Physics‐Informed Fusion Learning for Nowcasting Fault Detection of Hydrological Multi Sensor Network
- 1Department of Technology, Water, Energy and Environmental Engineering Research Unit University of Oulu Oulu Finland
- 2Finnish Environmental Institute (SYKE) Helsinki Finland
- 3Centre for Wireless Communications University of Oulu Oulu Finland
Abstract
Urban flood monitoring demands timely, reliable decisions under urgent and often unseen conditions. However, faults in sensor networks can undermine response, leading to human and economic losses. To address these issues, we propose a Nowcasting Physics‐Informed (NPI) model for fault detection in streamflow sensors, with a 100‐min memory window at 2‐min resolution. A Long Short‐Term Memory (LSTM) consumes fused features from measured signals and Storm Water Management Model simulator outputs to produce end‐of‐window fault probabilities. To ensure operational reliability, a lightweight per‐sensor eXtreme Gradient Boosting (XGBoost) stage calibrates the LSTM score using window‐local statistics of the measurement. In this study, we investigate whether and to what extent fusing a physical model with sensor data improves fault detection, and whether per‐sensor threshold calibration matters. Our results indicated yes. The Area under the precision–recall curve (AUC‐PR) was improved by 3.7%–12.5% over the physics‐informed model compared to baseline non‐physics‐informed learning. In addition, post‐hoc calibration improved AUC‐PR by an additional 0.2%–2.3%. Across five cross‐validation folds, F1‐score (F1) increased from 0.75 for the non‐physics LSTM baseline to 0.83 with calibration, and further to 0.88 for the physics informed model. Adding post‐hoc calibration to the physics informed model yielded the best result at 0.90. The NPI is causal and auditable, exporting calibrated probabilities and binary alerts for direct operator use without retraining the LSTM, and offers a practical template for warning systems and training strategies that remain robust to unseen conditions.
Related
- Physics-informed deep learning links geodetic data and fault friction
- Physics-Informed Machine Learning in Hydrology and Water Resources: A Systematic Bibliometric Analysis
- A comparison of data-driven and physics-informed statistical learning models for atmospheric remote sensing applications
- Daily global sea surface ageostrophic current dataset from 1993–2023 via physics-informed deep learning
- Plume Segmentation from MethaneSAT with Cross-Sensor Transfer Learning and Physics-Informed Postprocessing
BibTeX
@article{mousavi2026physicsinformed,
title = {{Physics‐Informed Fusion Learning for Nowcasting Fault Detection of Hydrological Multi Sensor Network}},
author = {Farid Mousavi and Ali Torabi Haghighi and Jari Silander and Mehdi Monemi and Mehdi Rasti},
journal = {Water Resources Research},
year = {2026},
month = oct,
doi = {10.1029/2026wr043436},
url = {https://doi.org/10.1029/2026wr043436},
note = {CC BY},
}
RIS
TY - JOUR
TI - Physics‐Informed Fusion Learning for Nowcasting Fault Detection of Hydrological Multi Sensor Network
AU - Farid Mousavi
AU - Ali Torabi Haghighi
AU - Jari Silander
AU - Mehdi Monemi
AU - Mehdi Rasti
PY - 2026
DA - 2026/10/28
JO - Water Resources Research
DO - 10.1029/2026wr043436
UR - https://doi.org/10.1029/2026wr043436
AB - Urban flood monitoring demands timely, reliable decisions under urgent and often unseen conditions. However, faults in sensor networks can undermine response, leading to human and economic losses. To address these issues, we propose a Nowcasting Physics‐Informed (NPI) model for fault detection in streamflow sensors, with a 100‐min memory window at 2‐min resolution. A Long Short‐Term Memory (LSTM) consumes fused features from measured signals and Storm Water Management Model simulator outputs to produce end‐of‐window fault probabilities. To ensure operational reliability, a lightweight per‐sensor eXtreme Gradient Boosting (XGBoost) stage calibrates the LSTM score using window‐local statistics of the measurement. In this study, we investigate whether and to what extent fusing a physical model with sensor data improves fault detection, and whether per‐sensor threshold calibration matters. Our results indicated yes. The Area under the precision–recall curve (AUC‐PR) was improved by 3.7%–12.5% over the physics‐informed model compared to baseline non‐physics‐informed learning. In addition, post‐hoc calibration improved AUC‐PR by an additional 0.2%–2.3%. Across five cross‐validation folds, F1‐score (F1) increased from 0.75 for the non‐physics LSTM baseline to 0.83 with calibration, and further to 0.88 for the physics informed model. Adding post‐hoc calibration to the physics informed model yielded the best result at 0.90. The NPI is causal and auditable, exporting calibrated probabilities and binary alerts for direct operator use without retraining the LSTM, and offers a practical template for warning systems and training strategies that remain robust to unseen conditions.
SN - 1944-7973
SN - 0043-1397
ER -