en
×

分享给微信好友或者朋友圈

使用微信“扫一扫”功能。
作者简介:

江洋(1989-),男,硕士,讲师,主要从事自主式水下机器人故障诊断研究。

通讯作者:

冯晨(1979-),女,博士,讲师,主要从事复杂系统故障诊断研究。

中图分类号:TP206+.3

文献标识码:A

文章编号:2096-5753(2024)05-0507-14

DOI:10.19838/j.issn.2096-5753.2024.05.007

参考文献 1
WANG T Q,YUAN P.Technological economic study for ocean energy development in China[J].IEEE Technology Management Council Singapore Chapter,2011:610-614.
参考文献 2
习近平.中国共产党第二十次全国代表大会上的报告[N].人民日报,2022-10-17(12).
参考文献 3
XU Q,TAN Z Y.Building a maritime community with a shared future:scholarly reflections on China’s new ocean vision[J].Marine Policy,2023,149:105508.
参考文献 4
CHEN Y H,XIAO S,LI D J.Stability analysis model for multi-node undersea observation networks[J].Simulation Modelling Practice and Theory,2019,97:101971.
参考文献 5
YUAN C R,SHUAI C G,MA J G,et al.An efficient control allocation algorithm for over-actuated AUVs trajectory tracking with fault-tolerant control[J].Ocean Engineering,2023,273:113976.
参考文献 6
YUAN C R,SHUAI C G,MA J G,et al.Adaptive optimal 3D nonlinear compound line-of-sight trajectory tracking control for over-actuated AUVs in attitude space[J].Ocean Engineering,2023,274:114056.
参考文献 7
GAO T Y,YANG J L,JIANG S D,et al.A novel fault diagnosis method for analog circuits based on conditional variational neural networks[J].Circuits,Systems,and Signal Processing,2021,40(6):2609-2633.
参考文献 8
LIU C Z,ANDRZEJ C,GRZEGORZ K,et al.Technology development and commercial applications of industrial fault diagnosis system:a review[J].The International Journal of Advanced Manufacturing Technology,2021,118(11-12):3497-3529.
参考文献 9
ABDULLAH A M,MAHATHIR M B,AYANTHA S M,et al.Multi-channel sensor fusion for real-time bearing fault diagnosis by frequency-domain multilinear principal component analysis[J].International Journal of Advanced Manufacturing Technology,2023,124(3-4):1321-1334.
参考文献 10
HASHIMOTO K,WATANABE M,TASHIRO S,et al.Missing of the ROV Kaiko vehicle-problem on the secondary cable[C]//Oceans’04 MTS/IEEE TechnoOcean’04.Kobe:IEEE,2004.
参考文献 11
ZHANG X Y,HONG J C,XU X M.Fault diagnosis of real-scenario battery systems based on modified entropy algorithms in electric vehicles[J].Journal of Energy Storage,2023,63:107079.
参考文献 12
PENG C,GAO H N,LIU X Y,et al.A visual vibration characterization method for intelligent fault diagnosis of rotating machinery[J].Mechanical Systems and Signal Processing,2023,192:110229.
参考文献 13
WANG J B,YANG B,ZENG C Y,et al.Recent advances and summarization of fault diagnosis techniques for proton exchange membrane fuel cell systems:a critical overview[J].Journal of Power Sources,2021,500:229932.
参考文献 14
KUMAR S S,MAHAPATRA D M,SING H L.Multifaceted Bio-sensing Technology[M].Salt Lake City:Academic Press,2023.
参考文献 15
WANG S H,CELEBI M E,ZHANGY D.Advances in data preprocessing for biomedical data fusion:an overview of the methods,challenges,and prospects[J].Information Fusion,2021,76:376-421.
参考文献 16
ZHONG H Y,YU S S,TRINH H,et al.Fine-tuning transfer learning based on DCGAN integrated with self-attention and spectral normalization for bearing fault diagnosis[J].Measurement,2023,210:112421.
参考文献 17
ALBAHRA S,GORBETT T,ROBERTSON SCOTT,et al.Artificial intelligence and machine learning overview in pathology & laboratory medicine:a general review of data preprocessing and basic supervised concepts[J].Seminars in Diagnostic Pathology,2023,40(2):71-87.
参考文献 18
DELISLE P L,BENOIT A R,DESROSIERS C,et al.Realistic image normalization for multi-domain segmentation[J].Medical Image Analysis,2021,74:102191.
参考文献 19
WANG J B,YANG B,ZENG C Y,et al.Recent advances and summarization of fault diagnosis techniques for proton exchange membrane fuel cell systems:a critical overview[J].Journal of Power Sources,2021,500:229932.
参考文献 20
RUAN W Y,SUN L.Robust latent discriminative adaptive graph preserving learning for image feature extraction[J].Knowledge-based Systems,2023,268:110487.
参考文献 21
GARCIA-ALVAREZ D,BREGON A,PULIDO B,et al.Integrating PCA and structural model decomposition to improve fault monitoring and diagnosis with varying operation points[J].Engineering Applications of Artificial Intelligence,2023,122:106145.
参考文献 22
SILVA R,MELO-PINTO P.T-SNE:a study on reducing the dimensionality of hyperspectral data for the regression problem of estimating oenological parameters[J].Artificial Intelligence in Agriculture,2023,7:58-68.
参考文献 23
ZHOU C,JIA Y,ZHAO S,et al.A mechanical part fault diagnosis method based on improved multiscale weighted permutation entropy and multiclass LSTSVM[J].Measurement,2023:112671.
参考文献 24
SARKAR S,PRAMANIK A,MAITI J.An integrated approach using rough set theory,ANFIS,and Z-number in occupational risk prediction[J].Engineering Applications of Artificial Intelligence,2023,117:105515.
参考文献 25
HUANG H,LI Y Y,LI W X.Dynamic reliability analysis of stochastic structures under non-stationary random excitations based on an explicit time-domain method[J].Structural Safety,2023,101:102313.
参考文献 26
HAN Y J,SONG T C,FENG J,et al.Grayscale-inversion and rotation invariant image description with sorted LBP features[J].Signal Processing:Image Communication,2021,99:116491.
参考文献 27
HAMDANI H,SEPTIARINI A,SUNYOTO A,et al.Detection of oil palm leaf disease based on color histogram and supervised classifier[J].Optik-International Journal for Light and Electron Optics,2021,245:167753.
参考文献 28
HUANG X,HAN B,NING Y Q,et al.Edge-based feature extraction module for 3D point cloud shape classification[J].Computers & Graphics,2023,112:31-39.
参考文献 29
YANG C Z,YU Q.Multiscale Fourier descriptor based on triangular features for shape retrieval[J].Signal Processing:Image Communication,2019,71:110-119.
参考文献 30
TAHER S A,LI J,FANG H Z.Simultaneous seismic input and state estimation with optimal sensor placement for building structures using incomplete acceleration measurements[J].Mechanical Systems and Signal Processing,2023,188:110047.
参考文献 31
ZENG C H,LAI W C,LIN H J,et al.Weak information extraction of gamma spectrum based on a two-dimensional wavelet transform[J].Radiation Physics and Chemistry,2023,208:110914.
参考文献 32
JIAO S B,SHI J Q,WANG Y,et al.A novel image noise reduction method for composite multistable stochastic resonance systems[J].Heliyon,2023,9(3):e14431.
参考文献 33
WEI D,FENG H D,HAN Q,et al.Fault detection and diagnosis for variable-air-volume systems using combined residual,qualitative and quantitative techniques[J].Energy & Buildings,2022,254:111491.
参考文献 34
HASSANNAYEBI E,NOURIAN R,MOUSAVI S M,et al.Predictive analytics for fault reasoning in gas flow control facility:a hybrid fuzzy theory and expert system approach[J].Journal of Loss Prevention in the Process Industries,2022,77:104796.
参考文献 35
LIU Y Q,RAMIN P,FLORES-ALSINA X,et al.Transforming data into actionable knowledge for fault detection,diagnosis and prognosis in urban wastewater systems with AI techniques:a mini-review[J].Process Safety and Environmental Protection,2023,172:501-512.
参考文献 36
MELLIT A,BENGHANEM M,KALOGIROU S,et al.An embedded system for remote monitoring and fault diagnosis of photovoltaic arrays using machine learning and the internet of things[J].Renewable Energy,2023,208:399-408.
参考文献 37
CAO H S,SUN P W,ZHAO L.PCA-SVM method with sliding window for online fault diagnosis of a small pressurized water reactor[J].Annals of Nuclear Energy,2022,171:109036.
参考文献 38
HUANG X,TENG Z S,TANG Q,et al.Fault diagnosis of automobile power seat with acoustic analysis and retrained SVM based on smartphone[J].Measurement,2022,202:111699.
参考文献 39
ZHANG H X,FENG W,ZHANG W J,et al.High anti-interference fault diagnosis method for PMSG drive system based on feature clustering optimization[J].Energy Reports,2023,9:576-582.
参考文献 40
WEN S Q,ZHANG W R,SUN Y F,et al.An enhanced principal component analysis method with Savitzky-Golay filter and clustering algorithm for sensor fault detection and diagnosis[J].Applied Energy,2023,337:120862.
参考文献 41
YUAN C R,SHUAI C G,MA J G,et al.An efficient control allocation algorithm for over-actuated AUVs trajectory tracking with fault-tolerant control[J].Ocean Engineering,2023,273:113976.
参考文献 42
ZHOU X,HUANG J Q,LU F,et al.A novel compound fault-tolerant method based on online sequential extreme learning machine with cycle reservoir for turbofan engine direct thrust control[J].Aerospace Science and Technology,2023,132:108059.
参考文献 43
LI Z Q,WANG W W,ZHANG C L,et al.Fault-tolerant control based on fractional sliding mode:crawler plant protection robot[J].Computers and Electrical Engineering,2023,105:108527.
参考文献 44
CONCHAS R F,SANCHEZ E N,RICALDE L J,et al.Sensor fault-tolerant control for a doubly fed induction generator in a smart grid[J].Engineering Applications of Artificial Intelligence,2023,117:105527.
参考文献 45
SANWALE J,DAHIYA S,TRIVEDI P,et al.Robust fault-tolerant adaptive integral dynamic sliding mode control using finite-time disturbance observer for coaxial octorotor UAVs[J].Control Engineering Practice,2023,135:105495.
参考文献 46
CHU Y H,WEI Z Q,YANG Z.New reweighted atomic norm minimization approach for line spectral estimation[J].Signal Processing,2023,206:108897.
参考文献 47
MENG Q H,ZHU S Y.Anomaly detection for construction vibration signals using unsupervised deep learning and cloud computing[J].Advanced Engineering Informatics,2023,55:101907.
参考文献 48
YAN Z P,ZHANG M Y,ZHANG C,et al.Decentralized formation trajectory tracking control of multi-AUV system with actuator saturation[J].Ocean Engineering,2022,255:111423.
