用深度学习加速海洋电磁数据反演，精度高且抗噪强。
DualTCN: A Physics-Constrained Temporal Convolutional Network for 2 Time-Domain Marine CSEM Inversion
- 设计物理约束的TCN网络，直接回归地层参数并重建电导率剖面。
- 比传统方法快2万倍，噪声下仍保持高精度（R²达0.858）。
- 适合地质勘探领域，尤其对快速反演与不确定性评估有需求者。
DualTCN是首个用于时域海洋可控源电磁（MCSEM）瞬变数据反演的深度学习框架。该框架摆脱传统地下介质离散化，直接回归四个地球模型参数——σ₁、σ₂、d₁、d₂，并通过可微分的软台阶解码器重构电导率-深度剖面。优化后的架构（379K参数）采用时序卷积网络（TCN）编码器，搭配晚期时间分支与辅助海床深度头，相较基线模型降低25.3%损失，预测精度高（σ₂的R² = 0.898），在A100 GPU上每样本反演耗时仅3.5~ms。通过课程式振幅增强，对噪声具有高度鲁棒性，在±2%随机振幅误差下仍保持平均R² = 0.858，而无增强时仅为0.363。该框架能有效推广至三层结构（海水/高阻层/基底），准确解析基底电导率（R² ≈ 0.88），但薄层分辨率受限（R² ≈ 0.23）。在对比基准中，显著优于传统局部优化方法（如Levenberg-Marquardt和L-BFGS-B），平均R²达0.877，远超多起点基线的0.129–0.439，计算成本降低至21,000倍。此外，通过蒙特卡洛丢弃实现不确定性量化，σ₁校准良好（PICP90 = 0.944），但短偏移（200m）信号限制导致d₂覆盖不足（PICP90 = 0.572），可通过后处理温度缩放或分裂共形预测缓解。
原文摘要 · Abstract (English)
DualTCN is the first deep-learning framework for inverting time-domain marine controlled-source electromagnetic (MCSEM) transient data. Moving away from traditional subsurface discretization, the framework regresses four earth-model parameters -- $σ_1$, $σ_2$, $d_1$, $d_2$ -- and reconstructs conductivity-depth profiles using a differentiable soft-step decoder. The optimized architecture (379K parameters) features a Temporal Convolutional Network (TCN) encoder paired with a late-time branch and an auxiliary seafloor-depth head. This design achieves a 25.3\% loss reduction over baseline models, with high predictive accuracy ($R^2 = 0.898$ for $σ_2$) and an inversion speed of 3.5~ms per sample on an A100 GPU. The framework demonstrates high robustness to noise through curriculum-based amplitude augmentation, maintaining a mean $\bar{R}^2$ of 0.858 at $\pm2\%$ random amplitude error, compared to $0.363$ without augmentation. DualTCN generalizes effectively to three-layer extensions (seawater/resistive layer/basement), accurately resolving basement conductivity ($R^2 \approx 0.88$), though thin-layer resolution remains a physical limitation ($R^2 \approx 0.23$). In comparative benchmarks, DualTCN significantly outperforms traditional local optimization methods like Levenberg-Marquardt and L-BFGS-B, yielding a mean $\bar{R}^2 = 0.877$ versus 0.129-0.439 for multi-start baselines, while operating at up to 21,000$\times$ lower computational cost. Finally, the framework incorporates uncertainty quantification via Monte Carlo (MC) Dropout. While well-calibrated for $σ_1$ (PICP90 = 0.944), inherent signal limitations at short offsets (200m) lead to under-coverage for $d_2$ (PICP90 = 0.572), which can be mitigated through post-hoc temperature scaling or split conformal prediction.
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。