Cycle-Aware Autoencoder with Cross-SignalConsistency for Railway Door Anomaly Detection

Ammar Bouketta; Smail Niar; Hamza Ouarnoughi; Eva Mutuzo Brindle

cs.LGSep 30, 2026

Cycle-Aware Autoencoder with Cross-SignalConsistency for Railway Door Anomaly Detection

Authors: Ammar Bouketta, Smail Niar, Hamza Ouarnoughi, Eva Mutuzo Brindle

Organizations: Universit´e Polytechnique Hauts-de-France, LAMIH CNRS UMR 8201, Valenciennes, France · Alstom, Crespin, France · Universit´e Polytechnique Hauts-de-France, LAMIH CNRS UMR 8201, INSA Hauts-de-France, Valenciennes, France · Computer Science Department, College of Computing and Informatics, University of Sharjah, Sharjah, UAE

Abstract

Passenger access doors are safety-critical subsystems in railway vehicles, yet detecting abnormal door behavior in real operation is challenging because faults are rare, diverse, and often unlabeled. This paper addresses railway door condition monitoring as a cycle-level unsupervised anomaly detection problem, where each complete opening-dwell-closing cycle is treated as a single monitoring unit. We propose the Temporal Cycle-Aware Attention Autoencoder with Cross-Signal Consistency (TCAA-CS), trained exclusively on nominal cycles. It combines a dual-stream encoder that processes continuous physical measurements (position, current, voltage) and binary logical states (door-closed, door-locked) through separate 1D-CNN branches, an LSTM encoder with temporal attention pooling, and a triple hybrid anomaly score fusing reconstruction error, latent-space deviation, and phase-aware cross-signal consistency. The consistency term helps identify cases where individual signals appear plausible but their inter-signal relationships become physically or logically inconsistent. On real industrial data from a passenger train in commercial service, TCAA-CS achieves 93.8% recall, 97.3% precision, and a 0.5% false-alarm rate, outperforming representative unsupervised baselines. System-level evaluation on an NVIDIA Jetson AGX Xavier supports the feasibility of real-time onboard deployment.

Figures & tables

Fig. 1: Cycle-level data representation. Left: segmentation of a day-long door recording into individual door operation cycles. Right: example of a temporally normalized cycle ( $T=256$ ) showing continuous signals and logical states aligned with opening and closing phases.

Days	Cars	Doors	Cycles	Features ( $F$ )	Cycle length ( $T$ )	Freq. (Hz)
30	5	16	4 960	5	256	50

TABLE I: Main characteristics of the railway door operation dataset.

Fig. 2: High-level overview of the TCAA-CS framework. Nominal cycles are used to train the reconstruction model, while inference combines signal fidelity, latent-space deviation, and cross-signal consistency into one anomaly decision per cycle.

Fig. 3: Detailed architecture of TCAA-CS for cycle-level railway door anomaly detection. The upper part shows dual-stream signal encoding, attention-based reconstruction, and latent representation learning. The lower part shows the triple hybrid anomaly scoring mechanism.

	Days	Cycles	Nominal	Anomalous
Training	20	3 179	3 179	0
Validation	4	620	620	0
Test	6	1 161	969	192

TABLE II: Day-wise train/validation/test split and final test-set composition used for evaluation.

Class	Type	Description
A	Energy / Load	Increased or biased motor current and/or voltage during opening or closing, simulating friction.
B	Temporal/ Transient	Local stretching or compression of motion phases, short spikes, dropouts, or sensor noise.
C	Logical	Inconsistent signals, e.g., door locked active while the door is moving.

TABLE III: Abnormal cycle classes used for test-time evaluation.

Method	$TP$	$FN$	$FP$	$TN$	Recall	Prec.	F1	FAR
LSTM-AE [ 7 ]	145	47	22	947	0.755	0.868	0.808	0.023
GRU-AE [ 11 ]	163	29	20	949	0.849	0.891	0.870	0.021
TCN-AE [ 12 ]	136	56	16	953	0.708	0.895	0.791	0.017
OC-SVM [ 13 ]	120	72	30	939	0.625	0.800	0.702	0.031
Transformer [ 10 ]	158	34	18	951	0.823	0.898	0.859	0.019
TCAA-CS	180	12	5	964	0.938	0.973	0.955	0.005

TABLE IV: Cycle-level anomaly detection results. FAR is computed on nominal test cycles only.

Configuration	Recall	Prec.	F1	FAR
Full TCAA-CS	0.938	0.973	0.955	0.005
w/o dual-stream	0.906	0.946	0.925	0.012
w/o attention ( $h_{T}$ only)	0.880	0.939	0.909	0.014
w/o $\mathcal{C}_{\text{cross}}$	0.891	0.950	0.920	0.011
w/o $\mathcal{D}_{\text{lat}}$	0.917	0.941	0.929	0.014
$\mathcal{L}_{\text{rec}}$ only	0.781	0.926	0.847	0.015

TABLE V: Component ablation. Each row removes one component from the full TCAA-CS model; “w/o” denotes “without”.

