Jian Wang

Cloud cover experiences rapid fluctuations, significantly impacting the irradiance reaching the ground and causing frequent variations in photovoltaic power output. Accurate detection of thin and fragmented clouds is crucial for reliable photovoltaic power generation forecasting. In this paper, we introduce a novel cloud detection method, termed Adaptive Laplacian Coordination Enhanced Cross-Feature U-Net (ALCU-Net). This method augments the traditional U-Net architecture with three innovative components: an Adaptive Feature Coordination (AFC) module, an Adaptive Laplacian Cross-Feature U-Net with a Multi-Grained Laplacian-Enhanced (MLE) feature module, and a Criss-Cross Feature Fused Detection (CCFE) module. The AFC module enhances spatial coherence and bridges semantic gaps across multi-channel images. The Adaptive Laplacian Cross-Feature U-Net integrates features from adjacent hierarchical levels, using the MLE module to refine cloud characteristics and edge details over time. The CCFE module, embedded in the U-Net decoder, leverages criss-cross features to improve detection accuracy. Experimental evaluations show that ALCU-Net consistently outperforms existing cloud detection methods, demonstrating superior accuracy in identifying both thick and thin clouds and in mapping fragmented cloud patches across various environments, including oceans, polar regions, and complex ocean-land mixtures.

As the persistent proliferation of renewable energy sources (RESs) will have a significant impact on the prediction of power system inertia, a statistically accurate prediction model for the system inertia, like the traditional prediction-then-optimize (PTO) model, might not necessarily lead to higher economics and security in the day-ahead scheduling. This paper proposes a predict-and-optimize (PAO) model for the day-ahead inertia prediction using distributionally robust unit commitment (DRUC). First, the probabilistic prediction of day-ahead inertia is obtained by the kernel density estimation (KDE), and the confidence set is used for the uncertainty quantization of DRUC. Then, the inertia prediction method is embedded into DRUC, combining with the relevant feature and the closed-loop feedback of DRUC cost increment. The cost-based day-ahead inertia prediction method is trained, in which the prediction quality is evaluated by inducing the DRUC cost increment rather than the statistical prediction error. Furthermore, the proposed PAO is trained by rolling updates to efficiently capture the time series characteristics of inertia, reducing the computational burden and enhancing the solution quality of DRUC. The case studies are carried out on the Finnish power system with a public dataset. The results show that the proposed PAO model offers a better prediction accuracy and incurs smaller re-dispatch cost increments as compared with those of the traditional PTO model, providing a higher security and economically efficient tool for transmission system operators (TSOs) in the day-ahead UC scheduling.

Machine learning algorithms are commonly used for quickly and efficiently counting people from a crowd. Test-time adaptation methods for crowd counting adjust model parameters and employ additional data augmentation to better adapt the model to the specific conditions encountered during testing. The majority of current studies concentrate on unsupervised domain adaptation. These approaches commonly perform hundreds of epochs of training iterations, requiring a sizable number of unannotated data of every new target domain apart from annotated data of the source domain. Unlike these methods, we propose a meta-test-time adaptive crowd counting approach called CrowdTTA, which integrates the concept of test-time adaptation into the meta-learning framework and makes it easier for the counting model to adapt to the unknown test distributions. To facilitate the reliable supervision signal at the pixel level, we introduce uncertainty by inserting the dropout layer into the counting model. The uncertainty is then used to generate valuable pseudo labels, serving as effective supervisory signals for adapting the model. In the context of meta-learning, one image can be regarded as one task for crowd counting. In each iteration, our approach is a dual-level optimization process. In the inner update, we employ a self-supervised consistency loss function to optimize the model so as to simulate the parameters update process that occurs during the test phase. In the outer update, we authentically update the parameters based on the image with ground truth, improving the model’s performance and making the pseudo labels more accurate in the next iteration. At test time, the input image is used for adapting the model before testing the image. In comparison to various supervised learning and domain adaptation methods, our results via extensive experiments on diverse datasets showcase the general adaptive capability of our approach across datasets with varying crowd densities and scales.

In recent years, with the continuous growth of global new energy installed capacity, it is imperative to improve the prediction accuracy of new energy power generation. Based upon the current development trends of photovoltaic power generation and deep learning, this paper proposes a solar irradiance prediction model based on TCB-GRU-MLP. First, based on the Pearson and Spearman correlation analysis methods, the influencing factors that are strongly related to photovoltaic meteorological factors and irradiance in the study area were selected; then, the solar irradiance is predicted, and the results are displayed in the visualization interface. Experimental results show that the trained prediction model performs better than the TCN and Transformer models.