近年来，云计算、深度学习等各种新兴技术，已经在故障诊断领域全面展开 应用。工程技术人员在处理智能船舶相关设备的故障监测和健康维护问题时，通 常需要收集大量数据，过去采用的是集中式的数据处理方法，但出于数据隐私安 全方面考虑，各数据拥有者不愿意甚至禁止分享本地的数据资源集中处理。为解 决上述问题，以船舶机舱设备为研究对象，引入船联网技术促进数据的收集，并 融合联邦学习技术打破数据壁垒，保护数据隐私，提出了基于船联网的动态验证 的隐私保护联邦故障诊断方法，最后验证了该方法的有效性并提出了该方法应 用于实际的构思。本文的研究内容主要体现在以下 3 个方面： （1）提出了基于船联网的联邦故障诊断模型。首先，给出了用于故障诊断 的卷积神经网络模型，并针对性地选用了模型的部件参数及相关算法。然后，为 适应船联网特殊通信环境提出模型稀疏化方法，用于减少模型参数。最后，将联 邦学习融合上述故障诊断模型得出基于船联网的联邦故障诊断模型，该模型将 船联网的各个参与者划分为服务器端和客户端，在本地客户端处理数据形成局 部模型，并在服务器端聚合本地模型形成全局模型，使用分布式方法处理数据从 而保护数据的隐私安全。 （2）为提升（1）中模型的性能，提出了动态验证和同态加密的方案。首先， 为提升故障诊断精度，位于服务器端的验证集会动态地验证客户端上传的模型 的有效性，并选择性地完成聚合过程。然后，为提升数据隐私保护能力，应用 Paillier 加密算法的加密方案将使用密钥保护模型参数的分发过程。最后，结合 （1）中模型与动态验证和同态加密方案，提出基于船联网的动态验证的隐私保 护联邦故障诊断模型，并给出了该模型的操作流程。 （3）为验证本文模型有效性和推广实际应用的可行性，进行了模拟实验并 提出了实际应用方案。首先，实验一针对本文模型进行了密码学成本和通信成本 的计算，结果表明本文模型相对传统联邦学习的成本下降了。然后，实验二针对 IID 和 Non-IID 两种不同数据场景对比了诊断故障的精度，结果表明本文模型相 比集中方案和基础方案，在诊断能力上面有了很大提升。最后，为将本文模型推 广实际应用，针对内河、近海和远洋 3 种水域场景给出了对应的通信方案。

[1] Park J S, Chen LJ, Fujii T, et al. Internet of Things over future internet [J]. International Journal

of Distributed Sensor Networks, 2016, 12(8): 113-123.

[2] 李金华. 德国“工业 4.0”与“中国制造 2025”的比较及启示 [J].中国地质大学学报(社会科学版),2015, 15(05):71-79.

[3] 孙武. 中国船级社发布《智能船舶规范（2020）》 [J].船舶工程, 2020, 42(03): 13-14.

[4] Zhang Z, Xu X, Gong W, et al. Efficient federated convolutional neural network with

information fusion for rolling bearing fault diagnosis [J]. Control Engineering Practice, 2021,

116(1013-1025).

[5] 曹雪晖. 从概念到实践:e 航海发展回顾 [J]. 中国海事, 2019, 2(09): 55-57.

[6] Liu G, Perez R, Muñoz J A, et al. Internet of Ships: the future ahead [J]. World Journal of

Engineering and Technology, 2016, 04(03): 220-227.

[7] Khan L U, Saad W, Han Z, et al. Federated learning for internet of things: recent advances, taxonomy, and open challenges [J]. IEEE Communications Surveys & Tutorials, 2021, 23(3): 1759-1799.

[8] Pérez Fernández R, Benayas Ayuso A, Pérez Arribas F L. Data management for smart ship or

how to reduce machine learning cost in IoS applications [M]. Smart Ship Technology 2018.

Londres, United Kingdom. 2018: 101-106.

[9] Tian Z, Liu F, Li Z, et al. The development of key technologies in applications of vessels

connected to the internet [J]. Symmetry, 2017, 9(10): 211-231.

[10] Hiekata K, Wanaka S, Mitsuyuki T, et al. Systems analysis for deployment of internet of things

(IoT) in the maritime industry [J]. Journal of Marine Science and Technology, 2020, 26(2):

459-469.

[11] Sattler F, Wiedemann S, Müller K R, et al. Robust and communication-efficient federated

learning from non-i.i.d. data [J]. IEEE Transactions on Neural Networks and Learning Systems,

2019, 31(3400-3413).

[12] Yang Q, Liu Y, Chen T, et al. Federated machine learning: concept and applications [J].

Machine Learning, 2019, 10(10-19).

[13] 梁天恺, 曾碧, 陈光，等. 联邦学习综述：概念、技术、应用与挑战 [J]. 计算机应用, 2021,

1-13.

[14] Zhao R, Yan R, Chen Z, et al. Deep learning and its applications to machine health monitoring

[J]. Mechanical Systems and Signal Processing, 2019, 115(15): 213-237.

