Virtual multi-camera sky images for regional solar irradiation forecasts

neobjavljeni prilog sa skupa

neobjavljeni prilog sa skupa

Virtual multi-camera sky images for regional solar irradiation forecasts

Vrsta prilog sa skupa (neobjavljen)
Tip neobjavljeni prilog sa skupa
Godina 2025
Status neobjavljeno

Sažetak

The integration of photovoltaic (PV) systems into power grids is a challenge due to the variability of
solar energy. With the global increase in the adoption of renewable energy sources, especially PV
systems, new challenges have arisen in managing and integrating these unpredictable energy sources
into existing power grids. Accurate short-term forecasting of solar irradiance is essential for grid
stability and efficient resource allocation. Conventional models rely on single sky cameras, which limits
regional applicability and fails to capture the spatial variability of cloud cover over larger areas. This
paper presents a synthetic dataset for regional solar forecasting created with the Unity Game Engine.
The simulation emulates cloud dynamics over a 50 × 50 km area with virtual sky cameras capturing
hemispherical images and solar irradiance data. This multi-camera setup enables the modelling of cloud
movements and their effects on solar radiation. The dataset provides a controlled environment for
testing and validating machine learning models. By simulating diverse meteorological conditions, it
provides insights into the spatial and temporal relationships between cloud cover and solar radiation.
An effective solution to mitigate the challenges of integrating PV systems into power grids is the
accurate short-term forecasting of solar irradiance and PV power. Initial analyses confirm the potential
of the dataset to improve regional forecast accuracy. More accurate solar irradiance forecasts enhance
PV integration, reduce reliance on backup power, and strengthen grid resilience. This synthetic
approach lays the groundwork for future advancements in renewable energy forecasting.

Ključne riječi

Multi-camera systems; Photovoltaic systems; Regional prediction; Renewable energy forecasting; Solar irradiance forecasting; Synthetic data set