Advanced Solar and Load Forecasting Incorporating HD Sky Imaging (Phase III)

As the deployment of solar power utilizing devices increases, accurate forecasts of solar energy become increasingly important for managing and stabilizing the electricity grid. This study investigates the validity of 7-day ahead global horizontal irradiance (GHI) forecasts for South Africa provided by the Global Forecast System (GFS). The clear-sky index was used to interpolate GFS mixed 3-hourly and 6-hourly averages to generate hourly forecasts. The model’s performance was validated against radiometric GHI data from 15 locations spread across South Africa’s six climatic zones. The results showed that the GFS model can satisfactorily forecast GHI in South Africa. Overall, the model performed best in interior climatic zones and relatively poorly in coastal climatic zones. The accuracy of the GFS model performance decreases as the forecast lead time increases. The model overestimates GHI in overcast conditions while it underestimates GHI in clear and partly cloudy conditions. On average, the model underestimates GHI in all seasons. Forecasting errors are highest in summer and lowest in winter and moderate in autumn and spring. The inadequacy of the GFS model to precisely determine the cloud quantities is the major source of error; errors increase as the cloud fraction increases. Despite this, the study’s findings suggest that GFS forecasts may be used for forecasting solar photovoltaic output for energy planning in South Africa. Significance Statement This paper provides the first comprehensive assessment of Global Forecast System (GFS) forecasts of global horizontal irradiance (GHI) in six climatic regions of South Africa, exploring how cloud cover, site location, climatic zone, season, forecast horizon, and hour of the day impact the model’s performance. This research was carried out in a country where GFS GHI models have never been validated before and that presents a wide range of climatic conditions. The results allow the accuracy of other global and mesoscale GHI forecasting models to be compared against the current results in the region. The outcomes of this study indicate the possibility of using GFS GHI forecasts for forecasting solar photovoltaic output for energy planning. South Africa has an electricity supply deficit, and being able to accurately forecast the output of the increasing capacity of solar PV is one of the things needed to stabilize the grid.

NYSolarCast: A solar power forecasting system for New York State

For many critical energy-related applications, such as reliable PV power production, accurate Global Horizontal Irradiance (GHI) short-term forecasts are crucial. Forecasts of GHI, as analyzed in this study, are produced by the state-of-the-art, high-resolution forecasts developed within the Destination Earth initiative, the ECMWF Digital Twin Engine using the ExtremeDT (Weather-Induced Extremes Digital Twin) dataset, that produces daily global simulations at resolutions of 4.4 km kilometres up to four days ahead. However the GHI ExtremesDT forecasts do not take into account the aerosol effects, which may introduce systematic biases, especially during periods of high aerosol load events. We investigated the impact of aerosols on DT GHI forecasts and the associated PV power predictions within the context of the DestinE Destination Renewable Energy (DRE) use case.We analyze almost one year of data comprising two-day-ahead DT GHI forecasts and the corresponding aerosol optical depth (AOD) forecasts from the Copernicus Atmosphere Monitoring Service (CAMS). The DT GHI forecasts are corrected for aerosol effects using fast radiative transfer model techniques utilising the lidRadtran [1, 2] package.Within the DRE project, and based on co-design activities with the project's end user, site-specific PV power production forecasts tailored to the user’s needs and infrastructure were developed, using DT GHI forecasts and a machine learning (ML) model that was trained on the user’s historical data. The impact of aerosol correction was also evaluated by comparing PV power production forecasts derived from GHI forecasts with and without aerosol correction against the actual PV power production of the plant.A machine learning (ML) model, trained on historical, site-specific production data from the QUEST PV park, was created within the DRE project to convert predicted solar irradiance into PV park power output in order to evaluate the impact on power production. The ML model propagates the original and aerosol-corrected GHI forecasts, which are then compared to actual production.The results show how CAMS-based aerosol correction of GHI forecasts can reduce bias and consistently improve PV power prediction. Results show the value of incorporating atmospheric composition data with ML-based power conversion models for operational energy applications, as well as the significance of aerosol representation in solar forecasting.AcknowledgmentsDRE project has received funding from the European Space Agency under the DESTINATION EARTH USE CASES – DESP USE CASES - ROUND 1. The duration of the project is 12 months (November 2023 - November 2024).We would like also to acknowledge the COST Action HARMONIA (International network for harmonization of atmospheric aerosol retrievals from ground based photometers), CA21119.Bibliography(1) Mayer, B.; Kylling, A. Atmospheric Chemistry and Physics 2005, 5, 1855–1877.(2) Emde, C.; Buras-Schnell, R.; Kylling, A.; Mayer, B.; Gasteiger, J.; Hamann, U.; Kylling, J.; Richter, B.; Pause, C.; Dowling, T.; Bugliaro, L. Geoscientific Model Development 2016, 9, 1647–1672.