参考文献 49
XIA S X,ZHOU X F,SHI H B,et al.A fault diagnosis method with multi-source data fusion based on hierarchical attention for AUV[J].Ocean Engineering,2022,266(1):112595.
参考文献 50
LI C Y,GUO S X.Characteristic evaluation via multi-sensor information fusion strategy for spherical underwater robots[J].Information Fusion,2023,95:199-214.
参考文献 51
WANG C,WANG Y H,ZHANG R F,et al.Analysis of roll center compensation method for underwater gliders based on deep learning[J].Ocean Engineering,2022,246:110529.
参考文献 52
YUAN C R,SHUAI C G,MA J G,et al.An efficient control allocation algorithm for over-actuated AUVs trajectory tracking with fault-tolerant control[J].Ocean Engineering,2023,273:113976.
参考文献 53
HACIZADE C,VURAL S Y.Active fault tolerant lateral control against actuator faults applied to AUV dynamics[C]//The 4th Conference on Control and Fault Tolerant Systems.Casablanca:IEEE,2019.
参考文献 54
WU P,HARRIS C A,SALAVASIDIS G,et al.Unsupervised anomaly detection for underwater gliders using generative adversarial networks[J].Engineering Applications of Artificial Intelligence,2021,104:104379.
参考文献 55
DAS D B,BIRANT D.GASEL:Genetic Algorithm-supported Ensemble Learning for fault detection in autonomous underwater vehicles[J].Ocean Engineering,2023,272:113844.
参考文献 56
TOPINI E,FANELLI F,TOPINI A,et al.An experimental comparison of Deep Learning strategies for AUV navigation in DVL-denied environments[J].Ocean Engineering,2023,274:114034.
参考文献 57
HUANG Z Y,SU Z Y,HUANG B.Quaternion-based finite-time fault-tolerant trajectory tracking control for autonomous underwater vehicle without unwinding[J].ISA Transactions,2022,131:15-30.
参考文献 58
YAN Z P,ZHANG M Y,ZHANG C,et al.Decentralized formation trajectory tracking control of multi-AUV system with actuator saturation[J].Ocean Engineering,2022,255:111423.
参考文献 59
LIU X X,JU Y T,LIU X H,et al.An IMU fault diagnosis and information reconstruction method based on analytical redundancy for autonomous underwater vehicle[J].IEEE Sensors Journal,2022,22(12):12127-12138.
参考文献 60
LI Z F,WANG M,MA G,et al.Adaptive reinforcement learning fault-tolerant control for AUVs with thruster faults based on the integral extended state observer[J].Ocean Engineering,2023,271:113722.
参考文献 61
WANG Q Y,LIU K Y,CAO Z Y.System noise variance matrix adaptive Kalman filter method for AUV INS/DVL navigation system[J].Ocean Engineering,2023,267:113269.
参考文献 62
SONG J H,HE X.Robust state estimation and fault detection for autonomous underwater vehicles considering hydrodynamic effects[J].Control Engineering Practice,2023,135:105497.
参考文献 63
LI X J,QIN H D,LI L Y.Fixed-time formation control for AUVs with unknown actuator faults based on lumped disturbance observer[J].Ocean Engineering,2023,269:113495.
参考文献 64
HAO L Y,ZHANG Y Q,LI H.Fault-tolerant control via integral sliding mode output feedback for unmanned marine vehicles[J].Applied Mathematics and Computation,2021,401:126078.
参考文献 65
ZHU C,HUANG B,ZHOU B,et al.Adaptive model-parameter-free fault-tolerant trajectory tracking control for autonomous underwater vehicles[J].ISA Transactions,2021,114:57-71.
参考文献 66
HAO L Y,ZHANG H,YUE W,et al.Fault-tolerant compensation control based on sliding mode technique of unmanned marine vehicles subject to unknown persistent ocean disturbances[J].International Journal of Control,Automation and Systems,2020,18(3):739-752.
参考文献 67
CHU Z Z,XIANG X B,ZHU D Q,et al.Adaptive fuzzy sliding mode diving control for autonomous underwater vehicle with input constraint[J].International Journal of Fuzzy Systems,2018,20(5):1460-1469.
参考文献 68
LIU X,ZHANG M,YAO F.Adaptive fault tolerant control and thruster fault reconstruction for autonomous underwater vehicle[J].Ocean Engineering,2018,155:10-23.
参考文献 69
BALDINI A,FASANO A,FELICETTI R,et al.A model-based active fault tolerant control scheme for a remotely operated vehicle[J].IFAC PapersOnLine,2018,51(24):798-805.
参考文献 70
CORRADINI M L,LONGHI S,MONTERIU A,et al.Observer-based fault tolerant sliding mode control for remotely operated vehicles[J].IFAC PapersOnLine,2010,43(20):173-178.
参考文献 71
ZHANG M J,LIU X,YIN B J,et al.Adaptive terminal sliding mode based thruster fault tolerant control for underwater vehicle in time-varying ocean currents[J].Journal of the Franklin Institute,2015,352(11):4935-4961.
参考文献 72
CHU Z Z,ZHANG M J.Fault reconstruction of thruster for autonomous underwater vehicle based on terminal sliding mode observer[J].Ocean Engineering,2014,88:426-434.
参考文献 73
SHUMSKY A.Data driven method for fault detection and isolation in nonlinear uncertain systems[J].IFAC PapersOnLinec,2007,7(1):283-286.
参考文献 74
FILARETOV V,ZHIRABOK A,ZUEV A,et al.The new approach for synthesis of diagnostic system for navigation sensors of underwater vehicles[J].Procedia Engineering,2014,69:822-829.
参考文献 75
ALVAREZ M D,LANE D M.A Hidden Markov Model application with Gaussian Mixture emissions for fault detection and diagnosis on a simulated AUV platform[C]//OCEANS 2016 MTS/IEEE Monterey.Monterey:IEEE,2016.
参考文献 76
FREDDI A,LONGHI S,MONTERIU A.Actuator fault detection system for a remotely operated vehicle[J].IFAC PapersOnLine,2013,46(33):356-361.
参考文献 77
CORRADINI M L,A MONTERIÙ,ORLANDO G.Actuator failure tolerant control design for underwater remotely operated vehicles[J].IFAC PapersOnLine,2010,43(16):25-30.
参考文献 78
BIAN X Q,MOU C H,YAN Z P,et al.Reliability analysis of AUV based on fuzzy fault tree[C]//2009 International Conference on Mechatronics and Automation.Changchun:IEEE,2009.
参考文献 79
MADSEN A L,KJÆRULFF U B.Applications of HUGIN to diagnosis and control of autonomous vehicles[J].Advances in Probabilistic Graphical Models,2007,213:313-332.
参考文献 80
LIU W X,WANG Y J,YIN B J,et al.Thruster fault identification based on fractal feature and multiresolution wavelet decomposition for autonomous underwater vehicle[J].Proceedings of the Institution of Mechanical Engineers Part C:Journal of Mechanical Engineering Science,2017,231(13):2528-2539.
参考文献 81
YIN B J,LIN X,TANG W X,et al.Thruster fault identification for autonomous underwater vehicle based on time-domain energy and time-frequency entropy of fusion signal[J].Intelligent Robotics and Applications,2019,11742:264-275.
参考文献 82
ZHANG M J,WU J,CHU Z Z.Multi-fault diagnosis for autonomous underwater vehicle based on fuzzy weighted support vector domain description[J].China Ocean Engineering Society,2014,28(5):599-616.
参考文献 83
NASCIMENTO S,VALDENEGRO-TORO M.Modeling and soft-fault diagnosis of underwater thrusters with Recurrent Neural Networks[J].IFAC PapersOnLine,2018,51(29):80-85.
参考文献 84
WANG Y J,ZHANG M J,WU J,et al.Research of the fault diagnosis method for the thruster of AUV based on information fusion[C]//The 3rd International Conference on Intelligent Computing.Qingdao:IEEE,2007.
参考文献 85
ZHANG Y J,ZHANG M J.Study of fuzzy neural networks model for system condition monitoring of AUV[J].Journal of Marine Science and Application,2002,1(2):42-45.
参考文献 86
LIU Q,ZHU D Q,YANG S X.Unmanned underwater vehicles fault identification and fault-tolerant control method based on FCA-CMAC neural networks,applied on an actuated vehicle[J].Journal of Intelligent & Robotic Systems,2012,66(4):463-475.
参考文献 87
SUN Y S,LI Y M,ZHANG G C,et al.Actuator fault diagnosis of autonomous underwater vehicle based on improved Elman neural network[J].Journal of Central South University,2016,23(4):808-816.
参考文献 88
LI X,SONG Y,GUO J,et al.Sensor fault diagnosis of autonomous underwater vehicle based on extreme learning machine[C]//2017 IEEE Underwater Technology.Busan:IEEE,2017.
参考文献 89
SUBHA T D,SUBASH T,JANE K,et al.Autonomous under water vehicle based on extreme learning machine for sensor fault diagnosistics[J].Materials Today:Proceedings,2020,24(2):2394-2402.
参考文献 90
WANG Y J,ZHAO J,ZHANG M J,et al.Research on the sensors condition monitoring method for AUV[J].Intelligent Robotics and Applications,2008,5314:427-436.
参考文献 91
TANG T,GANG Y.An intelligence fusion system for ship fault-tolerant control[J].IFAC PapersOnLine,2004,37(10):83-87.
参考文献 92
QIN X C,ZHANG W,GAO S,et al.Sensor fault diagnosis of autonomous underwater vehicle based on LSTM[C]//The 37th Chinese Control Conference.Wuhan:IEEE,2018.
参考文献 93
SHI C T,ZHANG R B.Software fault diagnosis model of AUV based on Bayesian Networks and its simplified method[C]//The 9th World Congress on Intelligent Control and Automation.Taipei:IEEE,2011.
参考文献 94
OMERDIC E,ROBERTS G.Thruster fault diagnosis and accommodation for open-frame underwater vehicles[J].Control Engineering Practice,2004,12(12):1575-1598.
参考文献 95
ZHENG X,FENG C,LI T,et al.Analysis of autonomous underwater vehicle navigational states based on complex networks[J].Ocean Engineering,2019,187:106141.
参考文献 96
LIU W X,WANG Y J,LIU X,et al.Weak thruster fault detection for AUV based on stochastic resonance and wavelet reconstruction[J].Journal of Central South University,2016,23(11):2883-2895.
参考文献 97
WANG Y J,ZHANG M J,WILSON P A,et al.Adaptive neural network-based backstepping fault tolerant control for underwater vehicles with thruster fault[J].Ocean Engineering,2015,110:15-24.
参考文献 98
ZHANG M J,WANG Y J,XU J A,et al.Thruster fault diagnosis in autonomous underwater vehicle based on grey qualitative simulation[J].Ocean Engineering,2015,105:247-255.
目录contents