Platform	Mean lat. (ms)	p95 lat. (ms)	Max lat. (ms)	GPU mem. (MB)	Energy
Intel i7-8850H CPU	8.52	10.41	14.8	–	–
Quadro P2000 GPU	5.21	6.08	8.23	32.7	–
Jetson AGX Xavier	4.98	6.52	8.92	32.7	8–12 W

TABLE VI: System-level performance of TCAA-CS under cycle-by-cycle inference.

Fig. 4: Representative door cycles: one nominal cycle and three anomalous cycles illustrating load-related, kinematic, and logical inconsistency fault mechanisms. Blue: observed signals. Red: TCAA-CS reconstruction. Red dashed boxes: anomaly regions.

Explore similar work

Deep Learning for Anomaly Detection in Railway Systems: A Structured Survey
Sep 30, 2026Ammar Bouketta, Smail Niar, Hamza OuarnoughiRelighting Unsupervised Detection
Université Polytechnique Hauts-de-France, LAMIH CNRS UMR 8201, Valenciennes, France · Alstom, Crespin, France · Université Polytechnique Hauts-de-France, LAMIH CNRS UMR 8201, INSA Hauts-de-France, Valenciennes, France +1
TPA-AD: A Two-Stage Pseudo Anomaly-Guided Method for Bearing Time-Series Anomaly Detection
Jun 2, 2026Xiancheng Wang, Zhibo Zhang, Ran Li +4Time-Series Anomaly Detection Graph Anomaly Detection
School of Ocean Engineering, Harbin Institute of Technology, West Wenhua Road, Weihai, 264209, Shandong, China · Technical Center, Bogie Development Department, CRRC Qingdao Sifang Locomotive and Rolling Stock Co., Ltd, Jinhong East Road, Qingdao, 266111, Shandong, China · Qingdao University, No. 308 Ningxia Road, Qingdao, 266071, Shandong, China
Real-time and adaptive anomaly detection algorithm for cyclostationary models
Sep 8, 2026Justyna Witulska, Tomasz Barszcz, Ireneusz Jabłoński +1Time-Series Anomaly Detection Graph Anomaly Detection
Faculty of Pure and Applied Mathematics and Hugo Steinhaus Center, Wrocław University of Science and Technology, Wrocław, 50-370, Poland · AGH University, Kraków, 30-059, Poland · Faculty of MINT, Brandenburg University of Technology, Cottbus, 03046, Germany

Sep 30, 2026, cs.LGJ/K move · Enter open · S save

Deep Learning for Anomaly Detection in Railway Systems: A Structured Survey

Ammar Bouketta, Smail Niar, Hamza Ouarnoughi

Université Polytechnique Hauts-de-France, LAMIH CNRS UMR 8201, Valenciennes, France · Alstom, Crespin, France · Université Polytechnique Hauts-de-France, LAMIH CNRS UMR 8201, INSA Hauts-de-France, Valenciennes, France+1

Relighting Unsupervised Detection Deep Learning Survey

Open paper AskPDF arXiv

Ensuring safe and reliable operation of modern railway systems increasingly relies on data-driven monitoring and intelligent fault detection. Deep learning has emerged as an effective paradigm for railway anomaly detection, driven by the growing availability of heterogeneous sensor data from rolling stock and infrastructure. This paper presents a structured survey of deep learning-based anomaly detection approaches for railway systems. The surveyed methods are organized using a unified taxonomy covering anomaly location, data representation and manifestation, sensing modality, and temporal characteristics. Existing approaches, including convolutional, recurrent and attention-based architectures, autoencoders, generative adversarial networks, and transformers, are structured into classification-based, prediction-based, reconstruction-based, and hybrid learning paradigms. The survey also examines data-centric challenges, evaluation practices, performance metrics, and practical deployment aspects, including edge-cloud architectures, computational constraints, and hardware-aware optimization. Finally, a decision-oriented framework links anomaly characteristics, data properties, and operational constraints to suitable detection paradigms and deployment configurations. This work provides a structured reference for selecting and deploying deep learning solutions for railway anomaly detection and highlights open challenges toward reliable and scalable intelligent monitoring systems.

Abstract

Explore similar work

Deep Learning for Anomaly Detection in Railway Systems: A Structured Survey

TPA-AD: A Two-Stage Pseudo Anomaly-Guided Method for Bearing Time-Series Anomaly Detection

Real-time and adaptive anomaly detection algorithm for cyclostationary models