Accurate short-term GHI forecasting using a novel temporal convolutional network model

Long-term average solar radiation prediction via seasonal Global Horizontal Irradiance (GHI) forecasting is increasingly leveraging Machine Learning (ML) to uncover complex relationships in historical GHI and climatic parameters surpassing traditional methods. This study investigates seasonal GHI forecasting across four climatically distinct Indian locations namely Chennai, Jaisalmer, Leh and Mawsynram using data from the National Solar Radiation Database. Following rigorous data quality assessment and feature selection based on Spearman's correlation and Mutual Information, eight key features are identified for training three ML models: Efficient Linear Regression (ELR), Regression Trees (RT) and Gaussian Process Regression (GPR). The GPR model demonstrates superior accuracy in seasonal GHI prediction across diverse Indian climatic conditions making it ideal for reliable and cost-efficient Stand-Alone Photovoltaic (SAPV) systems. It is observed that the GPR model achieves the lowest Root Mean Square Error (RMSE) of 0.0030, Mean Absolute Error (MAE) of 0.0022 and Coefficient of Determination (R²) of 0.9999 achieving a reduction of 189.1% in RMSE, 190.09% in MAE and 20.56% improvement in R² compared to ELR. Furthermore, the GPR model surpasses the RT model with a reduction of 124.05% in RMSE and 111.1% in MAE, together with an improvement of 0.2604% in R². While RT provides reasonably accurate forecasts, ELR shows the least reliability. These results confirm GPR's exceptional precision and ability to capture nearly all dataset variability.

A Transformer-based multimodal-learning framework using sky images for ultra-short-term solar irradiance forecasting

Aerosols affect surface solar radiation (SSR) by modifying cloud properties. For solar energy applications, either for photovoltaics (PV) or concentrating solar power (CSP) systems, understanding of spatio-temporal variability of aerosol is essential for short-term forecasts of solar radiation.  This study aims to quantify interday variability of the commonly used proxy for optical properties of aerosols - aerosol optical depth (AOD) at the wavelength of 500 nm and to assess the deviation in direct normal irradiance (DNI) and in global horizontal irradiance (GHI) caused by AOD variation using radiative transfer modeling. The following AOD data are used: AOD forecasts for the next day from the Copernicus Atmosphere Monitoring Service (CAMS); ground-based AOD measurements from the Aerosol Robotic Network (AERONET); ModIs/Terra high spatial resolution data set (MIDAS) which is based on satellite retrievals. We select locations with different aerosol types around the globe based on AERONET stations, where at each site more than 1500 daily values for consecutive days from 2010 to 2020 are available. We adopt the persistence approach using AOD data from AERONET and MIDAS and compare them with CAMS forecasts. The AOD data from AERONET serve as ground truth for validation. Under cloudless conditions, AOD variability informs the deviation of DNI and GHI forecasts from the observation. Furthermore, we attempt a worldwide clustering of aerosol types based on chemical modeling. Interday AOD variability is high in megacities with high aerosol load and in the proximity of highly variable aerosol sources such as desert dust and biomass burning (fires). This work should deepen our knowledge of spatio-temporal variability of aerosol in the context of AOD and SSR prediction, thus contributing to the quality and reliability of solar resource assessment.

