Prediction of Solar Irradiance and Photovoltaic Solar Energy Product Based on Cloud Coverage Estimation Using Machine Learning Methods

Abstract

Cloud cover estimation from images taken by sky-facing cameras can be an important input for analyzing current weather conditions and estimating photovoltaic power generation. The constant change in position, shape, and density of clouds, however, makes the development of a robust computational method for cloud cover estimation challenging. Accurately determining the edge of clouds and hence the separation between clouds and clear sky is difficult and often impossible. Toward determining cloud cover for estimating photovoltaic output, we propose using machine learning methods for cloud segmentation. We compare several methods including a classical regression model, deep learning methods, and boosting methods that combine results from the other machine learning models. To train each of the machine learning models with various sky conditions, we supplemented the existing Singapore whole sky imaging segmentation database with hazy and overcast images collected by a camera-equipped Waggle sensor node. We found that the U-Net architecture, one of the deep neural networks we utilized, segmented cloud pixels most accurately. However, the accuracy of segmenting cloud pixels did not guarantee high accuracy of estimating solar irradiance. We confirmed that the cloud cover ratio is directly related to solar irradiance. Additionally, we confirmed that solar irradiance and solar power output are closely related; hence, by predicting solar irradiance, we can estimate solar power output. This study demonstrates that sky-facing cameras with machine learning methods can be used to estimate solar power output. This ground-based approach provides an inexpensive way to understand solar irradiance and estimate production from photovoltaic solar facilities.

Wind, Solar, and Photovoltaic Renewable Energy Systems with and without Energy Storage Optimization: A Survey of Advanced Machine Learning and Deep Learning Techniques

“…The current Singapore's entire sky imaging separation database is updated with foggy and cloudy photos taken by a webcam Waggle sensor node to train each of the machine learning techniques with diverse sky circumstances [107]. One of the deep networks that have been used, the U-Net architecture, segregated cloud pixels one of the most correctly.…”

“…18 [107] Solar energy PV Solar power based on cloud coverage evaluation using machine learning approach.…”

Trendline Assessment of Solar Energy Potential in Hungary and Current Scenario of Renewable Energy in the Visegrád Countries for Future Sustainability

“…The irradiance is correlated with cloud cover predictions. Thus, correct analysis of irradiance can be considered in compliance with the sky cover [25]. For understanding the irradiance distribution of the Visegrád Countries, the PV-GIS interactive tool was used to observe the distribution in the four countries [26].…”

A Survey of the Researches on Grid-Connected Solar Power Generation Systems and Power Forecasting Methods Based on Ground-Based Cloud Atlas

“…Cloud map-based prediction also includes both satellite-based cloud maps and ground-based cloud maps, which can be used to predict solar radiation in the ultra-short-term or even in real-time by determining the distribution and movement trends of sky clouds. Due to time and spatial resolution constraints, the forecast technique based on satellite cloud maps is more accurate for the overall prediction of large areas within a few hours, but the local prediction error may be significant [5]. In contrast, ground-based cloud mapbased prediction techniques are more suited for predicting photovoltaic power over the next 0-4 h. In the ground-based cloud map prediction method, in order to improve the prediction accuracy of the model by taking into account the attenuation of cloud motion to solar radiation [6].…”