Data-driven combination of METAR observations and CAMS reanalysis aerosols to enhance satellite retrieval of surface solar irradiance. [PDF]
Roy A +3 more
europepmc +1 more source
Graph Neural Network‐Based Prediction of Building Energy Consumption
A graph neural network that encodes a multi‐zone building as a graph accurately predicts hourly cooling and heating loads across three distinct climates, outperforming Random Forest and XGBoost baselines and serving as a fast surrogate to EnergyPlus simulations for scalable building energy management.
Ali Maboudi Reveshti +4 more
wiley +1 more source
LSTM-Based Estimation of Solar Energy Production Using Meteorological and Environmental Data: Karabük Case Study. [PDF]
Gultekin F, Guneser MT, Yildirim MZ.
europepmc +1 more source
This study integrates climatic simulations with machine learning to predict solar and wind energy across Iraq. Results show Random Forest excels for solar (R2 = 0.98) and neural networks for wind (R2 = 0.97), enabling a practical web tool for renewable energy planning. ABSTRACT Driven by the global shift away from fossil fuels, solar and wind resources
Bassam Musheer Kareem +3 more
wiley +1 more source
A Survey of Current Operations-Ready Thermospheric Density Models for Drag Modeling in LEO Operations. [PDF]
Mutschler S +10 more
europepmc +1 more source
Integrated Solar and Wind Energy Systems for Domestic Power Generation and CO2 Emissions Reduction
A well‐structured and sequentially methodical procedure for evaluating the Wind and Photovoltaic Hybrid Energy System‐based technical, environmental, and economic viability. This study investigates the integration of solar photovoltaic(PV) and wind energy systems for domestic electricity generation in the Gaza Strip.
Mohamed Elnaggar +9 more
wiley +1 more source
Maximum dispatchable capacity evaluation of a VPP with hybrid wind-solar-gas-storage systems. [PDF]
Zhang C, Li D, Li C.
europepmc +1 more source
A hybrid PV–wind grid‐connected microgrid for Mongla Port is optimized using HOMER Pro and validated via DIgSILENT quasi‐dynamic load flow. The system achieves a low cost of energy (0.0341 USD/kWh), 70.9% renewable penetration, improved voltage stability, and significant emission reduction, demonstrating a reliable and cost‐effective solution for ...
Md. Behesty Hasan Sagor +6 more
wiley +1 more source
Analysis and Classification of Partial Shading Conditions in Photovoltaic Arrays
This study presents new mathematical models for describing P–V curve extrema under different shading scenarios and applies machine learning classifiers that use features derived from P–V characteristics for accurate fault identification. ABSTRACT With the escalating global transition toward renewable energy, ensuring the operational stability and ...
Hamid Reza Parsa, Mohammad Sarvi
wiley +1 more source
Optimizing solar irradiance forecasting: ANN models enhanced with ADAM and Cuckoo search algorithm. [PDF]
Sadiq M +6 more
europepmc +1 more source

