Frequency-Aware Spatiotemporal Graph Modeling of Multi-Pollutant Dynamics in Industrial Air Quality SystemsOA
Industrial air quality forecasting remains challenging due to nonlinear pollutant formation,localized emissions,meteorological variability,and nonstationary spatiotemporal dependencies among monitoring stations.This study proposes FFTGNet,a frequency-aware spatiotemporal graph neural network for multi-pollutant forecasting in industrial air quality systems.It integrates an FFT-guided dominant-period estimation and period-folding module with a temporal-to-spatial graph backbone composed of TemporalGLU and Chebyshev graph convolution.The frequency-guided module reorganizes input sequences into intra-period and inter-period representations,Temporal-GLU adaptively filters nonlinear temporal fluctuations and short-term spikes,and ChebGCN propagates information across inter-station spatial dependencies.Experiments were conducted using five years of hourly observations fromten Photochemical Assessment Monitoring Stations in Yunlin,Taiwan,covering O_(3),CO,PM10,meteorological variables,and PMF-derived VOC source features.Compared with conventional baselines and recent spatiotemporal models,includingMGCGRU-SANand AirNN-Adapted,the proposedmethod achieves themost consistent performance across pollutants and forecasting horizons.For 1 h forecasting,it obtains R^(2) values of 0.9509,0.8574,and 0.8556 forO_(3),CO,and PM10,respectively.For O_(3),it further maintains R^(2) values of 0.8116 and 0.6797 at the 3 and 6 h horizons,while showing smooth degradation rather than abrupt performance collapse up to 24 h.Ablation results confirm the complementary contributions of frequency-guided period folding,gated temporal modeling,and graph-based spatial propagation.Computational analysis further indicates that FFTGNet remains feasible for server-side near-real-time deployment,with an inference latency of 71.7ms per sample.These results demonstrate that coupling frequency-domain priors with spatiotemporal graph learning improves forecasting accuracy,stability,and applicability for industrial multi-pollutant air quality systems.
Chia-Hui Liu;Chen-Chuan Cheng
Department of Electronic Engineering,National Formosa University,Yunlin,TaiwanDepartment of Electronic Engineering,National Formosa University,Yunlin,Taiwan
资源环境
Frequency-aware learningspatiotemporal graphmodelsmulti-pollutantindustrial air quality systemsfast fourier transform
《Computer Modeling in Engineering & Sciences》 2026 (8)
P.1216-1241,26
评论