空氣污染對人體健康與環境品質具有深遠的影響，尤其是細懸浮微粒 (PM2.5) 與二氧化氮 (NO2) 等污染物，已被多項研究證實與呼吸道與心血管疾病高度相關。而台中火力發電廠為全台最大之燃煤電廠，其排放常被視為中部地區空氣污染的潛在重要來源。為釐清中火對區域空污的影響，並建立具預測力與解釋力之時空推估模型，本研究結合機器學習與地理資訊技術，發展出一套以 Geo-AI 為核心的空氣污染預測模型。
本研究以台灣中部空品區為研究範圍，整合空氣品質監測資料、氣象因子、土地利用與交通等多源資料，建構 PM2.5 與 NO2 濃度之時空推估模型。模型訓練採用多種機器學習演算法，並以集成學習方法結合各模型預測結果，同時透過 SHAP 值分析污染來源與重要變數影響，以填補現有研究在污染來源辨識與空間擴散特徵分析上的缺口。根據模型表現，PM2.5 與 NO2 主模型之決定係數 (R2) 分別達 0.85 與 0.84，顯示模型之預測效能良好；同時，中火變數亦在模型中被辨識為影響 PM2.5 及 NO2 的重要因子之一，其 SHAP 值在兩者污染物之主模型中佔比分別為 5.21 % 與 2.06 %。
本研究所建構之 Geo-AI 模型能有效掌握污染物在時間與空間上的變異性，並產出高解析度之濃度分布結果，補足監測站分布不足的限制。未來若結合即時資料，模型可延伸應用於操作型預測與空品預警系統，並作為污染熱區辨識、政策規劃與發電廠污染監控之重要參考依據。

Air pollution has significant impacts on human health and environmental quality, particularly fine particulate matter (PM2.5) and nitrogen dioxide (NO2), which are strongly associated with respiratory and cardiovascular diseases. This study developed a Geo-AI based spatiotemporal prediction model to evaluate air pollution in central Taiwan and assess the potential contribution of the Taichung Thermal Power Plant, the largest coal-fired power plant in Taiwan. Multiple data sources, including air quality monitoring, meteorology, land use, and traffic, were integrated to construct PM2.5 and NO2 prediction models using various machine learning algorithms and ensemble learning. SHAP value analysis identified the power plant as an important factor, contributing 5.21% to PM2.5 and 2.06% to NO2 in the main models (R2 = 0.85 and 0.84, respectively). The proposed model captures spatiotemporal variability with high-resolution outputs, providing valuable insights for pollution hotspot identification, policy development, and future real-time air quality forecasting.

Paraschiv, S., & Paraschiv, L. S. (2019). Analysis of traffic and industrial source contributions to ambient air pollution with nitrogen dioxide in two urban areas in Romania. Energy Procedia, 157, 1553-1560.
Fajersztajn, L., Saldiva, P., Pereira, L. A. A., Leite, V. F., & Buehler, A. M. (2017). Short-term effects of fine particulate matter pollution on daily health events in Latin America: a systematic review and meta-analysis. International journal of public health, 62, 729-738.
Achilleos, S., Kioumourtzoglou, M. A., Wu, C. D., Schwartz, J. D., Koutrakis, P., & Papatheodorou, S. I. (2017). Acute effects of fine particulate matter constituents on mortality: A systematic review and meta-regression analysis. Environment international, 109, 89-100.
Zhang, Z., Yan, W., Chen, Q., Zhou, N., & Xu, Y. (2019). The relationship between exposure to particulate matter and breast cancer incidence and mortality: a meta-analysis. Medicine, 98(50), e18349.
Zhao, L., Liang, H. R., Chen, F. Y., Chen, Z., Guan, W. J., & Li, J. H. (2017). Association between air pollution and cardiovascular mortality in China: a systematic review and meta-analysis. Oncotarget, 8(39), 66438.
Faustini, A., Rapp, R., & Forastiere, F. (2014). Nitrogen dioxide and mortality: review and meta-analysis of long-term studies. European Respiratory Journal, 44(3), 744-753.
