用神经网络动态生成火势参数，提升野火蔓延预测精度
Neural-Parameterized Cellular Automata for Wildfire Spread
- 用多尺度卷积神经网络实时生成影响火势的参数
- 72小时预测中交并比超0.6，10天数据拟合后效果稳定
- 兼顾物理可解释性与深度学习能力，适合灾害预警研究
传统野火模型依赖固定参数和静态燃料图，常低估火势蔓延。为此，我们提出一种基于JAX的混合深度学习概率细胞自动机框架。通过多尺度卷积神经网络，动态生成影响火势传播概率、风向对齐和坡度影响的空间变化参数。该设计在保留三态细胞自动机物理可解释性的同时，捕捉复杂非线性环境交互。JAX实现支持硬件加速与梯度参数校准。在美西六次大规模野火上评估，经10天数据同化窗口增量拟合后，72小时预测交并比（IoU）仍高于0.6；结果为已有扑救措施下的火势条件性预测。
原文摘要 · Abstract (English)
Traditional wildfire models rely on rigid, low-dimensional parameters and static fuel maps, frequently underpredicting fire spread. To address this weakness, we introduce a hybrid deep-learning parameterized Probabilistic Cellular Automata (CA) framework implemented in JAX. Our approach employs a Multi-Scale Convolutional Neural Network to dynamically generate spatially varying parameters that govern fire-spread probability, wind alignment, and slope influence. This hybrid design captures complex, nonlinear environmental interactions while preserving the physical interpretability of the underlying three-state CA. The JAX implementation enables hardware acceleration and gradient-based parameter calibration. Evaluated on six large-scale wildfires in the western United States, the model maintains IoU > 0.6 over 72-hour forecast horizons after a 10-day data assimilation window during which the model is fitted incrementally to observed perimeters; the resulting forecast is a conditional projection of fire growth under the suppression regime already ncoded in those observations.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。