Yuhao Nie

Sky image-based solar forecasting using deep learning has been recognized as a promising approach in reducing the uncertainty of solar power generation. However, a major challenge is the lack of large quantity of sky image data encompassing diverse sky conditions for model training. This study presents a comprehensive survey of open-source sky image datasets for solar forecasting and related research areas, including cloud segmentation, classification and motion prediction which could potentially enhance solar forecasting capabilities. In total, 72 open-source sky image datasets are identified globally that satisfy the needs of deep learning-based method development. A database containing information about various aspects of the identified datasets is constructed. A multi-criteria ranking system is further developed to evaluate each dataset based on eight dimensions which could have important impacts on the data usage. Finally, insights on the applications of these datasets are provided. This study streamlines the processes of identifying and selecting sky image datasets, and could potentially accelerate the method development and benchmark in solar forecasting and related fields including energy meteorology and atmospheric science.

In recent years, deep learning-based solar forecasting using all-sky images has emerged as a promising approach for alleviating uncertainty in PV power generation. However, the stochastic nature of cloud movement remains a major challenge for accurate and reliable solar forecasting. With the recent advances in generative artificial intelligence, the synthesis of visually plausible yet diversified sky videos has potential for aiding in forecasts. In this study, we introduce \emphSkyGPT, a physics-informed stochastic video prediction model that is able to generate multiple possible future images of the sky with diverse cloud motion patterns, by using past sky image sequences as input. Extensive experiments and comparison with benchmark video prediction models demonstrate the effectiveness of the proposed model in capturing cloud dynamics and generating future sky images with high realism and diversity. Furthermore, we feed the generated future sky images from the video prediction models for 15-minute-ahead probabilistic solar forecasting for a 30-kW roof-top PV system, and compare it with an end-to-end deep learning baseline model SUNSET and a smart persistence model. Better PV output prediction reliability and sharpness is observed by using the predicted sky images generated with SkyGPT compared with other benchmark models, achieving a continuous ranked probability score (CRPS) of 2.81 (13% better than SUNSET and 23% better than smart persistence) and a Winkler score of 26.70 for the test set. Although an arbitrary number of futures can be generated from a historical sky image sequence, the results suggest that 10 future scenarios is a good choice that balances probabilistic solar forecasting performance and computational cost.

Photovoltaics (PV), the primary use of solar energy, is growing rapidly. However, the variable output of PV under changing weather conditions may hinder the large-scale deployment of PV. In this study, we propose a two-stage classification-prediction framework to predict contemporaneous PV power output from sky images (a so-called “nowcast”), and compare it with an end-to-end convolution neural network (CNN). The proposed framework first classifies input images into different sky conditions and then the classified images are sent to specific sub-models for PV output prediction. Two types of classifiers are developed and compared: (1) a CNN-based classifier trained on clear sky index (CSI)-labeled sky images and (2) a physics-based non-parametric classifier based on a threshold of fractional cloudiness of sky images. Different numbers of classification categories are also examined. The results suggest that the cloudiness-based classifier is more suitable than the CSI-based classifier for the framework, and the 3-class classification (i.e., sunny, cloudy, overcast) is found to be the optimal choice. We then fine-tune the cloudiness threshold for the non-parametric classifier and tailor the architecture for each sky-condition-specific sub-model. Under the best design, the proposed framework can achieve a root mean squared error (RMSE) of 2.20 kW (relative to a 30 kW rated PV array) on the test set comprising 18 complete days (9 sunny, RMSE = 0.69 kW; 9 cloudy, RMSE = 3.06 kW). Compared with the end-to-end CNN baseline model, the overall prediction performance can be improved by 6% (7% in sunny and 6% in cloudy), with 6% fewer trainable parameters needed in the architecture.

Liquefied natural gas (LNG) is a promising alternative to coal to mitigate the greenhouse gas (GHG) and particulate emissions from power, industry, and district heating in China. While numerous existing life cycle assessment (LCA) studies estimate the GHG footprint of LNG, large variation exists in these results. Such variability could be caused by differing project designs, system boundaries, modeling methods and data sources. It is not clear which of these factors is the most important. Here, three research groups from Canada and the US performed independent LCAs of the same planned LNG supply chain from Canada to China. The teams applied different methods and assumptions but used aligned system boundaries and worked with a single upstream producer to obtain production data. The GHG emissions of Canadian LNG to China for power and heat generation were found to be 427–556 g CO2-eq/kWh and 81–92 g CO2-eq/MJ. Compared with Chinese coal for power generation, 291–687 g CO2-eq (34%–62%) reduction can be achieved per kWh of power generated. The central tendency in each study is aligned more closely than the overall uncertainty range: thus, uncertainty caused by fundamental data challenges likely outweighs variability caused by use of different LCA methods. Differences in assumptions and methods among the three teams lead to moderate variation at the stage level, but in better agreement at the life-cycle level, showing the existence of compensating variation. Given the robustness to very different LCA methods, existing literature variation may be explained by project-, location- and operator-dependent parameters.