Multi-Horizon AQI Forecasting Through the Heterogeneous Fusion of a Temporal Convolutional Network and a Graph-Conditioned Causal Transformer
Open Access
- 10 November 2026
- journal article
- research article
- Published by MDPI AG in Applied Sciences
- Vol. 16 (22) , 11119
Abstract
Recent station observations provide a useful basis for air quality index (AQI) forecasting, whereas regional measurements may supply additional context at longer forecast leads. In this study, the forecasts of an independently trained temporal convolutional network (TCN) and a graph-conditioned causal transformer (Graph-CT) are combined. Both use the same input window of 24 h to predict all the hourly leads from 1 to 24 h, and their corresponding outputs are averaged. Air-quality records from 26 stations in the Chengdu–Chongqing economic circle are reconstructed on a common hourly grid, with causal input filling and evaluation restricted to observed targets. Learning rates and training durations are selected on 2021 data, followed by retrospective evaluation on the previously inspected 2022 benchmark. Across five runs, equal fusion achieves a pooled root mean square error (RMSE) of 18.036 ± 0.103 AQI, compared with 19.491 ± 0.063 for TCN, 18.284 ± 0.140 for Graph-CT, 19.413 ± 0.174 for iTransformer, and 19.043 ± 0.455 for Graph WaveNet under the common task. The paired interval includes zero, leaving the advantage over a two-member Graph-CT ensemble unresolved. Validation calibration slightly decreases the RMSE but increases the mean absolute error (MAE), and removing the transformer branch improves the evaluated configuration. Forecast averaging is therefore useful on this benchmark, although the results favour neither equal coefficients as an optimum nor the full architecture over its simpler variants. The evaluation of untouched periods and regions remains necessary for assessing generalization.Keywords
Funding Information
- Science and Technology Research Program of Chongqing Municipal Education Commission (KJQN202502103)
This publication has 35 references indexed in Scilit:
- Interpretable long-horizon air pollution forecasting using a transformer-based frameworkExpert Systems with Applications, 2026
- A graph neural network and Transformer-based model for PM2.5 prediction through spatiotemporal correlationEnvironmental Modelling & Software, 2025
- Spatiotemporal PM2.5 forecasting via dynamic geographical Graph Neural NetworkEnvironmental Modelling & Software, 2025
- TKSTAGNet: A Top-K Spatio-Temporal Attention Gating Network for air pollution predictionExpert Systems with Applications, 2024
- A novel AQI forecasting method based on fusing temporal correlation forecasting with spatial correlation forecastingAtmospheric Pollution Research, 2023
- PM2.5 Forecast in Korea using the Long Short-Term Memory (LSTM) ModelAsia-Pacific Journal of Atmospheric Sciences, 2022
- Trend analysis and forecast of PM2.5 in Fuzhou, China using the ARIMA modelEcological Indicators, 2018
- The forecast combination puzzle: A simple theoretical explanationInternational Journal of Forecasting, 2016
- Another look at measures of forecast accuracyInternational Journal of Forecasting, 2006
- Combining forecasts: A review and annotated bibliographyInternational Journal of Forecasting, 1989