A Prediction of the Monthly Average Daily Solar Radiation on a Horizontal Surface in Saudi Arabia Using Artificial Neural Network Approach DOI Open Access
Waleed A. Almasoud,

Saleh M. Al-Sager,

Saad S. Almady

et al.

Processes, Journal Year: 2025, Volume and Issue: 13(4), P. 1149 - 1149

Published: April 10, 2025

When planning a solar energy conversion system, having sufficiently reliable values of the monthly average daily radiation (MADSR) on horizontal surface is essential. Traditionally, estimates based other climatological variables for which more information available have been relied upon to compensate lack direct measurements. Solar varies widely, requires creation site-specific forecast models. By using artificial neural network (ANN) models or similar methods historical datasets, can be easily assessed. To verify validity established ANN model, series analyses was performed mean squared error, coefficient determination (R2), and absolute error. The study used dataset collected from nine weather stations in Saudi Arabia 1985 2000. input parameters model were maximum air relative humidity, latitude, ambient temperature, longitude, minimum sunshine duration, location altitude, corresponding month. R2 whole test 0.8449. Furthermore, sensitivity analysis showed that site elevation (location altitude) had most significant effect MADSR surface, with contribution value 14.66%. results show accurately surfaces regardless seasonal variations conditions. this work important not only its shape forecasting but also establishing practical application ANNs renewable management. will help improve utilization support sustainable efforts. proposed believed useful predicting locations climatic conditions sites. approach may functional basic strategy arrangement suitable meteorological data.

Language: Английский

Estimation of Solar Diffuse Radiation in Chongqing Based on Random Forest DOI Creative Commons

Peng Wan,

Yongjian He,

Cheng Zheng

et al.

Energies, Journal Year: 2025, Volume and Issue: 18(4), P. 836 - 836

Published: Feb. 11, 2025

Solar diffuse radiation (DIFRA) is an important component of solar radiation, but current research into the estimation DIFRA relatively limited. This study, based on remote sensing data, topographic meteorological reanalysis materials, and measured data from observation stations in Chongqing, combined key factors such as elevation angle, water vapor, aerosols, cloud cover. A high-precision model was developed using random forest algorithm, a distributed simulation Chongqing achieved. The validated 8179 points, demonstrating good predictive capability with correlation coefficient (R2) 0.72, mean absolute error (MAE) 35.99 W/m2, root square (RMSE) 50.46 W/m2. Further validation conducted 14 stations, high stability applicability across different weather conditions. In particular, fit optimal for under overcast conditions, R2 = 0.70, MAE 32.20 RMSE 47.51 results indicate that can be effectively adapted to all calculations, providing scientific basis assessing exploiting energy resources complex terrains.

Language: Английский

Citations

0
Energy-Based Data-Driven Smart Sustainable Cities Using IoT, AI, and Big Data Analytics DOI
M. Nirmala Devi, Venkatesh Baskaran,

S. Arun Kumar

et al.

˜The œurban book series, Journal Year: 2025, Volume and Issue: unknown, P. 275 - 294

Published: Jan. 1, 2025

Language: Английский

Citations

0
Seasonal Dynamics in Soil Properties Along a Roadway Corridor: A Network Analysis Approach DOI Open Access
Ibrahim Haruna Umar,

Ahmad Muhammad,

Hang Lin

et al.

Materials, Journal Year: 2025, Volume and Issue: 18(8), P. 1708 - 1708

Published: April 9, 2025

Understanding soil properties’ spatial and temporal variability is essential for optimizing road construction maintenance practices. This study investigates the seasonal of properties along a 4.8 km roadway in Maiduguri, Nigeria. Using novel integration network analysis geotechnical testing, we analyzed nine parameters (e.g., particle size distribution (PSD), Atterberg limits, California bearing ratio) across wet (September 2024) dry (January 2021) seasons from 25 test stations. Average limits (LL: 22.8% vs. 17.5% dry; PL: 18.7% 14.7% PI: 4.2% 2.8% LS: 1.8% 2.3% dry), average compaction characteristics (MDD: 1.8 Mg/m3 2.1 OMC: 12.3% 10% CBR (18.9% 27.5% dry) were obtained. Network employed z-score standardization similarity metrics, with multi-threshold (θ = 0.05, 0.10, 0.15) revealing critical structural differences. During season, networks exhibited 5.0% reduction edges (321 to 305) density decline (1.07 1.02) as thresholds tightened, contrasting dry-season retaining 99.38% connectivity (324 322 edges) stable (0.99). Seasonal shifts classification (A-4(1)/ML A-2(1)/SM underscored moisture-driven plasticity changes. The findings highlight implications adaptive design, emphasizing moisture-resistant materials optimized periods.

Language: Английский

Citations

0
A Prediction of the Monthly Average Daily Solar Radiation on a Horizontal Surface in Saudi Arabia Using Artificial Neural Network Approach DOI Open Access
Waleed A. Almasoud,

Saleh M. Al-Sager,

Saad S. Almady

et al.

Processes, Journal Year: 2025, Volume and Issue: 13(4), P. 1149 - 1149

Published: April 10, 2025

When planning a solar energy conversion system, having sufficiently reliable values of the monthly average daily radiation (MADSR) on horizontal surface is essential. Traditionally, estimates based other climatological variables for which more information available have been relied upon to compensate lack direct measurements. Solar varies widely, requires creation site-specific forecast models. By using artificial neural network (ANN) models or similar methods historical datasets, can be easily assessed. To verify validity established ANN model, series analyses was performed mean squared error, coefficient determination (R2), and absolute error. The study used dataset collected from nine weather stations in Saudi Arabia 1985 2000. input parameters model were maximum air relative humidity, latitude, ambient temperature, longitude, minimum sunshine duration, location altitude, corresponding month. R2 whole test 0.8449. Furthermore, sensitivity analysis showed that site elevation (location altitude) had most significant effect MADSR surface, with contribution value 14.66%. results show accurately surfaces regardless seasonal variations conditions. this work important not only its shape forecasting but also establishing practical application ANNs renewable management. will help improve utilization support sustainable efforts. proposed believed useful predicting locations climatic conditions sites. approach may functional basic strategy arrangement suitable meteorological data.

Language: Английский

Citations

0