The performance of solar photovoltaic (PV) systems has been widely investigated based on the low-frequency data such as hourly or daily solar data. However, its short-term performance monitored at high frequency has been relatively less studied. This paper presents the short-term performance of a grid-connected solar photovoltaic (PV) system monitored at 1-min interval from July 1, 2010 to June 30, 2011 and compares with the daily performance. The summary results based on mean, standard deviation, and coefficient of variation (CV) are presented. The results indicate that the yearly means of the minutely array output power and the minutely global horizontal irradiance (GHI) in a typical day are bell-shaped. The CV pattern of the minutely array output power appears to be U-shaped while the CV pattern of the minutely GHI is varying within a horizontal zone of [0, 5%]. More importantly, the minutely data reveal that the output power responds nonlinearly to changes in solar irradiance while the hourly data conventionally used suggest a linear relationship between them. A piecewise quadratic model was built to fit the nonlinear relationship between array output power and GHI based on the minutely data. The model was proved to be accurate with high value of R2 and validated by the measured data.

The purpose of the article is to analyze changes in solar heat gains and wind characteristics across various regions of Ukraine based on historical, current, and projected data, in order to assess the potential for using renewable energy sources (RES), considering the challenges associated with extreme climatic conditions and the need for decentralized energy supply. The methodology includes the application of synthesis and analysis methods for climatic data. Findings an analysis and comparison of meteorological data from the global Meteonorm climate database, normative indicators according to DSTU 9190:2022, as well as climate scenarios RCP (2.6, 4.5, 8.5). This study analyzes the changes in solar radiation gains and wind characteristics across different regions of Ukraine, based on historical climatic data, Meteonorm meteorological data, national standards, and climate projections reflecting various levels of global warming. It was found that for most regions, current annual solar radiation values exceed the normative ones. Moreover, under all climate change scenarios for the cities of Odesa and Kyiv, a further increase in the global horizontal irradiance (GHI) is expected by 2100. The average hourly GHI maxima in 25 Ukrainian cities do not exceed the critical threshold of 1100 W/m²; however, potentially hazardous overirradiance events may occur at shorter (mi- nutescale) intervals and require further investigation. Based on the analysis of hourly wind direction data for the current period, four regions with similar prevailing wind directions were identified. In Region 1, the dominant wind directions are western and south-western; in Region 2 – western and north-western; in Region 3 – western; and in Region 4 – western, north-western, and northern. It was established that the territory of Ukraine is characterized by a low frequency of strong winds, with maximum wind speeds ranging from 13.5 to 16.8 m/s. The predominance of light winds (up to 5 m/s for 79 % of the time) indicates limited wind energy feasibility and minimal structural risk in terms of wind loads. The projected wind regime dynamics for Kyiv and Odesa up to 2100 under RCP 2.6, RCP 4.5, and RCP 8.5 scenarios do not show a significant increase in extreme values. Originality is found in the comprehensive comparison of climatic and normative GHI values and wind characteristics across temporal and spatial scales, with consideration of risks arising from potential extreme weather events. The practical value of the research results lies in the ability to account for regional climatic differences when designing decentralized energy systems based on renewable energy sources (RES), taking into account historical, current, and future climate change scenarios.