    摘要

    自主式水下机器人(Autonomous Underwater Vehicle,AUV)无人无缆运行在复杂的海洋环境中。 AUV 装备朝着大型化、高速化、精密化、系统化和自动化的方向发展。AUV 的安全、平稳以及长时间运行是需要研究的重点问题,推进器作为 AUV 的核心部件和负载最重的部件,其故障诊断技术对于保障 AUV 安全运行起到极为重要的作用。推进器故障特征微弱且易受到海流等外部随机特征的干扰,随着设备检测传感器与采样频率的增加,有利于建立新理论和共同认可的推进器故障诊断办法。研究推进器的故障诊断理论和方法,特别是开展推进器相关的故障识别和容错控制,能够有效增强 AUV 的运行可靠性,促进 AUV 实用化进程,相关研究的理论价值和工程应用价值非常大。

    Abstract

    The autonomous underwater vehicle operates in a complex marine environment without a cable. AUV equipment is developing towards large-scale,high-speed,precision,systematization and automation. The safety,stability and long-term operation of the AUV are the key issues that need to be studied. AUV thruster is the core and the heaviest component of the AUV. Fault diagnosis technology of the thruster plays an extremely important role in ensuring AUV safe operation. Fault features of the thruster are weak and vulnerable to the interference of external random features such as ocean currents. With the increase of equipment detection sensors and sampling frequency,it is conducive to the establishment of new theories and commonly recognized solutions for thruster fault diagnosis. Research on fault diagnosis theory and method of thrusters,especially on fault identification and fault tolerance control related to thrusters,can effectively enhance the reliability of AUV and promote the practical process of AUV. The theoretical value and engineering application value of relevant research are very great.

