Grounding Time-Series Foundation Models in Digital Twin Topology for Predictive Maintenance
作者: Sizhe Ma, Katherine A. Flanigan, Mario Bergés
分类: eess.SP, cs.AI
发布日期: 2026-09-30
备注: Submitted to Reliability Engineering \& System Safety (RESS)
💡 一句话要点
提出拓扑信息融合方法以提升数字双胞胎中的时间序列预测性能
🎯 匹配领域: 支柱九:具身大模型 (Embodied Foundation Models)
关键词: 时间序列模型 数字双胞胎 预测性维护 拓扑信息 交叉注意力 剩余使用寿命预测 多变量架构
📋 核心要点
- 现有的时间序列基础模型在未见任务(如回归)上表现不足,缺乏系统性的实证比较。
- 提出了一种拓扑信息融合方法,通过资产结构的拓扑约束来显式塑造交叉注意力机制。
- 实验结果表明,拓扑约束的注意力机制在回归任务中优于无约束融合,且不同来源之间具有互补性。
📝 摘要(中文)
数字双胞胎越来越多地支持依赖时间序列数据的分析任务,促使人们对时间序列基础模型(TSFMs)的关注。然而,现有的TSFMs主要针对时间延续进行预训练,在未见任务(如回归)上表现不佳。本文的贡献包括:基准测试五种TSFMs在剩余使用寿命(RUL)预测中的表现,提出拓扑信息融合方法,以及通过消融研究验证不同来源的互补性,结果表明拓扑约束的注意力机制在回归任务中表现优于无约束融合,且在某些情况下超越了最先进的性能。
🔬 方法详解
问题定义:本文旨在解决时间序列基础模型在数字双胞胎上下游任务中的表现不足,尤其是在回归任务中的应用痛点。现有方法多依赖于时间延续预训练,缺乏对特定任务的适应性。
核心思路:论文提出了一种拓扑信息融合方法,利用数字双胞胎中存储的资产结构信息,通过拓扑约束来引导交叉注意力机制,从而提高模型对物理系统局部连接性的尊重。
技术框架:整体架构包括三个主要模块:首先是基于C-MAPSS数据集的TSFM基准测试;其次是拓扑信息的提取与融合;最后是消融研究以验证不同来源的互补性。
关键创新:最重要的创新点在于拓扑约束的交叉注意力机制,该机制显著区别于传统的无约束全连接交互,能够更好地捕捉物理系统的特性。
关键设计:在模型设计中,采用了冻结的TSFM骨干网络,并通过拓扑约束调整交叉注意力的计算方式,确保融合表示符合物理系统的局部连接性。
🖼️ 关键图片
📊 实验亮点
实验结果显示,采用拓扑约束的注意力机制的模型在C-MAPSS数据集上表现优于传统的无约束融合方法,尤其是在复杂操作条件下,提升幅度虽小但显著,证明了不同来源的互补性。
🎯 应用场景
该研究在工业设备的预测性维护领域具有广泛的应用潜力,能够帮助企业更准确地预测设备的剩余使用寿命,从而优化维护策略,降低停机时间和维护成本。未来,该方法还可以扩展到其他依赖时间序列数据的领域,如智能制造和物联网。
📄 摘要(原文)
Digital twins increasingly support downstream analytical tasks that depend on time-series data, motivating interest in time-series foundation models (TSFMs) as scalable backbones. However, TSFMs are primarily pretrained for temporal continuation and often underperform on unseen tasks such as regression, and systematic empirical comparisons against state-of-the-art dedicated models in digital twin contexts remain limited. This paper makes three contributions. First, we benchmark five well-known TSFMs with frozen backbones on remaining useful life (RUL) prediction using the C-MAPSS dataset, finding that multivariate architectures substantially outperform univariate ones, particularly under varying operating conditions. This raises a deeper question: when cross-channel dependencies can be modeled through pretrained weights, target-task adaptation, and digital twin-derived representations, how much does each contribute, and are they complementary? Second, we propose a topology-informed fusion approach in which topological constraints, derived from the asset structure the digital twin stores among its information models, explicitly shape cross-attention, so that fused representations respect the physical system's local connectivity rather than relying on unconstrained all-to-all interactions. Third, we conduct an ablation study across C-MAPSS subsets of varying operational complexity that isolates the three sources and their interactions. The sources prove complementary rather than redundant, and topology-constrained attention outperforms unconstrained fusion, though by a small margin, enabling a frozen TSFM informed by digital twin representations to remain competitive or in some cases exceed state-of-the-art performance on this regression task.