用神经算子加速油藏模拟，30步预测误差可控且快1万倍。
Sequential Physics-Constrained Neural Operator Forward Modeling for the $\textit{Norne}$ Reservoir System
- 基于傅里叶神经算子构建物理约束模型，实现多相流动态高效逼近。
- 3298天全程预测精度达油相R²>0.99、气相>0.90、压力≈0.80，水相持续提升。
- 训练仅需一小时，1000样本集合单卡运行不到一分钟，提速近万倍。
我们构建了一个用于三相黑油油藏动态的顺序代理建模综合数学与计算框架，重点使用傅里叶神经算子（FNO）及其物理信息变体（PINO），以挪威诺尔内基准油藏为应用对象，该油藏网格为46×112×22（共113,344个单元），生产历史覆盖30个时间步（3298天）。理论贡献涵盖四个相互关联的问题：（1）在乘积Sobolev空间设定下的泛函分析框架，包含隐式时间步映射的存在性与局部利普希茨估计；（2）协变量偏移量化，证明Wasserstein-2距离满足 $W_2 \ leq \varepsilon(L^n-1)/(L-1)$，当 $L>1$ 时出现指数级种群风险偏差；（3）物理约束下的谱稳定性，表明PINO训练中若 $λ_R \ geq λ^*_R$，则学习到的雅可比谱半径降至 $ρ_F + Cλ_R^{-1/2}$，从而保证一致时间滚动误差 $|δ_n| \ leq \varepsilon/(1-ρ)$；（4）K步时间反向传播梯度分析，推导出几何偏差衰减 $O(ρ^K)$，最优窗口 $K^ = O(\log(T/σ^2))$，以及Adam收敛率 $O(1/\sqrt{t}) + O(ρ^{K^*})$。实证验证全部理论预测：自回归PINO代理模型在3298天内保持油相R²>0.99、气相>0.90、压力≈0.80，水相持续提升，八张NVIDIA B200 GPU训练不足一小时。1000样本集成在单张B200 GPU上运行不到一分钟，相较OPM有限体积模拟器提速约10⁴倍。
原文摘要 · Abstract (English)
We develop a comprehensive mathematical and computational framework for sequential surrogate modeling of three-phase black-oil reservoir dynamics using neural operators, with particular emphasis on Fourier Neural Operators (FNO) and their physics-informed variant (PINO). The application focus is the Norne benchmark reservoir, defined on a heterogeneous $46\times112\times22$ grid ($N=113,344$ cells), with a production history spanning $T=30$ timesteps covering 3298 days. Our theoretical contributions are organized around four interlocking problems: (1) functional-analytic formulation in a product-Sobolev-space setting, including well-posedness of the implicit timestep map and sharp local Lipschitz estimates; (2) covariate shift quantification, proving that the Wasserstein-2 distance grows as $W_2 \leq \varepsilon(L^n-1)/(L-1)$, with exponential population-risk discrepancy for $L>1$; (3) physics-constrained spectral stability, showing PINO training with $λ_R \geq λ^*_R$ reduces the learned Jacobian spectral radius to $ρ_F + Cλ_R^{-1/2}$, yielding uniform-in-time rollout error $|δ_n| \leq \varepsilon/(1-ρ)$; and (4) $K$-step TBPTT gradient analysis, deriving geometric bias decay $O(ρ^K)$, optimal window $K^ = O(\log(T/σ^2))$, and Adam convergence $O(1/\sqrt{t}) + O(ρ^{K^*})$. Empirical validation confirms all theoretical predictions: autoregressive PINO surrogates sustain $R^2>0.99$ (oil), $R^2>0.90$ (gas), $R^2\approx 0.80$ (pressure), and monotonically improving $R^2$ (water) across the full 3298-day horizon, trained on eight NVIDIA B200 GPUs in under one hour. A 1000-member ensemble runs in under one minute on a single B200 GPU, giving a ${\sim}10^4\times$ wall-clock speedup over the OPM finite-volume simulator.
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。