  • 0 引言

  • 地球上的陆地和海洋总面积约 510 000 000 km2,其中海洋面积约 361 000 000 km2,占地球总面积的 71%,如何保护、开发海洋,成为 21 世纪人类关注的重要问题[1]。海洋是生命的源泉、资源的宝库、未来的希望。二十大报告中指出:我国在深海、深地探测等取得重大成果,进入创新型国家行列[2]。维护我国 300 多万平方千米的海洋权益,保护蓝色国土的安全,建立海洋新秩序,促进海洋资源公平、有效利用的国际责任,均要求我国深度参与全球海洋治理、有效维护和拓展国家海洋权益[3]。为满足建设海洋强国、推进实施“21 世纪海上丝绸之路” 的迫切需求,以《中国制造 2025》为指导思想,顺应“人工智能+”发展趋势,需要大力加强海洋信息化建设,响应海洋物联网信息化建设。我国所建设的海底科学观测网是人类建立的第三种地球科学观测平台,海底观测网旨在将水下传感器、无人潜航器、空中无人机和卫星等以“传感器”的形式构建节点网络,该网络能覆盖大面积的海域,实现了海洋信息智能采集和智能应用的互联、集成和共享。完善的海底观测网系统将海上信息资源整合,实现对海上繁杂数据的监测和系统化管理,为各种海洋经济活动提供安全保障[4]

  • 自主式水下机器人未来有可能成为海底观测网和海洋信息化建设的重要节点,同时也是开发海洋、保护海洋、管控海洋的核心平台,在海洋军事应用中也扮演着重要角色,被誉为未来海军力量倍增器。AUV 技术正向着更大航程、更大深度、更高航速、更高智能方向发展。在未来诸多海洋应用中,AUV 长期自主执行任务、自主维护也将成为必然趋势。目前,影响其广泛应用和推广的主要因素应该是自主生存能力,故障诊断对于提高自主生存能力具有举足轻重的地位。绝大多数用户,无论是民用或军用,都需要对 AUV 状态进行实时监测与检测,经统计,AUV 缠绕(渔网、绳索等杂物)、碰撞、气蚀已经逐渐成为 AUV 损坏或者丢失的主要原因。障碍物感知、自主规避,以及缠绕物摆脱技术直接关系到 AUV 生存安全,也决定着 AUV 自主生存能力水平的高低。因此,提高 AUV 的自主检测能力至关重要。

  • 1 故障诊断技术概述

  • 随着工业生产和信息化技术的快速发展,设备的结构和智能化日益完善,功能越来越强大,自动化程度也越来越高,可靠性、可用性、维护与管理、安全性等问题亟待解决[5-6]

  • 由于多种不可抗力因素的影响,各种各样的故障都有可能在设备中出现,国外曾经发生各种空难、丢失、爆炸、泄漏、燃烧等重大事故,造成人员的伤亡[7-8]。美国、日本等其它国家曾经发生过多次重大灾难性事故,造成了严重的人员伤亡和环境污染[9]。例如,2011 年,日本附近海域发生 9.0 级地震引发海啸,福岛第一核电站氢气爆炸,导致放射性物质泄漏。虽然事故已经过去 10 多年了,但是对于当地生态环境、人们健康和社会经济都产生了长期的严重不良影响。美国也发生过多次航天事故,1986 年挑战者号航天飞机事故,升空后右侧固体火箭助推器的 O 形环密封圈失效。1981 年哥伦比亚号航天飞机事故,事故原因至今仍然没有定论,疑似因航天飞机隔热瓦部缺损而引起事故。日本“海沟号”的神秘消失,无人驾驶潜艇坠入海底深处,通信设备损坏[10]。国内也发生过不少重大事故,2022 年 3 月 21 日,东方航空云南有限公司波音 737-800 型客机,升空后第 64 min 偏离巡航高度 8 900 m 并快速下降,坠毁于广西壮族自治区梧州市藤县埌南镇莫埌村附近。2011 年 7 月 23 日,温州南站沿线铁路牵引供电接触网或附近大地遭受雷击,LKD2-T1 型列控中心设备采集驱动单元采集电路电源回路中的保险管 F2 熔断。列控中心设备仍按照熔断前无车占用状态进行控制输出,致使温州南站列控中心设备控制的区间信号机故障,始终保持绿灯状态,造成事故的发生。2003 年 4 月 24 日海军 361 号潜艇事故,发电机的通气管阀门因故障未能开启,导致外界的氧气无法进入发电机。柴油发电机在整个工作过程中,使用的都是潜艇内的氧气,导致潜艇内严重缺氧造成严重的事故。

  • 为保证设备安全平稳运行,无数科研工作人员加入故障诊断的研究领域中来,相关文献中关于故障诊断的研究主题分布如图1 所示。故障诊断对于保障设备的安全运行起到关键作用,通过各种检查和测试方法,获取设备或系统的运行状态信息与模式,为进一步的维护与保养、任务决策、故障定位提供信息支撑[11-12],故障诊断能够有效地提高设备的使用寿命,防患于未然。航空航天、机械工程、控制与决策、电子信息、交通运输等多个方面早已开始故障诊断的研究。

  • 图1 相关文献中关于故障诊断研究主题分布

  • Fig.1 Distribution of research topics on fault diagnosis in relevant literature

  • 从相关文献中故障诊断研究的主题分布能够看出,多数研究围绕故障诊断方法开展,故障检测、故障诊断、故障定位是研究的重点和热点,故障诊断系统、介质、电子设备、发动机、物联网、传感器、风电机组、配电网、服务器、监控系统、控制装置等设备和装置是故障诊断研究的硬件基础。

  • 从故障诊断的过程来看,诊断系统一般由信号采集、数据处理、特征提取、状态识别和容错控制 5 个部分构成[13]。信号采集模块主要通过传感器等手段来采集数据,传感器将某些变化过程的物理量转换成电信号等形式输出以供分析[14]

  • 数据处理模块因不同评价指标往往具有不同的量纲或量纲单位,需要对数据进行标准化处理[15]。数据处理模块主要功能是数据的标准化,数据的标准化是将数据规则缩放,使之集中在小的固定区间[16-17]。在一些需要进行比较评价的指标时,需要去除数据的单位限制[18],将其转化为无量纲的纯数值,便于不同单位或量级的指标能够进行比较和加权[19]

  • 特征提取方法很多,就涉及的数据类型而言可以分为 4 类[20]

  • 1)基于变量间信息的特征提取方法包括:主分量分析法及其改进算法、独立分量分析法及其改进算法、奇异值分解法、主成分分析[21]、多维尺度变换、等距映射算法、局部线性嵌入、T 分布随机邻域嵌入 [22]、线性判别式分析、自编码器 (AutoEncoder,AE)、拉普拉斯特征映射等。

  • 2)基于变量与模式间信息的特征提取方法。特征与模式之间的响应是 2 个不同的概念,但其中特征对模型具有较大影响,则该特征可以认为是主特征量。根据相关原则把有关特征量选择出来,形成新的数据组,进行数据降维,完成特征提取。该类方法主要有基于神经网络技术、基于互信息熵[23]、基于模糊理论、基于粗糙集理论[24] 等技术的特征选择方法。