Chen, X., Liu, Q., Sheng, T., Li, F., Xu, Z., Han, D., ... & Cheng, J. (2019). A high temporal-spatial emission inventory and updated emission factors for coal-fired power plants in Shanghai, China. Science of the Total Environment, 688, 94-102.
Hannun, R. M., & Razzaq, A. H. A. (2022, March). Air pollution resulted from coal, oil and gas firing in thermal power plants and treatment: a review. In IOP conference series: earth and environmental science (Vol. 1002, No. 1, p. 012008). IOP Publishing.
Artun, G. K., Polat, N., Yay, O. D., Üzmez, Ö. Ö., Arı, A., Tuygun, G. T., ... & Gaga, E. O. (2017). An integrative approach for determination of air pollution and its health effects in a coal fired power plant area by passive sampling. Atmospheric Environment, 150, 331-345.
Tsai, J. H., Chen, S. H., Chen, S. F., & Chiang, H. L. (2021). Air pollutant emission abatement of the fossil-fuel power plants by multiple control strategies in Taiwan. Energies, 14(18), 5716.
Adappa, S., Tiwari, R. R., Kamath, R., & Guddattu, V. (2017). Health effects and environmental issues in residents around coal-fired thermal power plant, Padubidri: a cross-sectional study. J Environ Occup Sci, 6(1), 8-11.
Vig, N., Ravindra, K., & Mor, S. (2023). Environmental impacts of Indian coal thermal power plants and associated human health risk to the nearby residential communities: A potential review. Chemosphere, 341, 140103.
Kamath, R., Udayar, S. E., Jagadish, G., Prabhakaran, P., & Madhipatla, K. K. (2022). Assessment of health status and impact of pollution from thermal power plant on health of population and environment around the plant in Udupi District, Karnataka. Indian Journal of Public Health, 66(2), 91-97.
Wu, Y. S., Tsai, C. Y., Chang, K. H., & Chiang, C. F. (2021). Impact of air pollutants emitted by taichung power plant on atmospheric PM2.5 in Central Taiwan. Aerosol and Air Quality Research, 21(4), 200358.
Tsai, C. Y., Chen, T. F., & Chang, K. H. (2023). Role of an Ultra-Large Coal-Fired Power Plant in PM2. 5 Pollution in Taiwan. Atmosphere, 15(1), 56.
Briggs, D. J., Collins, S., Elliott, P., Fischer, P., Kingham, S., Lebret, E., ... & Van Der Veen, A. (1997). Mapping urban air pollution using GIS: a regression-based approach. International Journal of Geographical Information Science, 11(7), 699-718.
Liu, C., Henderson, B. H., Wang, D., Yang, X., & Peng, Z. R. (2016). A land use regression application into assessing spatial variation of intra-urban fine particulate matter (PM2. 5) and nitrogen dioxide (NO2) concentrations in City of Shanghai, China. Science of the Total Environment, 565, 607-615.
Wu, C. D., Chen, Y. C., Pan, W. C., Zeng, Y. T., Chen, M. J., Guo, Y. L., & Lung, S. C. C. (2017). Land-use regression with long-term satellite-based greenness index and culture-specific sources to model PM2. 5 spatial-temporal variability. Environmental pollution, 224, 148-157.
Naughton, O., Donnelly, A., Nolan, P., Pilla, F., Misstear, B. D., & Broderick, B. (2018). A land use regression model for explaining spatial variation in air pollution levels using a wind sector based approach. Science of the Total Environment, 630, 1324-1334.
Wong, P. Y., Su, H. J., Lee, H. Y., Chen, Y. C., Hsiao, Y. P., Huang, J. W., ... & Spengler, J. D. (2021). Using land-use machine learning models to estimate daily NO2 concentration variations in Taiwan. Journal of Cleaner Production, 317, 128411.
Wong, P. Y., Lee, H. Y., Chen, Y. C., Zeng, Y. T., Chern, Y. R., Chen, N. T., ... & Wu, C. D. (2021). Using a land use regression model with machine learning to estimate ground level PM2. 5. Environmental Pollution, 277, 116846.
Wong, P. Y., Su, H. J., Lung, S. C. C., & Wu, C. D. (2023). An ensemble mixed spatial model in estimating long-term and diurnal variations of PM2. 5 in Taiwan. Science of The Total Environment, 866, 161336.