MGSB: Manifold Gated Signature Branch Pressure-Domain Baseline Architecture for Two-Phase Pipeline Flows Under Distributional Shift

2026-08-05

AI Summary

This study addresses the significant performance degradation of multiphase pipeline leak detection models under distribution shifts, such as the transition from bubbly to slug flow regimes. To enhance out-of-distribution generalization without relying on specific sensor configurations, the authors propose a novel operating-condition-aware architecture that explicitly incorporates condition modeling into leak detection for the first time. The approach integrates condition-conditioned feature fusion, TT-RoughPath trajectory encoding, and Mean-Teacher consistency regularization. Evaluated under a leave-one-condition-out protocol, the method achieves an F1 score of 0.930 in-distribution and 0.783 out-of-distribution, substantially outperforming CNN-LSTM and fully connected baselines.

Abstract

Leak detection models for multiphase pipelines often degrade when deployed under flow regimes that differ from training. Existing evaluations typically assess performance under in-distribution operating conditions, masking failures caused by regime transitions such as bubble-to-slug flow. We propose the Manifold Gated Signature Bias (MGSB), a regime-aware architecture combining regime-conditioned feature fusion, a TT-RoughPath encoder, and Mean-Teacher consistency regularization to improve robustness under distribution shift. Under leave-one-group-out evaluation, MGSB achieves a detection F1 of 0.930 and an OOD F1 of 0.783, substantially outperforming CNN-LSTM and fully connected baselines under severe feature corruption. Ablations show the proposed architecture, not the training procedure, is the primary contributor to OOD robustness, while Mahalanobis-distance analysis confirms the held-out conditions are genuinely out-of-distribution. These results show that explicit regime-aware modelling is a practical path toward robust, sensor-agnostic leak detection in industrial multiphase pipelines.

Problem

• multiphase pipelines
• leak detection
• distributional shift
• flow regime transition
• out-of-distribution

Innovation

• distributional shift
• regime-aware modeling
• TT-RoughPath encoder
• Mean-Teacher regularization
• multiphase pipeline leak detection