  • 3)监测设备给出一系列设备振动信息频谱,采用一定的算法将信号进行分解,进行谐波分析。频谱型故障特征提取方法有:信号的各种频域、时域分析方法[25]

  • 4)图像处理的特征提取方法。随着新一代计算机技术的发展,已有将图像特征提取的方法应用到故障诊断中。灰度特征提取方法主要有:灰度直方图[26]、颜色直方图[27]、灰度平均值、均值、方差等。常用的提取纹理特征的方法有:灰度差分统计法、灰度共生矩阵、灰度–梯度共生矩阵、Tamura 纹理特征、自回归纹理特征、小波变换等方法。形状特征提取方法有:不变矩特征尺度[28]、旋转、平移不变性、傅里叶形状描述符[29]、小波描述符等。

  • 故障诊断状态识别方法可以分为解析模型、信号处理、知识 3 个方面,基于解析模型的故障诊断方法有参数估计、状态估计[30](观测器法、滤波器法)、等价空间等;基于信号处理的故障诊断方法有小波变换[31]、谱分析、变分模态分解、经验模式分解、形态信号处理、随机共振[32]、滤波等;基于知识的故障诊断方法有图论方法(符号有向图、故障树、复杂网络、贝叶斯网络、马尔可夫链等)、定性仿真[33]、专家系统[34]、人工智能[35-36]、粗糙集、神经网络(循环神经网络(Recurrent Neural Network,RNN)、卷积神经网络(Convolutional Neural Network,CNN)、生成对抗网络,BP 神经网络(Back Propagation,BP)、长短记忆网络(Long Short Term Memory Networks,LSTM))等网络、支持向量机[37-38]、支持向量域描述、信息融合、统计分析方法(多元统计分析(聚类、主成分析)、时间序列分析、灰色理论法(灰色预测)、残差分析、格兰杰因果)等[39-40]

  • 容错控制主要作用是根据检测的故障数据,针对不同的故障原因和故障特征,在仪器即将发生故障或者发生故障以后,采取相应的容错控制策略,保证仪器的正常工作,或者以最小的性能损失为代价[41-42],保证设备在要求的时段内完成基本的功能,并具有能够允许的性能指标[43-44]

  • 容错控制方法有神经网络信息融合、硬件冗余、有限脉冲响应滤波器、传感器信号替代、控制率离线设计在线调度、滑模容错控制[45]、最小化 l-范数[46]、控制率伪逆重构、小脑神经网络等方法。随着大数据、云计算[47]、深度学习、信号处理、动力学等技术的快速发展,也有较多的新方法应用到信号采集、数据处理、特征提取、状态识别和容错控制相关模块中。

  • 2 AUV 故障诊断技术研究现状和发展趋势

  • 2.1 AUV 故障诊断技术概述

  • AUV 通常在复杂多变的海况中运行,在无人无缆情况下加之通信时滞性大,保证 AUV 顺利完成任务成为学者们研究的热点。查阅相关文献资料故障诊断主要从动态系统故障诊断的定性和定量角度进行分析。大多数 AUV 强故障都能获得有效的诊断,但针对设备故障概率和位置以及弱故障诊断研究相对较少。主要原因是:强故障特征较为明显,故障诊断相对容易,但 AUV 本身及其环境是一个复杂的动态系统,外部干扰会引起 AUV 传感器与实际状态值的短期偏差,使故障特征被掩盖,如何提取描述无关紧要的时域和频域符号的行为特征,以及建立基于预先的知识库进行故障诊断,难以满足未知类型的故障诊断。数据驱动方法的故障诊断不完全依赖于精确的模型,在外部干扰使数据扰动时仍能够适用, AUV 故障诊断大多集中在模块级,较少对系统健康管理及状态监控开展研究。

  • 2.2 AUV 故障诊断技术国内外研究现状和发展趋势

  • 总结国内外近几年研究现状和AUV故障诊断资料,将 AUV 故障诊断方法分为解析模型、信号处理、知识 3 个方面(如图2 所示),分别从这 3 个方面介绍国内外 AUV 故障诊断方法的研究现状。

  • 图2 AUV 故障诊断常用方法分类

  • Fig.2 Classification of common methods for AUV fault diagnosis

  • 基于信号处理的 AUV 故障诊断方法如图3 所示:主要集中在信息融合、信号处理 2 个方面,随着可用数据越来越多,信息融合成为 AUV 故障诊断研究的热点,YAN 研究了复杂海洋环境下多种自主水下航行器系统的三维形成跟踪问题。设计了一种基于分散估计和控制的观察器–控制器方案。为每个 AUV 开发一个观察者来估计系统的整体状态,然后将其与集体任务变量集成,以产生控制输入,它使多 AUV 系统能够执行依赖于全局信息的编队轨迹跟踪任务。在设计阶段,利用双曲切函数降低了执行器饱和的风险,并解决了系统的测量噪声和非线性问题。其次,在连通无向拓扑结构下,以及在不饱和和饱和控制输入的情况下,基于李雅普诺夫方法的稳定性分析证明了估计误差和任务误差的收敛[48]

  • 图3 基于信号处理的 AUV 故障诊断方法

  • Fig.3 AUV fault diagnosis method based on signal processing

  • XIA 等研究人员提出了一个通用的数据层次结构。在此基础上,提出了一种基于层次注意的多源数据融合故障诊断方法。该方法由编码–解码器网络、叠加编码器和注意机制的融合网络以及基于注意分布的故障识别方法组成。融合网络利用编码器和层次注意来提取深度特征,并分层融合特征。采用多层注意分布来解释故障评估,实现了故障识别。设计了一种针对冗余性的随机掩模融合策略,并提出了一种针对多数据源间强耦合的特征正交化方法。根据南海试航获得的“潜龙”2 号 AUV 监测数据,验证了该方法有效[49]

  • 目前,现有的融合方法大多忽略了水下未知环境中不同类型传感器节点丰富的多源信息,这使得自主水下航行器准确感知外部环境并做出可操作的决策具有挑战性。考虑到 AUV 任务中涉及的姿态估计、定位和避障等关键问题,LI 和 GUO 提出了一种基于各种低成本传感器的球形水下机器人多源信息融合模型。来自惯性测量单元、压力传感器阵列、避障传感器阵列、深度传感器、下行摄像机和声通信系统的多源信息融合,使球形水下机器人能够获得姿态估计、定位和避障等高精度的估计数据。更准确地说,根据传感器之间的相关性,建立了角度误差补偿、速度误差、方向误差等优化模型。随后,提出了一种基于反向传播神经网络的机器学习方法,通过特征选择、数据训练和特征估计等方法来提高多源信息融合模型的精度和有效性。最后,在不同的场景下进行了一系列的实验,如运动实验和避障实验。理论推导和综合评价证明了该模型的有效性和可行性,为解决球形水下机器人的姿态估计、定位和避障等问题提供了新的参考方法[50]

  • 信号处理在谱分析、小波变换、信号滤波、信号参数估计等方向取得不少研究成果,WANG 等建立了一个精确的滚转调节单元,并利用中国“海燕”–1 型滑翔机在南海收集的试验数据来验证新的滚动中心补偿方法的准确性。为了使滚转调节单元的工作频率最小化,提出了一种基于变分模态分解和长短期记忆的滚动中心补偿方法[51]。YUAN 和其他 3位研究人员介绍了控制典型方向舵丢失、卡住或变形的容错控制方法来重构控制输入,提出了一种基于滑模观测器的自诊断和容错控制方案[52]。HACIZADE 提出了一种基于参数识别的自主水下航行器动力学容错控制方法。使用卡尔曼滤波器来识别与故障执行器相对应的控制分布矩阵元件,使用该被识别出的控制分布矩阵进行控制重构。为模型 AUV 设计了一个线性二次调节器作为控制器。卡尔曼滤波器识别执行器中的故障,线性二次调节器控制器为已识别的模型重新配置。在仿真中,考虑了 AUV 转向子系统动力学的线性化模型,并研究了所提出的执行器故障识别和可重构控制技术的性能[53]

