GDSAN: A physics-informed generative network for zero-shot labeled fault data generation

Abstract

Aero-engines are critical industrial assets whose failures can lead to severe consequences, highlighting the necessity of effective Prognostics and Health Management (PHM). However, existing approaches suffer from limitations in data availability and model accuracy, particularly when real fault samples are scarce or absent, hindering reliable diagnostics. This study develops a novel physics-informed network, named Generative Data-Simulation Adversarial Network (GDSAN), to generate labeled fault vibration signals for reliable aero-engine rotor systems health monitoring. This model introduces a learnable modifying matrix to systematically reconcile discrepancies between simulated and measured data across four error dimensions. After that, physics-informed spectral and energy constraints are embedded into the loss function to enhance both model training stability and the physical plausibility of generated signals. Furthermore, a hybrid-driven PHM framework is constructed, leverages former generated labeled fault data to realize zero-shot fault diagnosis, thereby reducing reliance on high-fidelity simulation models or extensive measured fault samples. The following experimental validation on an aero-engine test bench demonstrates that the proposed framework successfully generates labeled fault signals closely aligned with experimental measurements in both the feature space and frequency spectrum, and eliminates the desperate need for enormous but expensive measured fault samples in model training process. Moreover, the proposed physics-informed terms in the loss function significantly improve the physical plausibility of generated signals.

Graphical abstract

Keywords

aero-engine / rotor system / data generation / physics-informed / hybrid-driven / prognostics and health management

References

[1] Cano T C , Castro I , Rodríguez A , Lamar D G , Khalil Y F , Albiol-Tendillo L , Kshirsagar P . Future of electrical aircraft energy power systems: An architecture review. IEEE Transactions on Transportation Electrification, 2021, 7(3): 1915–1929

[2] Xia X Z , Fu X Y , Zhong S S , Li Z , Fu S , Bai Z F , Liu X . A multi-agent convolution deep reinforcement learning network for aeroengine fleet maintenance strategy optimization. Journal of Manufacturing Systems, 2023, 68: 410–425

[3] Dong J , Li D Y , Cong Z Y , Peng K . A new fault detection method based on an updatable hybrid model for hard-to-detect faults in nonstationary processes. Reliability Engineering & System Safety, 2025, 259: 110920

[4] Meng T , Jiang X X , Song Q Y , Hu Z H , Guo J F , Zhu Z K . Modified feature mode decomposition guided by spectral structure information for machinery fault diagnosis. IEEE Transactions on Instrumentation and Measurement, 2024, 73: 3525810

[5] Barbieri N , de Sant’Anna Vitor Barbieri G , Martins B M , de Sant’Anna Vitor Barbieri L , de Lima K F . Analysis of automotive gearbox faults using vibration signal. Mechanical Systems and Signal Processing, 2019, 129: 148–163

[6] Rajpoot S C , Singhai S K . Efficient detection and analysis of shunt faults in electric power distribution systems (EPDS) using DCFC and EFDC algorithm: a modeling and simulation approach. International Journal of Modelling and Simulation, 2024, 7: 1–32

[7] Pan W J , Li X M , Kong K S , Li J , Wang J Y , Bao J W , Gao P , Zeng X , Nie P . Prediction model and loosening characteristics of a bolted connection system under torsional excitation. Frontiers of Mechanical Engineering, 2025, 20(3): 20

[8] Hu C , Shi T L , Yang J , Xiong F , Duan J . Nonlinear dynamic analysis of a dual-rotor system with deteriorating rub-impact fault caused by loose disc. Nonlinear Dynamics, 2024, 112(18): 15987–16001

[9] Hu C , Jiang S , Duan J , Shi T L . Dynamic characteristics of a multiple-crack rotor system versus crack relative orientation angle. Journal of the Brazilian Society of Mechanical Sciences and Engineering, 2025, 47(4): 193

[10] Zeng W , Lu W , Xu F . Model-based low-speed rotation error prediction for the rigid shaft-bearing system considering the assembly deviation. Mechanism and Machine Theory, 2023, 184: 105294

[11] Bernardi E , Adam E J . Observer-based fault detection and diagnosis strategy for industrial processes. Journal of the Franklin Institute, 2020, 357(14): 10054–10081

[12] Xia C J , Wang S L , Wang S B , Ma C , Xu K . Geometric error identification and compensation for rotary worktable of gear profile grinding machines based on single-axis motion measurement and actual inverse kinematic model. Mechanism and Machine Theory, 2021, 155: 104042

[13] Duan J , Hu C , Zhan X B , Zhou H D , Liao G L , Shi T L . MS-SSPCANet: A powerful deep learning framework for tool wear prediction. Robotics and Computer-integrated Manufacturing, 2022, 78: 102391

[14] Duan J , Liang J Q , Yu X J , Si Y , Zhan X B , Shi T L . Toward practical tool wear prediction paradigm with optimized regressive siamese neural network. Advanced Engineering Informatics, 2023, 58: 102200

[15] Yan X A , She D M , Xu Y D , Jia M P . Deep regularized variational autoencoder for intelligent fault diagnosis of rotor–bearing system within entire life-cycle process. Knowledge-Based Systems, 2021, 226: 107142

[16] Zhao K , Jiang H K , Liu C Q , Wang Y F , Zhu K . A new data generation approach with modified Wasserstein auto-encoder for rotating machinery fault diagnosis with limited fault data. Knowledge-Based Systems, 2022, 238: 107892