Recent advances and the widespread availability of smartphones have ushered in a new wave of innovations in healthcare. We present our initial experience with Endockscope, a new docking system that optimizes the coupling of the iPhone 4S with modern endoscopes. Using the United States Air Force resolution target, we compared the image resolution (line pairs/mm) of a flexible cystoscope coupled to the Endockscope+iPhone to the Storz high definition (HD) camera (H3-Z Versatile). We then used the Munsell ColorChecker chart to compare the color resolution with a 0° laparoscope. Furthermore, 12 expert endoscopists blindly compared and evaluated images from a porcine model using a cystoscope and ureteroscope for both systems. Finally, we also compared the cost (average of two company listed prices) and weight (lb) of the two systems. Overall, the image resolution allowed by the Endockscope was identical to the traditional HD camera (4.49 vs 4.49 lp/mm). Red (ΔE=9.26 vs 9.69) demonstrated better color resolution for iPhone, but green (ΔE=7.76 vs 10.95), and blue (ΔE=12.35 vs 14.66) revealed better color resolution with the Storz HD camera. Expert reviews of cystoscopic images acquired with the HD camera were superior in image, color, and overall quality (P=0.002, 0.042, and 0.003). In contrast, the ureteroscopic reviews yielded no statistical difference in image, color, and overall (P=1, 0.203, and 0.120) quality. The overall cost of the Endockscope+iPhone was $154 compared with $46,623 for a standard HD system. The weight of the mobile-coupled system was 0.47 lb and 1.01 lb for the Storz HD camera. Endockscope demonstrated feasibility of coupling endoscopes to a smartphone. The lighter and inexpensive Endockscope acquired images of the same resolution and acceptable color resolution. When evaluated by expert endoscopists, the quality of the images overall were equivalent for flexible ureteroscopy and somewhat inferior, but still acceptable for flexible cystoscopy.

The long-term trends of total surface solar radiation (SSR), surface diffuse radiation, and surface air temperature were analyzed in this study based on updated 48-yr data from 55 observational stations in China, and then the correlation between SSR and the diurnal temperature range (DTR) was studied. The effect of total solar radiation on surface air temperature in China was investigated on the basis of the above analyses. A strong correlation between SSR and DTR was found for the period 1961–2008 in China. The highest correlation and steepest regression line slope occurred in winter, indicating that the solar radiation effect on DTR was the largest in this season. Clouds and water vapor have strong influences on both SSR and DTR, and hence on their relationship. The largest correlations between SSR and DTR occurred in wintertime in northern China, regardless of all-day (including clear days and cloudy days) or clear-day cases. Our results also showed that radiation arriving at the surface in China decreased significantly during 1961–1989 (dimming period), but began to increase during 1990–2008 (brightening period), in agreement with previous global studies. The reduction of total SSR offset partially the greenhouse warming during 1961–1989. However, with the increase of SSR after 1990, this offsetting effect vanished; on the contrary, it even made a contribution to the accelerated warming. Nonetheless, the greenhouse warming still played a controlling role because of the increasing of minimum and mean surface temperatures in the whole study period of 1961–2008. We estimated that the greenhouse gases alone may have caused surface temperatures to rise by 0.31–0.46°C (10 yr)−1 during 1961–2008, which is higher than previously estimated. Analysis of the corresponding changes in total solar radiation, diffuse radiation, and total cloud cover indicated that the dimming and brightening phenomena in China were likely attributable to increases in absorptive and scattering aerosols in the atmosphere, respectively.

This study investigates the possibility of changes in daily scale solar radiation and precipitation variability. Coefficients of variation (CVs) were computed for the daily downward surface solar radiation product from the International Satellite Cloud Climatology Project and the daily precipitation product from the Global Precipitation Climatology Project. Regression analysis was used to identify trends in CVs. Statistically significant changes in solar radiation variability were found for 35% of the globe, and particularly large increases were found for tropical Africa and the Maritime Continent. These increases in solar radiation variability were correlated with increases in precipitation variability and increases in deep convective cloud amount. The changes in high-frequency climate variability identified here have consequences for any process depending nonlinearly on climate, including solar energy production and terrestrial ecosystem photosynthesis. To assess these consequences, additional work is needed to understand how high-frequency climate variability will change in the coming decades.