  • 谱分析是一种将模态分析结果和已知谱联系起来的计算结构响应的分析方法,可分为时间–历程分析和频域的谱分析。WU 和其他研究人员提出了一种无监督异常检测系统,采用双向生成对抗网络,其时间序列数据由多个传感器收集。在这项研究中,水下滑翔机舰队的异常检测系统在 2 个健康的部署数据集上进行训练,并在由选择的航行器收集的其它 9 个部署数据集上进行测试。该系统已成功的应用于检测 9 个测试部署中的异常,其中包括几种不同类型的异常和健康行为。此外,对数据抽取设置的灵敏度研究表明,该系统对水下滑翔机的近实时异常检测具有鲁棒性[54]

  • 还有通过信息融合的方法进行故障诊断的, DAS 提出了一种结合遗传算法和集成学习的方法,用于 AUV 的故障检测。在特征选择阶段使用遗传算法提取了 97.7%的特征。提出了一种有效的通过去除冗余和不相关的数据来解决这一问题的方法,降低计算复杂度,从而更好地理解数据和学习模型。结果表明:该方法以 99.7%的准确率成功预测了 AUV 的故障状态,故障类型检测(严重螺旋桨故障、轻微螺旋桨故障、负载增加故障、传感器故障)准确率为 98.96%[55]。TOPINI 等通过深度神经网络来估计航行器的自身框架速度,而无需在导航策略中使用多普勒计程仪(Doppler Velocity Log,DVL)[56]

  • 基于解析模型的 AUV 故障诊断主要有参数估计、状态估计、等价空间 3 个方面,如图4 所示。参数估计是从总体中抽取的随机样本来估计总体分布中未知参数,HUANG 讨论了自主水下航行器遇到模型不确定性、海洋扰动和执行器故障的有限时间展开轨迹跟踪控制问题。首先,平移和旋转运动的模型是用单位四元数代替经典的欧拉角来描述的,这样 AUV 的动力学就可以全局和唯一地表述。随后,提出了 2 种有限时间控制策略,使 AUV 的状态变量可以收敛到一个可调区域。此外,利用被动容错技术,自适应定律可以适用于执行器故障的存在。通过将标量四元数的初始值积分到滑模表面,使所提出的控制器具有反解卷特性。然后,利用双曲切函数,得到了无奇异性的跟踪误差的有限时间稳定特点。利用李亚普诺夫定理验证了闭环系统的稳定性。最后,通过数值模拟验证了该方法的有效性[57]

  • 图4 基于解析模型的 AUV 故障诊断

  • Fig.4 AUV fault diagnosis based on analytical model

  • 状态估计方法中卡尔曼滤波和滑模观测器应用到 AUV 故障诊断中的较多,由于 Kalman 滤波对故障检测残差及模型的残差较为敏感,属于分散式结构,具有工程实现简单、故障容错能力强等优点,便于各子系统的故障诊断和隔离。因此,通过调整航行器的模型以及滤波器参数的非线性优化,能够在一定的区间内检测到航行器舵片的故障[58]。利用不同的辅助传感器组估计同一状态向量的卡尔曼滤波器组,增加系统中的冗余,使其更容易检测和隔离故障[59]。文献提出了一种针对有推进器故障的自主水下航行器轨迹跟踪的强化学习容错控制方法[60]。WANG 提出了一种用于 AUV 惯性导航系统(Inertial Navigation,INS)/DVL 导航系统的自适应系统噪声方差矩阵卡尔曼滤波方法[61]。 SONG 和 HE 将外部扰动建模为具有与噪声相同特征的随机模型。还应用了一种改进的鲁棒滤波方法来估计系统的状态,并减轻了随机系统不确定性的影响。基于所提出的滤波方法,基于观测值的重构误差计算残差信号,并利用滑动窗口技术进行平滑处理,直接应用于故障检测问题。最后,仿真实验证明了该方法在几种情况下检测执行器故障的有效[62]。LI 研究了在模型不确定性、洋流扰动和未知执行器故障情况下,多 AUV 系统的自适应固定时间的层控制。为了提高系统的鲁棒性,设计了一个自适应固定时间积分滑动模扰动观测器来估计由模型不确定性、洋流扰动和未知执行器故障组成的集。基于设计的观测器,结合非线性命令滤波器和反步技术,设计了自适应固定时间容错控制器,保证了轨迹跟踪误差和相应的虚拟领导者在固定时间内收敛到原点。与传统的有限时间控制相比,闭环系统的收敛时间与初始状态无关。仿真结果表明该控制算法具有较好的控制性能[63]

  • 滑模观测器可依据系统外部变量的实际测量值获得状态变量估计值。HAO 等 3 位研究人员研究了无人航行器推进器故障容错控制问题。主要的方法是首次基于积分滑模控制技术,其中只使用无人驾驶水上航行器的可测量输出。利用矩阵全秩分解和自适应机制,构造了输出积分滑动面,设计了具有全阶补偿器的积分滑动模控制器,以衰减推进器故障和海洋外部干扰作用下偏航角和偏航速度误差的振荡振幅。通过一种典型的浮式船,仿真实例验证了该方法的有效性[64]。ZHU 和其他 4 位研究者,针对自主水下航行器受外界干扰和执行器故障的轨迹跟踪问题,提出了一种无模型参数的控制策略。构建了 2 种控制架构,使系统状态可以被强制执行到具有可接受的性能的期望轨迹。通过将滑模控制技术和自适应算法相结合,开发了第 1 个用于健康执行器下跟踪任务的控制架构。考虑到执行器故障的情况,通过在第 2 控制器中使用无源容错技术,大大提高了系统的可靠性。由于欧拉–拉格朗日系统的特性,可以正确地处理水下航行器的非线性动力学,从而使所提出的控制器可以在没有模型参数的情况下建立起来。最后,通过理论分析和数值仿真,验证了该控制器的有效性[65]。HAO 研究了一种滑模技术的鲁棒自适应容错补偿控制方法,建立了包括部分卡死、全卡死和时变卡死在内的通用推力器故障模型对推力器结构矩阵进行全秩分解,在此基础上构造线性滑动面,并在滑模趋近律中引入自适应机制[66]

  • CHU 考虑到动态模型的建模不确定性,提出了一种基于模糊逻辑系统的自适应滑模控制器[67]。 LIU 针对容错控制框架下的故障重构问题,构造了一个改进的二阶滑模观测器来估计推力器的故障影响。该观测器的反馈由不同的估计误差函数组成,包括分数函数、特征函数和积分函数等,并基于李雅普诺夫函数(Lyapunov)理论对其稳定性进行了分析。最后,针对不同的期望轨迹和不同类型的推进器故障进行了仿真[68]。BALDINI 用滑模观测器产生 3 个残差,该方案可适用于航行器动力学的任何控制律,并用动力学的反演控制器和实际参数进行了仿真试验[69]。CORRADINI 将输入解耦非线性状态变换应用于水下机器人模型,利用变结构系统观测器实现故障检测,同时采用基于滑模的方法进行故障隔离和控制[70]。张铭钧在终端滑模中引入自适应策略,在线估计集中不确定性的上界,包括海流扰动和建模不确定性,以及推力器故障引起的推力器分配增益的变化。该方法的最大优点是不需要集中不确定性的先验知识,且不依赖于故障检测与诊断模块。基于 Lyapunov 理论和芭芭拉引理 (Barbalat),该方法能够适应推进器故障,并保证跟踪误差的有限时间稳定性。此外,针对抖振现象,提出了一种基于分数功率的连续开关项来代替不连续切换项。在所提出的抖振抑制方法中,切换项的连续性是基于分数次幂的,而切换项的增益是基于滑模函数的欧氏范数的指数形式来更新的[71]。CHU 基于径向基函数网络(RBF)的 AUV 运动建模方法并提出了一种终端滑模观测器,以保证每个估计的状态信号在有限时间内收敛。根据终端滑模观测器的输出,采用等效输出注入法重构推力器故障[72]

