All Issue

2025 Vol.3, Issue 2 Preview Page

Research Article

Rapid Wildfire Spread Prediction Model Based on Probabilistic Conditional GAN

확률론적 조건부 GAN 기반 신속 산불 확산 예측 모델

Taehoon Kang1*
,
Taeyong Kim2
강 태훈1*
,
김 태용2
1Undergraduate Student, Department of Civil Systems Engineering, Ajou University, Suwon-si 16499, Korea
2Assistant Professor, Department of Civil Systems Engineering, Ajou University, Suwon-si 16499, Korea
1아주대학교 건설시스템공학과 학사과정
2아주대학교 건설시스템공학과 조교수
*Corresponding Author
31 December 2025. pp. 170-181
PDF XML Citation
Abstract

To address the increasing scale of wildfires due to recent climate change, this study proposes a rapid and probabilistic wildfire spread prediction model using a conditional Generative Adversarial Network. The proposed model is trained on a large-scale dataset constructed via FARSITE simulations, securing both the precision of physics-based models and the computational speed of deep learning. Experimental results demonstrate that the model addresses the ambiguity of spread boundaries and under-estimation issues found in existing Autoencoder-based models by generating sharp fire shapes, while achieving a breakthrough in prediction speed compared to physics-based simulations. Furthermore, by introducing an ensemble technique to quantify uncertainty and providing various spread scenarios, it is confirmed that the model can effectively support the decision-making of field commanders. This study suggests that generative Artificial Intelligence (AI) technology can be utilized as a key component of real-time wildfire disaster prevention systems.

Keywords
wildfire spread prediction
conditional GAN
FARSITE
probabilistic prediction
ensemble analysis

최근 기후변화로 인한 산불의 대형화에 대응하기 위해, 본 연구는 조건부 생성적 적대 신경망을 활용한 신속하고 확률론적인 산불 확산 예측 모델을 제안한다. 제안된 모델은 FARSITE 시뮬레이션을 통해 구축된 대규모 데이터셋을 학습하여 물리 기반 모델의 정밀함과 딥러닝의 연산 속도를 동시에 확보하였다. 실험 결과 본 모델은 기존 오토인코더 기반 모델에서 나타나는 확산 경계의 모호성 및 과소예측 문제를 극복하고 선명한 화재 형상을 생성하였으며 물리 기반 시뮬레이션 대비 획기적인 예측 속도를 달성하였다. 또한 앙상블 기법을 도입하여 불확실성을 정량화하고 다양한 확산 시나리오를 제공함으로써 현장 지휘관의 의사결정을 효과적으로 지원할 수 있음을 확인하였다. 본 연구는 생성형 AI 기술이 실시간 산불 방재 시스템의 핵심 요소로 활용될 수 있음을 시사한다.

키워드
산불 확산 예측
조건부 생성적 적대 신경망
FARSITE
확률론적 예측
앙상블 분석
References
1

Airey-Lauvaux, C., Pierce, A. D., Skinner, C. N. and Taylor, A. H. (2022) Changes in fire behavior caused by fire exclusion and fuel build-up vary with topography in California montane forests, USA. Journal of Environmental Management 304, 114255.

10.1016/j.jenvman.2021.114255
2

Bank, D., Koenigstein, N. and Giryes, R. (2023) Autoencoders. Machine Learning for Data Science Handbook: Data Mining and Knowledge Discovery Handbook, 353-374.

10.1007/978-3-031-24628-9_16
3

CAL FIRE. (2021) 2020 Fire Siege. California Department of Forestry and Fire Protection.

4

Carvalho, R. B., Menor, I. O., Schmidt, I. B., Berlinck, C. N., Genes, L. and Dirzo, R. (2025) Brazil on fire: Igniting awareness of the 2024 wildfire crisis. Journal of Environmental Management 389, 126190.

