaState Grid Jiangsu Electric Power Company Ltd. Research Institute, Nanjing 211103, ChinaState Grid Jiangsu Electric Power Company Ltd. Research Institute, Nanjing 211103, ChinaState Grid Jiangsu Electric Power Company Ltd. Research Institute, Nanjing 211103, ChinaState Grid Jiangsu Electric Power Company Ltd. Research Institute, Nanjing 211103, ChinaState Grid Jiangsu Electric Power Company Ltd. Zhenjiang Power Supply Branch, Zhenjiang 212002, ChinaState Grid Jiangsu Electric Power Company Ltd. Zhenjiang Power Supply Branch, Zhenjiang 212002, ChinaSolar radiation forecasting is the basis of building a robust solar power system. Most ground-based forecasting methods are unable to consider the impact of cloud changes on future solar radiation. To alleviate this limitation, this study develops a hybrid network which relies on a convolutional neural network to extract cloud motion patterns from time series of satellite observations and a long short-term memory neural network to establish the relationship between future solar radiation and cloud information, as well as antecedent measurements. We carefully select the optimal scales to consider the spatial and temporal correlations of solar radiation and design test experiments at ten stations to check the model performance in various climate zones. The results demonstrate that the solar radiation forecasting accuracy is considerably improved, particularly in cloudy conditions, compared with purely ground-based models. The maximum magnitude of improvements reaches up to 50 W/m<sup>2</sup> (15%) in terms of the (relative) root mean squared error (RMSE) for 1 h ahead forecasts. The network achieves superior forecasts with correlation coefficients varying from 0.96 at 1 h ahead to 0.85 at 6 h ahead. Forecast errors are related to cloud regimes, of which the cloud amount leads to a maximum relative RMSE difference of about 50% with an additional 5% from cloud variability. This study ascertains that multi-source data fusion contributes to a better simulation of cloud impacts and a combination of different deep learning techniques enables more reliable forecasts of solar radiation. In addition, multi-step forecasts with a low latency make the advance planning and management of solar energy possible in practical applications.https://www.mdpi.com/1996-1073/17/24/6222solar radiation forecastingconvolutional neural networkcloud amounttemporal and spatial scalesolar energy
spellingShingle Qiangsheng Bu
Shuyi Zhuang
Fei Luo
Zhigang Ye
Yubo Yuan
Tianrui Ma
Tao Da
Improving Solar Radiation Forecasting in Cloudy Conditions by Integrating Satellite Observations
Energies
solar radiation forecasting
convolutional neural network
cloud amount
temporal and spatial scale
solar energy
title Improving Solar Radiation Forecasting in Cloudy Conditions by Integrating Satellite Observations
title_full Improving Solar Radiation Forecasting in Cloudy Conditions by Integrating Satellite Observations
title_fullStr Improving Solar Radiation Forecasting in Cloudy Conditions by Integrating Satellite Observations
title_full_unstemmed Improving Solar Radiation Forecasting in Cloudy Conditions by Integrating Satellite Observations
title_short Improving Solar Radiation Forecasting in Cloudy Conditions by Integrating Satellite Observations
title_sort improving solar radiation forecasting in cloudy conditions by integrating satellite observations
topic solar radiation forecasting
convolutional neural network
cloud amount
temporal and spatial scale
solar energy
url https://www.mdpi.com/1996-1073/17/24/6222
work_keys_str_mv AT qiangshengbu improvingsolarradiationforecastingincloudyconditionsbyintegratingsatelliteobservations
AT shuyizhuang improvingsolarradiationforecastingincloudyconditionsbyintegratingsatelliteobservations
AT feiluo improvingsolarradiationforecastingincloudyconditionsbyintegratingsatelliteobservations
AT zhigangye improvingsolarradiationforecastingincloudyconditionsbyintegratingsatelliteobservations
AT yuboyuan improvingsolarradiationforecastingincloudyconditionsbyintegratingsatelliteobservations
AT tianruima improvingsolarradiationforecastingincloudyconditionsbyintegratingsatelliteobservations
AT taoda improvingsolarradiationforecastingincloudyconditionsbyintegratingsatelliteobservations