Intelligent spatiotemporal-fused deep learning resilient framework for maritime cyber-physical systems

Abstract

Modern shipboard power systems have increasingly integrated cyber-physical systems with the digital transformation of the maritime industry, thereby exposing them to unprecedented cybersecurity threats. Among these, false data injection attacks (FDIA) present a particularly severe challenge to vessel operational safety because they can evade conventional detection methods and subtly manipulate system state measurements while remaining concealed. To address this critical challenge, this study proposes an integrated framework referred to as CNN-BiLSTM-GAM, which is designed for effective FDIA detection and accurate data recovery in shipboard power systems. The framework synergistically integrates multiple deep learning architectures: a convolutional neural network (CNN) extracts spatial features from multivariate sensor data affected by attacks, a bidirectional long short-term memory (BiLSTM) network captures bidirectional temporal dependencies in power system operational data, and a global attention mechanism (GAM) dynamically weighs the spatiotemporal features, enabling the precise identification of subtle anomaly patterns induced by attacks in the highly dynamic and noisy environment of shipboard power systems. Compared with existing studies, experimental results demonstrate that the proposed CNN-BiLSTM-GAM framework achieves outstanding performance in attack detection, with an accuracy of 97.71% and an F1-score of 0.9548. In data recovery, it achieves a coefficient of determination (R²) of 0.9686, significantly outperforming various baseline models.

Keywords

Shipboard power systems / False data injection attack / CNN-BiLSTM-GAM / Attack detection and recovery

References

[1] Abbasimehr H, Paki R. Improving time series forecasting using LSTM and attention models. J Ambient Intell Humaniz Comput, 2022, 13: 673-691

[2] Aboelezz AM, El-Saadawi MM, Eladl AA, El-Afifi MI, Bureš V, Sedhom BE. Improved protection scheme for shipboard microgrids based on high frequency impedance method with experimental validation. Int J Electr Power Energy Syst, 2025, 164: 110448

[3] Androjna A, Brcko T, Pavic I, Greidanus H. Assessing cyber challenges of maritime navigation. J Mar Sci Eng, 2020, 8(10): 776

[4] Baul A, Sarker GC, Sadhu PK, Yanambaka VP, Abdelgawad A. XTM: a novel transformer and LSTM-based model for detection and localization of formally verified FDI attack in smart grid. Electronics, 2023, 12(4): 797

[5] Broussolle F. State estimation in power systems: detecting bad data through the sparse inverse matrix method. IEEE Trans Power Appar Syst, 1978, PAS-97(3): 678-682

[6] Callenes J, Poshtan M. Dynamic reconfiguration for resilient state estimation against cyber attacks. IEEE Trans Emerg Top Comput, 2024, 12(2): 559-571

[7] Cao J, Wang QM, Qu ZY, Chen CL, Dong YC. Method for identifying false data injection attacks in power grid based on improved CNN-LSTM. Electr Eng, 2025, 107: 9309-9334

[8] Cheng G, Lin YZ, Chen YB, Bi TS. Adaptive state estimation for power systems measured by PMUs with unknown and time-varying error statistics. IEEE Trans Power Syst, 2021, 36(5): 4482-4491

[9] Chung J, Jang B. Accurate prediction of electricity consumption using a hybrid CNN-LSTM model based on multivariable data. PLoS ONE, 2022, 17(11): e0278071

[10] Cofta P. Simplifying the operation of hybrid ad-hoc sensor networks with neural networks as the sole data reconstruction method. Expert Syst Appl, 2025, 296: 128981

[11] Du SD, Li TR, Yang Y, Horng SJ. Multivariate time series forecasting via attention-based encoder-decoder framework. Neurocomputing, 2020, 388: 269-279

[12] Esmalifalak M, Liu LC, Nguyen N, Zheng R, Han Z. Detecting stealthy false data injection using machine learning in smart grid. IEEE Syst J, 2017, 11(3): 1644-1652

[13] Haghshenas SH, Hasnat MA, Naeini M (2023) A temporal graph neural network for cyber attack detection and localization in smart grids. In: 2023 IEEE Power & Energy Society Innovative Smart Grid Technologies Conference (ISGT). IEEE, pp 1–5

[14] He YB, Mendis GJ, Wei J. Real-time detection of false data injection attacks in smart grid: a deep learning-based intelligent mechanism. IEEE Trans Smart Grid, 2017, 8(5): 2505-2516

[15] Hu D, Zhang C, Yang T, Chen G. Anomaly detection of power plant equipment using long short-term memory based autoencoder neural network. Sensors, 2020, 20(21): 6164

[16] Karim MS. Maritime cybersecurity and the IMO legal instruments: sluggish response to an escalating threat?. Mar Pol, 2022, 143: 105138

[17] Kalangi RR, Unhelkar B, Chakrabarti C, Ganesh C, Vidyullatha P, Shaik A (2025) Hybrid CNN-LSTM architecture for robust cloud security through anomaly detection and threat mitigation. J Inf Syst Eng Manag 10(27s):217–227. https://doi.org/10.52783/jisem.v10i27s.4398