  • SHUMSKY 提出了一种新的基于右手构造多项式方程组基的方法,该方法将初始系统模型非线性转化为规则的反馈,用于自主式水下机器人运动模型的故障检测与隔离上[73]。FILARETOV 利用水下机器人的运动学模型和传感器信号的数据融合进行水下航行器导航传感器诊断[74]。DE 介绍了隐马尔可夫模型在 AUV 推进器系统仿真,使用期望最大化算法学习高斯混合模型,利用混合高斯模型分类器进行诊断[75]。FREDDI 建立了水下航行器的动力学模型,测量摇摆、喘振和偏航(及其速度) 作为输入信号,形成非线性观测器的残差产生和基于序列变化检测算法的残差评估诊断系统[76]。 CORRADINI在将输入解耦非线性状态变换应用于水下机器人模型后,检测由残差产生模块负责,而基于滑动模态的方法则用于机器人控制和故障检测。最后,利用执行器的固有冗余进行控制重构[77]

  • 基于知识的 AUV 故障诊断,大致可以分为定性的诊断方法、人工智能、统计分析 3 个方面,如图5 所示。定性故障诊断可以分为图论、定性仿真、专家系统。图论中故障树较多的应用在 AUV 故障诊断中,BIAN 建立了水下机器人的模糊故障树,利用操作规则计算了 AUV 不能正常工作的顶层事件的模糊概率[78]。MADSEN 为了处理不确定性,利用有限记忆影响图来表示和解决自主式水下机器人诊断与控制的复杂问题[79]

  • 图5 基于知识的 AUV 故障诊断

  • Fig.5 Knowledge-based AUV fault diagnosis

  • 人工智能中的支持向量机方法也较多应用在 AUV 故障诊断中,相比于时域、频域故障诊断方法[80],人工智能方法可不考虑模型的中间处理过程而实现故障检测,YIN 针对基于单一故障特征的推力器故障识别中分类精度不高的问题,提出了一种基于融合信号时域能量和时频熵的故障识别算法。首先将浪涌速度和控制电压 2 个单一方面的故障奇异信号融合成融合信号,以更全面地反映故障信息。然后,在时域提取融合信号的峰值区域能量特征,在时频域提取融合信号的熵特征,得到多域故障特征。最后,基于支持向量数据描述算法,建立了多分类器,并计算了故障样本与多分类器中各超球体之间的相对距离。故障样本对应的故障严重程度由最小相对距离确定。水池实验样机的实验结果表明,该方法的分类精度为 95.2%。与浪涌速度和控制电压对应的分类模型相比,分类精度分别提高了 5.2%和 22.8%[81]。王玉甲在基于支持向量域描述(Support Vector Data Description,SVDD)的自主式水下机器人推进器故障诊断方法研究中,传统的参数优化方法很难得到 SVDD 分类模型的最优核函数参数。针对这些问题,研究了一种改进的 SVDD 故障模式分类方法。张铭钧研究了水下机器人推进器和传感器的多故障诊断问题,传统的支持向量域描述在 AUV 多故障模式分类过程中,由于样本稀疏密度和样本分布不均匀等因素的影响,分类精度较低。该方法在故障诊断过程中,为各推力器和传感器增加故障检测面,以隔离故障部件[82]

  • NASCIMENTO 利用测量的控制输入、电压、转速和电流信号对推力器的标称特性进行 RNN 建模[83]。王玉甲改进了动态递归神经网络和基于径向基函数网络,通过将运动状态模型的输出值与实际速度和角度的实测值进行比较,得到了整体故障信息[84]。文献提出了一种用于系统状态监测的模糊神经网络结构等效模型,其输出为系统某些部分的状态或故障程度。该网络由 6 层神经元组成,分别表示隶属函数、模糊规则和输出。通过离线学习得到结构参数和权值,并根据经验推导出模糊规则[85]。除此之外,还采用径向基函数神经网络对一般不确定性进行逼近。根据 Lyapunov 理论,推导了径向基神经网络的控制律和自适应律,以保证跟踪误差渐近收敛于零。LIU 提出了一种新的无人水下航行器推进器故障诊断与调节方法。采用基于信号分配的模糊小脑模型关节控制器神经网络实现了对推进器连续和不确定干扰故障情况的故障识别。采用基于加权伪逆的重构算法求解控制分配问题的有效解[86]。孙玉山也提出了一种改进的动态递归神经网络[87]

  • 另一方面用于 AUV 传感器故障诊断的部分神经网络,LI 建立了极限学习机(Extreme Learning Machine,ELM)预测模型用于 AUV 传感器故障诊断[88],SUBHA 将相空间重构与 ELM 相结合,用于预测水下机器人的传感器输出[89]。WANG 对每个传感器都用径向基网络辨识网络,通过比较实际输出和模型输出在规定时间内的离散度和经验阈值,可以检测出被检测传感器发生的故障[90]。 TANG 提出了一种前馈神经网络与人工神经网络混合神经网络的智能融合系统,用于系统故障检测和容错控制[91]。QIN 提出了一种基于预测模型的长短期记忆网络故障诊断方法。首先,利用 AUV 实验数据训练 LSTM 网络,如加速导航、改变导航深度、改变导航方向和改变导航速度。然后利用训练好的网络对传感器系统进行拟合,建立水下机器人的运动模型[92]

  • 基于统计分析的方法也广泛地应用在 AUV 故障诊断中,SHI 提出了一种基于贝叶斯网络的故障诊断软件。在此基础上,根据 AUV 的特点,提出了一种基于任务上下文的不相关节点切割方法,该方法简化了网络,降低了后果计算的复杂性,有效地提高了实时故障诊断的能力[93]。OMERDIC 介绍了一种新型的开架式水下航行器推进器故障诊断与调节系统[94]。为了确定自主式水下机器人的航行状态,郑旭提出了一种基于复杂网络的 AUV 导航数据分析方法[95]。刘维新提出了一种基于单稳态随机共振的改善缺陷性能的方法。该方法通过将常规的双稳态电势函数更改为更适合非周期性信号的单稳态电势函数,改善了弱推力器误差特性的性能。使用传统的小波重构方法检测弱推力器误差可能会导致无法检测到的误差或错误警报。为了成功地检测出弱推力器的缺陷,提出了一种将随机共振与小波重构相结合的弱推力器检测方法[96]。王玉甲针对水下航行器存在建模不确定性、外部扰动和未知推力器故障的情况,提出了一种推力器容错控制方法,将推进器故障作为一般不确定性的一部分,并结合建模不确定性和外部扰动,采用径向基函数神经网络逼近不确定性。根据李雅普诺夫理论,推导了径向基函数网络的控制律和自适应律,保证了跟踪误差渐近收敛到零[97]

