Day-Ahead Hourly Solar Irradiance Forecasting Based on Multi-Attributed Spatio-Temporal Graph Convolutional Network

Abstract

Solar irradiance forecasting is fundamental and essential for commercializing solar energy generation by overcoming output variability. Accurate forecasting depends on historical solar irradiance data, correlations between various meteorological variables (e.g., wind speed, humidity, and cloudiness), and influences between the weather contexts of spatially adjacent regions. However, existing studies have been limited to spatiotemporal analysis of a few variables, which have clear correlations with solar irradiance (e.g., sunshine duration), and do not attempt to establish atmospheric contextual information from a variety of meteorological variables. Therefore, this study proposes a novel solar irradiance forecasting model that represents atmospheric parameters observed from multiple stations as an attributed dynamic network and analyzes temporal changes in the network by extending existing spatio-temporal graph convolutional network (ST-GCN) models. By comparing the proposed model with existing models, we also investigated the contributions of (i) the spatial adjacency of the stations, (ii) temporal changes in the meteorological variables, and (iii) the variety of variables to the forecasting performance. We evaluated the performance of the proposed and existing models by predicting the hourly solar irradiance at observation stations in the Korean Peninsula. The experimental results showed that the three features are synergistic and have correlations that are difficult to establish using single-aspect analysis.

A Combined Persistence and Physical Approach for Ultra-Short-Term Photovoltaic Power Forecasting Using Distributed Sensors

“…Advancements in forecasting models, including deep learning approaches and graph convolutional networks (GCNs), further refine prediction capabilities by capturing temporal and spatiotemporal patterns within the generated data [20]. For accurate medium-term PV power forecasts, utilizing high-quality data sources and continuously refining forecasting models is essential [8,21].…”

An Improved Method for Photovoltaic Forecasting Model Training Based on Similarity

“…Mainly, historical data are classified or solar irradiance is analyzed to narrow down the influence of weather on data performance [31][32][33][34][35][36][37][38]. Alternatively, the spatio-temporal characteristics of data can be analyzed and extracted [39][40][41]. Iraklis C amplified seasonal data to enhance possible instability of renewable energy production, then predicted the day-ahead flexibility provided by VPP in distributed energy systems [42].…”

Transitivity-Preserving Graph Representation Learning for Bridging Local Connectivity and Role-Based Similarity

“…The Thirty-Eighth AAAI Conference on Artificial Intelligence gation ranges in a layer than GNNs, neighbourhood aggregation has little ability to reveal long-range dependencies or role-based similarities between nodes (Lee and Jung 2020a;Jeon, Choi, and Lee 2022). Third, the existing models lack the ability to integrate local and global structural features into a unified vector representation, which could then benefit various downstream tasks.…”