Understanding the influence of climatological factors on solar energy is crucial for optimizing solar potential assessment, power generation, and load variation analysis. Despite Ethiopia's significant solar energy potential, the impacts of key climatic variables such as temperature (TΔ, Ta, Tmax), Cloud Cover (CC), and Relative Humidity (RH) fluctuations on Global Horizontal Irradiance (GHI) remain understudied, particularly in remote regions. This research bridges this gap by evaluating how these factors affect GHI distribution across Ethiopia. The study employs a multiple linear regression model, integrating seasonal climatic parameters with a Prescott sunshine-based model to predict GHI. Additionally, spatial interpolation via ArcGIS mapping analyzes the variations in cloud cover (35 sites of NASA data and nine sites from meteorological data), and relative humidity of (26 sites of meteorological data) against 73 sites' solar radiation patterns. Key uniqueness includes the integration of multi-climatic predictors on solar modeling and a spatially explicit assessment of Ethiopia’s solar potential under varying weather conditions. Results reveal that monthly average solar irradiance ranges from 3.56 kWh/m²/day (July, Awasa) to 6.78 kWh/m²/day (February, Mekele). When climatic factors (Ta, RH, CC) are incorporated, predicted GHI varies between 3.81 kWh/m²/day (July, Jimma) and 6.59 kWh/m²/day (February, Bahir Dar). The model demonstrates high accuracy, with a multiple correlation coefficient of R = 0.999 Awasa. Statistical validation confirms robustness, showing minimal errors. Mean Bias Error (MBE: (-0.65 to -0.48) kWh/m²/day; Mean Absolute Percentage Error MAPE: (0.004–0.344%). Spatial analysis indicates that cloud cover and RH reduce GHI in western, southwestern, and central highlands, whereas drier northeastern/southeastern regions receive higher radiation. The discussion underscores the implications for PV system sizing, energy demand management, and equipment selection based on climate-solar relationships. In conclusion, this study provides critical insights into Ethiopia’s solar energy dynamics, emphasizing the need for climate-informed solar planning. The findings support policymakers and energy developers in optimizing solar resource utilization under varying meteorological conditions.

When considering the amount of shortwave radiation incident on a photovoltaic solar array and, therefore, the amount and stability of the energy output from the system, clouds represent the greatest source of short-term (i.e., scale of minutes to hours) variability through scattering and reflection of incoming solar radiation. Providing estimates of this short-term variability is important for determining and regulating the output from large solar arrays as they connect with the larger power infrastructure. In support of the installation of a 37-MW solar array on the grounds of Brookhaven National Laboratory (BNL), a study of the impacts of clouds on the output of the solar array has been undertaken. The study emphasis is on predicting the change in surface solar radiation resulting from the observed/forecast cloud field on a 5-minute time scale. At these time scales, advection of cloud elements over the solar array is of particular importance. As part of the BNL Aerosol Life Cycle Intensive Operational Period (IOP), a 915-MHz Radar Wind Profiler (RWP) was deployed to determine the profile of low-level horizontal winds and the depth of the planetary boundary layer. The initial deployment mission of the 915-MHz RWP for cloud forecasting has been expanded the deployment to provide horizontal wind measurements for estimating and constraining cloud advection speeds. A secondary focus is on the observation of dynamics and microphysics of precipitation during cold season/winter storms on Long Island. In total, the profiler was deployed at BNL for 1 year from May 2011 through May 2012.

Purpose: This study is to derive energy demand and solar power generation potential through CEA(City Energy Analyst) modeling for mixed type apartments in Gangnam-gu, where energy consumption is high. Method: For urban building energy modeling, EPW(EnergyPlus Weather) files were used to analyze the building energy demand and potential after installation of solar panels through CEA modeling. In the case of building energy and solar panels, solar radiation and energy-use types are derived for each building through the Solar Radiation and Schedule model in the CEA model. Result: The cooling and heating energy of buildings in the study site is greatly affected by ambient temperature and humidity, and it was found that there was a large fluctuation by season due to changes in solar radiation due to dynamics with surrounding building s and building structure.The study site had a mixed type apartment structure, so it was found that the potential for soloar power was high when mainly installed on the east wall. Using the results of this study, it is expected to be used as apreliminary data for stakeholders in future decision-making and research by presenting the necessity of research on building energy structure in cities for efficient building energy consumption and solar power generation.