10.1016/j.jenvman.2025.126190
5

Chen, G., Qiu, M., Wang, P., Zhang, Y., Shindell, D. and Zhang, H. (2024) Continuous wildfires threaten public and ecosystem health under climate change across continents. Frontiers of Environmental Science & Engineering, 18(10), 130.

10.1007/s11783-024-1890-6
6

Finney, M. A. (1998) FARSITE: Fire Area Simulator—model development and evaluation. U.S. Department of Agriculture, Forest Service, Rocky Mountain Research Station.

10.2737/RMRS-RP-4
7

Goodfellow, I. J., Pouget-Abadie, J., Mirza, M., Xu, B., Warde-Farley, D., Ozair, S., Courville, A. and Bengio, Y. (2014) Generative adversarial nets. Advances in Neural Information Processing Systems 27.

8

Isola, P., Zhu, J. Y., Zhou, T. and Efros, A. A. (2017) Image-to-image translation with conditional adversarial networks. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, 1125-1134.

10.1109/CVPR.2017.632
9

Jiang, W., Qiao, Y., Su, G., Li, X., Meng, Q., Wu, H., Song, G. and Wang, F. (2023) WFNet: A hierarchical convolutional neural network for wildfire spread prediction. Environmental Modelling & Software 170, 105841.

10.1016/j.envsoft.2023.105841
10

Kim, N. G., Yeom, C. H. and Kwon, C. G. (2024) Exploring Directions for Establishing Forest Fire Containment Lines through Forest Fire Spread Prediction Simulations-A Case Study of 2022 Uljin-Samcheok Wildfire. Crisisonomy 20(5), 73-84.

10.14251/crisisonomy.2024.20.5.73
11

Mirza, M., and Osindero, S. (2014) Conditional generative adversarial nets. arXiv preprint arXiv:1411.1784.

12

Rothermel, R. C. (1972) A mathematical model for predicting fire spread in wildland fuels. Intermountain Forest & Range Experiment Station, Forest Service, US Department of Agriculture.

13

Shadrin, D., Illarionova, S., Gubanov, F., Evteeva, K., Mironenko, M., Levchunets, I., ... and Burnaev, E. (2024) Wildfire spreading prediction using multimodal data and deep neural network approach. Scientific Reports 14(1), 2606.

10.1038/s41598-024-52821-x38297034PMC10831103
14

Sullivan, A. L. (2009) Wildland surface fire spread modelling, 1990–2007. 1: Physical and quasi-physical models. International Journal of Wildland Fire 18(4), 349-368.

10.1071/WF06143
15

Thedy, J., Liao, K. W. and Kim, T. (2025) Auto regressive neural network-driven reliability optimization in base-isolated building design. Results in Engineering 26, 104713.

10.1016/j.rineng.2025.104713
16

Wang, Q., Ihme, M., Gazen, C., Chen, Y. F. and Anderson, J. (2024) A high-fidelity ensemble simulation framework for interrogating wildland-fire behaviour and benchmarking machine learning models. International Journal of Wildland Fire 33(12).

10.1071/WF24097
17

Zhu, Z. and Woodcock, C. E. (2014) Automated cloud, cloud shadow, and snow detection in multitemporal Landsat data: An algorithm designed specifically for monitoring land cover change. Remote Sensing of Environment 152, 217-234.

10.1016/j.rse.2014.06.012
Information
  • Publisher :Korean Institute of Bridge and Structural Engineers
  • Publisher(Ko) :한국교량및구조공학회
  • Journal Title :Journal of Structure Research and Practice
  • Journal Title(Ko) :한국교량및구조공학회 논문집
  • Volume : 3
  • No :2
  • Pages :170-181
  • Received Date : 2025-11-28
  • Revised Date : 2025-12-05
  • Accepted Date : 2025-12-09
  • DOI :https://doi.org/10.22725/JSRP.2025.3.2.170
Journal Informaiton Journal of Structure Research and Practice Journal of Structure Research and Practice
Online Submission & Review System
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • KCI 문헌 유사도 검사
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close