What question did this study set out to answer?

The aim is to enhance early fault detection in industrial plants by improving accuracy and interpretability of alarm systems.

April 18, 2026Scientific Reports8 citationsOpen Access

Early-warning industrial fault detection based on physics-guided residual learning and calibrated CRNNs

AKAbuzar KhanCECOS University FFFahmid Al FaridMultimedia University AJAhmad JunaidCECOS University

Key Points

The aim is to enhance early fault detection in industrial plants by improving accuracy and interpretability of alarm systems.
Developed a physics-guided pipeline combining twin residuals and compact descriptors.
Utilized a convolutional recurrent neural network (CRNN) for temporal structure modeling.
Employed SHAP for feature selection and Platt scaling for calibration.
Quantified reliability using bootstrap bias-corrected confidence intervals.
Achieved approximately 99.0% accuracy and near 1.00 AUC-ROC.
Macro F1-score measured at 0.93 ± 0.02 after calibration.
Improved Numenta Anomaly Benchmark score by 17% over Isolation Forest.
Reduced expected calibration error to approximately 0.03.

Abstract

Industrial plants generate high volume multivariate sensor time series where early fault detection must be accurate, governable and explainable under regime changes, sensor correlations and drift. Many published Tennessee Eastman Process (TEP) studies report strong receiver operating characteristic (ROC) or accuracy but provide limited evidence for calibrated probabilities, uncertainty bounds and actionable sensor level diagnostics, so alarm thresholds remain ad hoc and difficult to defend in control rooms. To address this issue, we propose a physics guided pipeline that combines twin based residuals with compact rolling and spectral descriptors and an attention based convolutional recurrent neural network (CRNN). The proposed method is also compared against state-of-the-art (SOTA) baselines to establish its relative effectiveness under the same evaluation setting. Residuals highlight deviations from physics consistent behavior, while a one-dimensional convolutional neural network (Conv1D) and bidirectional gated recurrent unit (BiGRU) encoder with attention models short and long temporal structure over 120 step windows. SHapley Additive exPlanations (SHAP) guided feature selection reduces redundancy and improves interpretability, Optuna stabilizes training and Platt scaling calibrates anomaly probabilities for policy driven thresholding under a false alarm budget. Reliability is quantified with 1000 block bootstrap bias-corrected and accelerated (BCa) confidence intervals. On a run level 70/20/10 split of Tennessee Eastman Process runs, the method achieves about 99.0% accuracy with area under the receiver operating characteristic curve (AUC-ROC) and area under the precision-recall curve (AUC-PR) near 1.00, macro F1-score of 0.93 ± 0.02 and expected calibration error (ECE) reduced to ≈ 0.03 after calibration. Early warning governance improves the Numenta Anomaly Benchmark (NAB) score by 17% over an Isolation Forest baseline and supports low nuisance alarm operation. These results show that residual-guided learning plus calibrated decision governance can support deployment-oriented industrial monitoring by enabling auditable thresholds and operator-facing explanations within the Tennessee Eastman benchmark setting.

Perguntar à IA

Bookmark

View Full Paper

Cite This Study

Khan et al. (Wed,) studied this question.

synapsesocial.com/papers/69e31f1a40886becb653e901 https://doi.org/https://doi.org/10.1038/s41598-026-48227-6

Perguntar à IA

Bookmark

View Full Paper