Cloud Detection for PV Power Forecast based on Colour Components of Sky Images

Integrating and operating PV in a system-efficient way is the key to increasing penetration levels to evolve toward a more sustainable and clean energy system. Reliable generation forecasting is the solution to tackle fluctuations in PV power generation and its negative effects on power systems and energy markets. Clouds play a key role in solar power, so cloud detection is important in PV power generation forecast. A color-based cloud detection algorithm is proposed in this paper. It considers 12 different color channels together with the K-means algorithm to detect clouds in sky images from an All-Sky Camera. The result of the proposed method has been compared to both classic and state-of-the-art methods. The experimental results show that the proposed approach has better performance in comparison to state-of-the-art methods.