[15] Bertino E, Choo K R, Georgakopolous D, et al. Internet of Things (IoT): smart and secure

service delivery [J]. ACM Transactions on Internet Technology, 2016, 16(ACM Trans. Internet

Technol.): Article 22.

[16] Li J Q, Yu F R, Deng G, et al. Industrial internet: a survey on the enabling technologies,

applications, and challenges [J]. IEEE Communications Surveys & Tutorials, 2017, 19(3):

1504-1526.

[17] Lin B, Zhao L, Suraweera H A, et al. Guest editorial special issue on Internet of Things for

smart ocean [J]. IEEE Internet of Things Journal, 2020, 7(10): 9675-9687.

[18] Te W, Wei F, Yunfei C, et al. Hybrid satellite-terrestrial communication networks for the

maritime Internet of Things: key technologies, opportunities, and challenges [J]. IEEE Internet

of Things Journal, 2021, 8(11): 8910-8934.

[19] 覃闻铭, 王晓峰. 船联网组网技术综述 [J]. 中国航海, 2015, 38(02): 1-4+8.

[20] Xia T, Wang M M, Zhang J, et al. Maritime Internet of Things: challenges and solutions [J].

IEEE Wireless Communications, 2020, 27(2): 188-196.

[21] Kairouz P, Avent B. Advances and open problems in federated learning [J]. Foundations and

Trends® in Machine Learning, 2021, 14(5):256-275.

[22] Domingo Ferrer J, Blanco Justicia A, Manjon J, et al. Secure and privacy-preserving federated

learning via co-utility [J]. IEEE Internet of Things Journal, 2022, 9(5): 3988-4000.

[23] Phong L T, Phuong T T. Privacy-Preserving deep learning via weight transmission [J]. IEEE

Transactions on Information Forensics and Security, 2019, 14(11): 3003-3015.

[24]王凯, 剡昌锋, 王风涛, 等. 深沟球轴承复合故障动力学特征 [J]. 哈尔滨工业大学学报,

2020, 52(01): 133-140.

[25] Xiaoxi D, Qingbo H. Time–frequency manifold sparse reconstruction: A novel method for

bearing fault feature extraction [J]. Mechanical Systems and Signal Processing, 2016, 80(1):

392-413.

[26] 李舜酩, 郭海东, 李殿荣，等. 振动信号处理方法综述 [J]. 仪器仪表学报, 2013, 34(08):

1907-1915.

[27] Yaguo L, Bin Y, Xinwei J, et al. Applications of machine learning to machine fault diagnosis:

A review and roadmap [J]. Mechanical Systems and Signal Processing, 2020, 138(10): 382-

401

[28] Rui Z, Ruqiang Y, Zhenghua C, et al. Deep learning and its applications to machine health

monitoring [J]. Mechanical Systems and Signal Processing, 2019, 115(213-237).

[29] Mirza M B, Mian M A, Elsayedme A, et al. AdaBoost-based artificial neural network learning

[J]. Neurocomputing, 2017, 248(120-126).

[30] 张燕霞, 户文刚. 基于 VMD-SVD 和 SVM 的旋转机械故障诊断研究 [J]. 机电工程,

2022, 39(03): 324-329.

[31] 张炎亮, 颜健勇. 基于 G-DPSO 算法的决策树轴承故障诊断方法 [J]. 工业工程, 2021,

24(06): 41-47.

[32] 查园园, 王亭岭, 上官伟，等. 基于贝叶斯网络的列控车载设备故障诊断 [J]. 北京交通

大学学报, 2021, 45(05): 37-45.

[33] Wan S, Zhang X. Bearing fault diagnosis based on teager energy entropy and mean-shift fuzzy

C-means [J]. Structural Health Monitoring, 2020, 19(6): 1976-1988.

[34] Olivier J, Viktor S, Bram V, et al. Convolutional neural network based fault detection for

rotating machinery [J]. Journal of Sound and Vibration, 2016, 377(331-345).

[35] Wenjun S, Rui Z, Ruqiang Y, et al. Convolutional discriminative feature learning for induction

motor fault diagnosis[J]. IEEE Transactions on Industrial Informatics, 2017, 13(3): 1350-1359.

[36] Tim De B, Kim V, Robert B. Railway track circuit fault diagnosis using recurrent neural

networks [J]. IEEE Transactions on Neural Networks and Learning Systems, 2016, 28(3): 523-

533.

[37] 李巍华, 单外平, 曾雪琼，等. 基于深度信念网络的轴承故障分类识别 [J]. 振动工程学

报, 2016, 29(02): 340-247.

[38] 陈国成, 张建, 菅光雷，等. 基于改进叠加自动编码器轴承智能故障诊断方法 [J]. 噪声

与振动控制, 2022, 42(01): 156-161.

[39] Nan Z, Shifei D, JIAN Z, et al. An overview on restricted boltzmann machines [J].

Neurocomputing, 2018, 275(1186-99).

