[论文解读] A multi-fidelity deep operator network (DeepONet) for fusing simulation and monitoring data: Application to real-time settlement prediction during tunnel construction
本文提出了一种多保真度DeepONet框架,通过融合低保真度的有限元仿真数据与高保真度、稀疏且含噪的监测数据,实现实时、全场的地基沉降预测,应用于隧道掘进机(TBM)掘进过程中。通过利用因果感知预处理与迁移学习,将物理信息仿真与现场测量数据相结合,即使在有限且含噪的现场观测条件下,模型仍能实现高精度预测(R² > 0.9,误差数据占比30%;>0.8,误差数据占比50%),并实现每步掘进小于一分钟的预测速度。
Ground settlement prediction during the process of mechanized tunneling is of paramount importance and remains a challenging research topic. Typically, two paradigms are existing: a physics-driven approach utilizing process-oriented computational simulation models for the tunnel-soil interaction and the settlement prediction, and a data-driven approach employing machine learning techniques to establish mappings between influencing factors and the ground settlement. To integrate the advantages of both approaches and to assimilate the data from different sources, we propose a multi-fidelity deep operator network (DeepONet) framework, leveraging the recently developed operator learning methods. The presented framework comprises of two components: a low-fidelity subnet that captures the fundamental ground settlement patterns obtained from finite element simulations, and a high-fidelity subnet that learns the nonlinear correlation between numerical models and real engineering monitoring data. A pre-processing strategy for causality is adopted to consider the spatio-temporal characteristics of the settlement during tunnel excavation. Transfer learning is utilized to reduce the training cost for the low-fidelity subnet. The results show that the proposed method can effectively capture the physical information provided by the numerical simulations and accurately fit measured data as well. Remarkably, even with very limited noisy monitoring data, the proposed model can achieve rapid, accurate, and robust predictions of the full-field ground settlement in real-time during mechanized tunnel excavation.
研究动机与目标
- 解决机械隧道掘进过程中实时、全场地基沉降预测的挑战,其中监测数据稀疏且含噪。
- 将基于物理的仿真数据(低保真度)与真实世界监测数据(高保真度)相结合,以提升预测精度。
- 开发一种数据高效、具备实时预测能力的模型,通过提前预测隧道面前方的沉降,支持动态隧道掘进控制。
- 通过迁移学习与因果感知数据处理,降低训练成本并提升泛化能力。
提出的方法
- 该框架采用DeepONet架构,包含两个并行分支:一个在有限元仿真数据上训练的低保真度子网络,以及一个在真实监测数据上训练的高保真度子网络。
- 应用因果感知预处理,对齐时空输入与输出,确保模型尊重隧道掘进引起的沉降的时间演进规律。
- 采用迁移学习,利用仿真数据的预训练权重初始化低保真度子网络,显著降低训练时间与计算成本。
- 通过学习输入参数(如掌子面支护力与注浆压力)与全场沉降输出之间的非线性映射,融合两个子网络的预测结果。
- 采用多保真度损失函数,结合仿真与监测数据指导训练,增强物理一致性与数据保真度。
- 架构经过优化,推理速度极快,实现每步掘进的预测时间低于一分钟。
实验结果
研究问题
- RQ1多保真度DeepONet能否有效融合低保真度仿真数据与高保真度监测数据,从而提升隧道掘进过程中的实时沉降预测性能?
- RQ2所提出的因果感知预处理策略在多大程度上提升了模型捕捉时变沉降动态的能力?
- RQ3在数据稀缺环境下,迁移学习在多大程度上可降低训练成本,同时保持预测精度?
- RQ4当监测数据误差超过30%时,模型对噪声和数据稀疏性的鲁棒性如何?
- RQ5模型实现稳定且准确的全场沉降场重建,所需的最少时间步数是多少?
主要发现
- 当使用仅32个时间步的监测数据(误差水平30%)进行训练时,模型的R²得分超过0.9。
- 即使在50%误差水平下,模型仍能保持R²得分高于0.8,表明对噪声测量具有强大鲁棒性。
- 使用“低估”数据(k > 1)可获得比高估或随机扰动数据更高的预测精度,可能是因为绝对值更大,从而降低了相对噪声的影响。
- 模型每步掘进的实时预测时间少于一分钟,具备在隧道掘进作业中实际部署的可行性。
- 迁移学习显著降低了训练成本,使得每次新掘进步均可高效微调,而无需从头开始训练。
- 该框架成功捕捉了仿真数据中的物理趋势以及稀疏监测数据中的实际场行为,实现了高保真度的场重建。
更好的研究,从现在开始
从阅读论文到最终审阅,大幅缩短您的研究时间。
无需绑定信用卡
本解读由 AI 生成,并经人工编辑审核。