[17] Chen Y J , Liu X M , Rao M , Qin Y , Wang Z P , Ji Y J . Explicit speed-integrated LSTM network for non-stationary gearbox vibration representation and fault detection under varying speed conditions. Reliability Engineering & System Safety, 2025, 254: 110596

[18] de Castro-Cros M , Velasco M , Angulo C . Machine-learning-based condition assessment of gas turbines—A review. Energies, 2021, 14(24): 8468

[19] Duan J , Zhang X , Shi T L . A hybrid attention-based paralleled deep learning model for tool wear prediction. Expert Systems with Applications, 2023, 211: 118548

[20] Duan J , Hu C , Zhou H D , Zhan X B , Xiong F , Shi T L . High-speed bearing health monitoring method based on attention mechanism optimized siamese deep residual network. IEEE Sensors Journal, 2023, 23(19): 23191–23200

[21] Baradieh B , Zainuri M , Kamari M , Yusof Y , Abdullah H , Zaman M , Zulkifley M . RSO based optimization of random forest classifier for fault detection and classification in photovoltaic arrays. International Arab Journal of Information Technology, 2024, 21(4): 636–660

[22] Kusiak A . Generative artificial intelligence in smart manufacturing. Journal of Intelligent Manufacturing, 2025, 36(1): 1–3

[23] Li B Y , Xie X , Yu B , Liao Y W , Fan D . Data-driven friction modeling and compensation for rotary servo actuators. Frontiers of Mechanical Engineering, 2024, 19(6): 41

[24] Cheng Z B , Yu T X , Jia G F , Shi Z F . A physics-informed neural network-based method for dispersion calculations. International Journal of Mechanical Sciences, 2025, 291–292: 110111

[25] Chen K K , Dong X J , Gao P L , Chen Q , Peng Z K , Meng G . Physics-informed neural networks for topological metamaterial design and mechanical applications. International Journal of Mechanical Sciences, 2025, 301: 110489

[26] Banerjee C , Nguyen K , Fookes C , Raissi M . A survey on physics informed reinforcement learning: Review and open problems. Expert Systems with Applications, 2025, 287: 128166

[27] Cai S Z , Wang Z C , Wang S F , Perdikaris P , Karniadakis G E . Physics-informed neural networks for heat transfer problems. Journal of Heat Transfer, 2021, 143(6): 060801

[28] Stiefelmaier J , Böhm M , Sawodny O , Tarín C . Hybrid model- and learning-based fault diagnosis in adaptive buildings. Control Engineering Practice, 2024, 151: 106037

[29] Han X , Zhao P , Zhao X R , Zi B . Review on machine learning-based approaches for the kinematic analysis and synthesis of mechanisms. Frontiers of Mechanical Engineering, 2025, 20(2): 11

[30] Benninger M , Liebschner M . Optimization of practicality for modeling- and machine learning-based framework for early fault detection of induction motors. Energies, 2024, 17(15): 3723

[31] Yang H , Yang X D , Sun D , Hu Y J . Construction health indicator using physically-informed 1D-WGAN-GP joint attention LSTM-DenseNet method. Measurement Science and Technology, 2024, 35(7): 076204

[32] Benninger M , Liebschner M , Kreischer C . Fault detection of induction motors with combined modeling and machine-learning-based framework. Energies, 2023, 16(8): 3429

[33] Akbalık F , Yıldız A , Ertuğrul Ö F , Zan H . Engine fault detection by sound analysis and machine learning. Applied Sciences, 2024, 14(15): 6532

[34] Li X L , Xi Y H , Zhou F . Power quality disturbances classification with imbalanced/insufficient samples based on WGAN-GP-SA and DCNN. Digital Signal Processing, 2024, 151: 104518

[35] Tian X X , Li S Y , Zhang H , Li Q B , Ai S G. . A digital twin framework for turbine blade crack propagation prediction.. International Journal of Mechanical Sciences, 2025, 297–298: 110396

[36] Sun H , Peng F Y , Zhou L , Yan R , Zhao S Q . A hybrid driven approach to integrate surrogate model and Bayesian framework for the prediction of machining errors of thin-walled parts. International Journal of Mechanical Sciences, 2021, 192: 106111

[37] Li Q , Chu L Y , Sun Q , Tang Y N , Zhang Y P . Fault identification of centrifugal pump using WGAN-GP method with unbalanced datasets based on kinematics simulation and experimental case. Measurement Science and Technology, 2024, 35(9): 096108

[38] Liao W J , Wu L K , Xu S H , Fujimura S . A novel zero-shot learning method with feature generation for intelligent fault diagnosis. IEEE Transactions on Industrial Informatics, 2025, 21(4): 3386–3395

[39] Mu G Q , Liu C L , Chen J H . Integrating physical knowledge for generative-based zero-shot learning models in process fault diagnosis. Reliability Engineering & System Safety, 2025, 258: 110852

[40] Zhang X G , Yang J Z , Yang X M . Residual life prediction of rolling bearings based on a CEEMDAN algorithm fused with CNN–attention-based bidirectional LSTM modeling. Processes, 2023, 12(1): 8

[41] Yang R , Zhang Z Y , Chen Y S . Analysis of vibration signals for a ball bearing-rotor system with raceway local defects and rotor eccentricity. Mechanism and Machine Theory, 2022, 169: 104594

RIGHTS & PERMISSIONS

Higher Education Press