[40] Zuo B, Zhang Z, Cheng J, et al. Data-driven flooding fault diagnosis method for protonexchange membrane fuel cells using deep learning technologies [J]. Energy Conversion and

Management, 2022, 251(1125-1136).

[41] He W, Li Z, Malekian R, et al. An Internet of Things approach for extracting featured data

using ais database: an application based on the viewpoint of connected ships [J]. Symmetry,

2017, 9(9): 186-201.

[42] Shan H, J. Peterson I, Hathorn S, et al. The rfid connection: rfid technology for sensing and

the internet of things [J]. IEEE Microwave Magazine, 2018, 19(7): 63-79.

[43]Bohan L, Yutai H, Wan Xiang C. Data augmentation approaches in natural language processing:

a survey [J]. Machine Learning, 2021, arXiv:2110.01852.

[44] 朱晓慧, 钱丽萍, 傅伟，等. 图像数据增强技术研究综述 [J]. 软件导刊, 2021, 20(05):

230-236.

[45] Zhang X, Yao L, Huang C, et al. Multi-modality sensor data classification with select

ive attention [J]. Computer Vision and Pattern Recognition, 2018, arXiv e-prints(arXiv:

1804.05493).

[46] Du J, Li X, Gao Y, et al. Integrated gradient-based continuous wavelet transform for bearing

fault diagnosis [J] Sensors 2022, 22(22):8760-8774

[47] Pouyanfar S, Sadiq S, Yan Y, et al. A survey on deep learning: algorithms, techniques, and

applications [J]. ACM Computing Surveys 2018, 51(ACM Comput. Surv.): Article 92.

[48] Hoang D T, Kang H J. Rolling element bearing fault diagnosis using convolutional neural

network and vibration image [J]. Cognitive Systems Research, 2019, 53(42-50).

[49] Zhou D X. Universality of deep convolutional neural networks [J]. Applied and Computational

Harmonic Analysis, 2020, 48(2): 787-794.

[50] 宫文峰, 陈辉, 张泽辉，等. 基于改进卷积神经网络的滚动轴承智能故障诊断研究 [J].

振动工程学报, 2020, 33(02): 400-413.

[51] Baldi P, Sadowski P. The dropout learning algorithm [J]. Artificial Intelligence, 2014, 210(78-

122).

[52] Liu X, Li H, Xu G, et al. Privacy-Enhanced federated learning against poisoning adversaries

[J]. IEEE Transactions on Information Forensics and Security, 2021, 16(4574-4588).

[53] Zhang Z, Guan C, Chen H, et al. Adaptive privacy-preserving federated learning for fault

diagnosis in internet of ships [J]. IEEE Internet of Things Journal, 2022, 9(9): 6844-6854.

[54] Mcmahan H B, Moore E, Ramage D, et al. Communication-Efficient learning of deep

networks from decentralized data[C]; proceedings of the International Conference on Artificial

Intelligence and Statistics, F, 2016.

[55] Marcolla C, Sucasas V, Manzano M, et al. Survey on fully homomorphic encryption, theory,

and applications [J]. Proceedings of the IEEE, 2022, 110(10): 1572-1609.

[56] Phong L T, Aono Y, Hayashi T, et al. Privacy-Preserving deep learning via additively

homomorphic encryption [J]. IEEE Transactions on Information Forensics and Security, 2018,

13(5): 1333-1345.

[57] Sadeghikhorami L, Safavi A A. Secure distributed Kalman filter using partially homomorphic

encryption [J]. Journal of the Franklin Institute, 2021, 358(5): 2801-2825.

[58] Smith W A, Randall R B. Rolling element bearing diagnostics using the Case Western Reserve

University data: A benchmark study [J]. Mechanical Systems and Signal Processing, 2015, 64-

65(100-131).

[59] Amiri M M, Gündüz D. Federated learning over wireless fading channels [J]. IEEE

Transactions on Wireless Communications, 2020, 19(5): 3546-3557.

[60] Sattler F, Wiedemann S, Müller K-R, et al. Robust and communication-efficient federated

learning from non-i.i.d. data [J]. Distributed, Parallel, and Cluster Computing 2019, 31(3400-

3413).

[61] Liu W, Chen L, Chen Y, et al. Accelerating federated learning via momentum gradient descent

[J]. IEEE Transactions on Parallel and Distributed Systems, 2020, 31(8): 1754-1766.

[62] Sanghai G, Jingjing W, Chunxiao J, et al. MagicNet: the maritime giant cellular network [J].

IEEE Communications Magazine, 2021, 59(117-123).

[63] Wang M M, Zhang J, You X. Machine-Type communication for maritime internet of things: a

design [J]. IEEE Communications Surveys & Tutorials, 2020, 22(4): 2550-2585.

[64] Te W, Wei F, Yunfei C, et al. Hybrid satellite-terrestrial communication networks for the

maritime internet of things: key technologies, opportunities, and challenges [J]. IEEE Internet

of Things Journal, 2021, 8(8910-8934)