[18] Kausar F, Deo S, Hussain S, Ul Haque Z. Federated deep learning model for false data injection attack detection in cyber physical power systems. Energies, 2024, 17(21): 5337

[19] Kayan H, Nunes M, Rana O, Burnap P, Perera C. Cybersecurity of industrial cyber-physical systems: a review. ACM Comput Surv, 2022, 54(11S): 229

[20] Li C, Wang X, Chen X, Han A, Zhang X. Data-driven attack detection mechanism against false data injection attacks in DC microgrids using CNN-LSTM-Attention. Symmetry, 2025, 17(7): 1140

[21] Li XP, Hu LB, Lu ZG (2024) Detection of false data injection attack in power grid based on spatial-temporal transformer network. Expert Syst Appl 238(Part A):121706. https://doi.org/10.1016/j.eswa.2023.121706

[22] Li XP, Wang YK, Lu ZG. Graph-based detection for false data injection attacks in power grid. Energy, 2023, 263: 125865

[23] Lin ZC (2023) A hybrid CNN-LSTM-Attention model for electric energy consumption forecasting. In: APIT’23: Proceedings of the 2023 5th Asia Pacific Information Technology Conference. ACM, pp 167–173. https://doi.org/10.1145/3588155.3588182

[24] Liu Y, Ning P, Reiter MK. False data injection attacks against state estimation in electric power grids. ACM Trans Inf Syst Secur, 2011, 14(1): 13

[25] Liu YC, Shao ZR, Hoffmann N (2021) Global attention mechanism: retain information to enhance channel-spatial interactions. Preprint at arXiv:2112.05561

[26] Luo C, Zhou ZY, Jiang R, Zheng BC. Attentional convolutional neural network based on distinction enhancement and information fusion for FDIA detection in power systems. Electronics, 2024, 13(24): 4862

[27] Mai VT, Mohammadzadeh A, Alattas KA, Taghavifar H, Ghaderpour E. Cybersecurity in maritime power systems: a comprehensive review of cyber threats and mitigation techniques. Electr Power Syst Res, 2025, 247: 111797

[28] Manandhar K, Cao XJ, Hu F, Liu Y. Detection of faults and attacks including false data injection attack in smart grid using Kalman filter. IEEE Trans Control Netw Syst, 2014, 1(4): 370-379

[29] Mohamed AS, Arani MFM, Jahromi AA, Kundur D. False data injection attacks against synchronization systems in microgrids. IEEE Trans Smart Grid, 2021, 12(5): 4471-4483

[30] Nguyen DD, Le MT, Cung TL. Improving intrusion detection in SCADA systems using stacking ensemble of tree-based models. Bull Electr Eng Inform, 2022, 11(1): 119-127

[31] Oruc A, Kavallieratos G, Gkioulos V, Katsikas S. Perspectives on the cybersecurity of the integrated navigation system. J Mar Sci Eng, 2025, 13(6): 1087

[32] Riyadh M, Syaifullah M (2025) Literature review: development of maritime systems in the digital era. Maritime Technol Soc 4(1):17–22. https://doi.org/10.62012/mp.v4i1.43273

[33] Saleem F, Ahmad Z, Kim JM. Real-time pipeline leak detection: a hybrid deep learning approach using acoustic emission signals. Appl Sci-Basel, 2025, 15(1): 185

[34] Shin AH, Kim ST, Park GM. Time series anomaly detection using transformer-based GAN with two-step masking. IEEE Access, 2023, 11: 74035-74047

[35] Su XJ, Deng C, Yang JJ, Li FY, Li CJ, Fu Y, et al. . DAMGAT-based interpretable detection of false data injection attacks in smart grids. IEEE Trans Smart Grid, 2024, 15(4): 4182-4195

[36] Xiao YT, Yin HS, Zhang YD, Qi HG, Zhang YD, Liu ZY. A dual-stage attention-based Conv-LSTM network for spatio-temporal correlation and multivariate time series prediction. Int J Intell Syst, 2021, 36(5): 2036-2057

[37] Xie KM, Liu JC, Liu YB. A power system timing data recovery method based on improved VMD and attention mechanism bi-directional CNN-GRU. Electronics, 2023, 12(7): 1590

[38] Yang FY, Xie XP, Sun QY, Yue D. FDI attack estimation and event-triggered resilient control of DC microgrids under hybrid attacks. IEEE Trans Smart Grid, 2024, 15(4): 4207-4216

[39] Zhang CY, Fu XQ, Qiu DW, Badihi H, Gu HT. Robust imputation of missing photovoltaic power data using a weather- and context-aware hybrid transformer framework. Renew Energy, 2025, 256: 124576

[40] Zhou K, Tong YF, Li XT, Wei XR, Huang H, Song K, et al. . Exploring global attention mechanism on fault detection and diagnosis for complex engineering processes. Process Saf Environ Protect, 2023, 170: 660-669

Funding

National Nature Science Foundation of Chin(62473328)

RIGHTS & PERMISSIONS

The Author(s)