MVG-KAN: Multi-View Geo-Wind Guided KAN for PM$_{2.5}$ Forecasting

2026-06-23

AI Summary

Existing approaches struggle to effectively integrate heterogeneous factors—such as human activities, meteorological patterns, and wind-driven cross-regional transport—to accurately model PM₂.₅ dynamics. This work proposes a tri-perspective modeling framework that decouples local periodic trends from station-specific nonlinear residual dynamics via a periodic-residual decomposition. It further introduces a Geo-Wind directed graph as a spatial prior, encoding geographical decay alongside wind direction and speed to characterize pollutant dispersion pathways. The de-seasonalized residual sequences are then modeled using a spatiotemporal Kolmogorov–Arnold network (TKAN), enhanced by joint learning across multiple pollutants to improve representation capacity. The method substantially advances short-term PM₂.₅ forecasting accuracy, particularly excelling in capturing wind-guided inter-station transport and local residual inertia.

Abstract

Accurate short-term PM$_{2.5}$ forecasting is important for public health protection, air-quality early warning, and urban environmental management. However, PM$_{2.5}$ variation is driven by multiple coupled factors, including stable periodic changes induced by human activities and meteorological regularity, station-specific short-term concentration evolution, and meteorology-driven pollutant dispersion among monitoring stations. Existing spatio-temporal forecasting methods may capture station relationships to some extent, but distance-only, correlation-based, or purely adaptive graphs are often insufficient to comprehensively represent these heterogeneous factors, especially wind-direction-dependent pollutant transport. To address this problem, we propose a Multi-View Geo-Wind Guided KAN model for PM$_{2.5}$ forecasting, named \textbf{MVG-KAN}, which models station-level PM$_{2.5}$ evolution from three complementary views: local periodic regularity, station-wise residual temporal dynamics, and meteorological-environment-guided spatial dispersion. Specifically, the periodic-residual forecasting backbone first separates stable daily and weekly patterns from non-periodic residual variations. A Geo-Wind Graph is constructed by combining geographic distance decay with wind-direction- and wind-speed-aware transport, providing a lightweight physically motivated directed spatial prior for residual propagation among stations. In addition, a temporal Kolmogorov-Arnold network (TKAN) residual head is then introduced to learn station-wise nonlinear autoregressive correction from de-periodized PM$_{2.5}$ residuals and historical multi-pollutant sequences, thereby enhancing the modeling of local residual inertia and pollutant co-variation.

Problem

• PM2.5 forecasting
• spatio-temporal modeling
• wind-driven dispersion
• heterogeneous factors
• air quality prediction

Innovation

• Geo-Wind Graph
• Kolmogorov-Arnold Network
• PM2.5 forecasting
• spatio-temporal modeling
• multi-view decomposition