  • AUV 自身及所在环境是一个非线性动态系统,外部干扰会使 AUV 传感器短时偏离实际状态值,使得故障特征被掩盖,定性分析较难从不显著的时域和频域征兆中提取描述定性行为的知识,并且定性分析方法需要建立基于预先建立的定性知识库进行 AUV 故障诊断,难以满足未知类型的故障诊断。定量分析的方法不依赖于精确的模型,更适合外部干扰下的故障诊断[98]。本文以故障诊断方法为综述脉络对推进器的故障诊断方法和技术进行梳理和总结,能够得出已应用在 AUV 推进器中的故障诊断方法有强化学习、滑模控制技术、李雅普诺夫函数、隐马尔可夫模型、支持向量域描述、模糊小脑模型、径向基函数神经网络、高斯混合模型以及其它方法,仍有很多故障诊断方法可用于 AUV 推进器故障诊断,例如直接对信号分解、信号统计等方法,诊断出故障后的容错控制也是非常值得研究的内容。AUV 故障诊断亟待解决的问题是故障数据相对较少、特征微弱、故障难定位等问题。

  • 能够总结出对 AUV 推进器故障诊断的文献相对较少,基于信号处理推进器故障的诊断方法不需要建立精确的系统模型,信息融合方法适合于多参数、多信息的推进器故障诊断,在复杂的系统中也可以实现,但此类方法在大尺度数据环境下可能会发生延迟现象。信号方法可将信号分解为多维,挖掘出更多的特征信息,但转换过程中部分有用信息可能被淹没。

  • 基于解析模型 AUV 推进器的诊断方法能有效并实时地判断传感器数据异常情况。然而,基于解析模型的 AUV 故障诊断也存在一些缺点。解析模型方法不能检测出系统故障的具体原因。解析模型方法可能受到模型复杂性和准确性的限制,特别是在处理非线性和不确定性问题时。此外,对于未知和复杂的水下环境,这种方法可能无法提供足够的可扩展性和普适性。

  • 基于知识 AUV 推进器的诊断方法在数据分类和信息预测方面有独特的优势,同时还需要已知系统的精确数学模型,这就进一步缩小了此方法的应用范围。部分方法也需依赖于训练数据集质量以及神经网络训练精度。

  • AUV 推进器故障诊断的发展趋势主要集中在提高诊断的准确性、可靠性和减少诊断时间,同时采用 3 种模型方法建立基于预先的知识库进行故障诊断和数据驱动方法的故障诊断方法,实现 AUV 推进器故障诊断大模型。

  • 3 结束语

  • 本文根据动态系统的故障诊断理论,结合 AUV 故障诊断技术的发展现状和发展趋势,分析并归纳目前 AUV 故障诊断技术尚需解决的主要问题,阐述如下。

  • 1)AUV 推进器多传感器、多信息、多参数融合故障诊断。

  • 随着 AUV 系统变得越来越复杂,系统状态信号呈现出非线性、时变和不确定性等特点。当系统发生故障时,往往表现出多种症状,仅依靠单一的理论方法和信息很难对系统做出准确的判断,从而导致误报、漏报和其它现象的高概率。AUV 故障诊断多停留在数据级的融合,对于特征级融合研究较少,以及采用什么方法从推进器电流、推进器电压、推进器转速、AUV 航向、AUV 航速、电池电压、电池电流等数据中选取有用信息也是非常需要研究的。对于特征级融合的研究重点需要考虑提取哪些典型特征作为推进器的故障特征。

  • 2)AUV 推进器的弱故障诊断问题。

  • 尽快发现系统故障对推进器的快速诊断和应急处理具有非常重要的作用,特别在 AUV 推进器的早期失效,其特性相对较弱,对系统的影响较小,维护成本较低。如果可以及早诊断,将有助于设备的安全平稳运行,但是,在实际的 AUV 运行过程中,大多数工作环境都具有很强的噪声,并且容易受到洋流、工作深度和水密度等环境干扰的影响。而且,AUV 的各个组件的输出信号存在彼此叠加的现象,使得难以提取有效特征。因此,相对先进的微弱信号检测技术的研究在提高 AUV 推进器故障诊断的准确性方面起着非常重要的作用。

  • 3)AUV 推进器故障数据少、小样本、自适应问题。

  • 大多数深度学习的故障诊断模型有效的前提是训练集和测试集的数据分布相同或接近且目标数据样本量充足。然而,在实际的应用中,AUV 存在工况环境复杂、运行状态多变以及信号难以直接测量的问题。例如,在 AUV 的整个生命周期中,大多数情况下都处于正常工作的状态,有时只能采集有限的故障数据,甚至没有;同一型号的 AUV 在不同的运行状态下,变负载、变转速的数据分布也是不同的。这些往往导致训练集数据和测试集数据的分布存在差异,进而导致已训练的模型泛化能力低,甚至不再适用。如何在小样本、变负载下自适应实现 AUV 状态的准确识别是目前亟待解决的问题。

  • 4)AUV 推进器故障诊断可视化问题。

  • 较多的神经网络故障诊断模型已经应用在 AUV 推进器故障诊断中,神经网络诊断过程可以理解为完全不需要考虑内部结构和逻辑结构。但是,存在神经网络内部运行机制和可解释性差的问题,在自动驾驶、智慧医疗、信息安全等场景却需要准确地解释网络运行机制。为了明确模型可解释性,对神经网络各激活层情况、提取的特征、特征分布情况等的可视化研究,成为需要研究的热点和重点。

  • 5)AUV 推进器故障诊断位置难定位、诊断时间长、算法计算复杂度高的问题。

  • 推进器故障诊断过程中怎样快速、准确地定位出故障位置是极其重要的,对于快速维修和自主决策都要求能快速找出具体的故障位置和原因。人工智能、统计分析等方法虽然能够作为推进器故障诊断方法,但是诊断时间相对较长,急需降低诊断时间,同时降低算法存在的复杂度以及寻找性能更好的运算处理器。因此,如何快速、准确地定位故障,降低算法复杂度,降低诊断时间是 AUV 推进器故障诊断重要的研究内容。

  • 参考文献

    • [1] WANG T Q,YUAN P.Technological economic study for ocean energy development in China[J].IEEE Technology Management Council Singapore Chapter,2011:610-614.

    • [2] 习近平.中国共产党第二十次全国代表大会上的报告[N].人民日报,2022-10-17(12).

    • [3] XU Q,TAN Z Y.Building a maritime community with a shared future:scholarly reflections on China’s new ocean vision[J].Marine Policy,2023,149:105508.

    • [4] CHEN Y H,XIAO S,LI D J.Stability analysis model for multi-node undersea observation networks[J].Simulation Modelling Practice and Theory,2019,97:101971.

    • [5] YUAN C R,SHUAI C G,MA J G,et al.An efficient control allocation algorithm for over-actuated AUVs trajectory tracking with fault-tolerant control[J].Ocean Engineering,2023,273:113976.

    • [6] YUAN C R,SHUAI C G,MA J G,et al.Adaptive optimal 3D nonlinear compound line-of-sight trajectory tracking control for over-actuated AUVs in attitude space[J].Ocean Engineering,2023,274:114056.

    • [7] GAO T Y,YANG J L,JIANG S D,et al.A novel fault diagnosis method for analog circuits based on conditional variational neural networks[J].Circuits,Systems,and Signal Processing,2021,40(6):2609-2633.

    • [8] LIU C Z,ANDRZEJ C,GRZEGORZ K,et al.Technology development and commercial applications of industrial fault diagnosis system:a review[J].The International Journal of Advanced Manufacturing Technology,2021,118(11-12):3497-3529.

    • [9] ABDULLAH A M,MAHATHIR M B,AYANTHA S M,et al.Multi-channel sensor fusion for real-time bearing fault diagnosis by frequency-domain multilinear principal component analysis[J].International Journal of Advanced Manufacturing Technology,2023,124(3-4):1321-1334.

    • [10] HASHIMOTO K,WATANABE M,TASHIRO S,et al.Missing of the ROV Kaiko vehicle-problem on the secondary cable[C]//Oceans’04 MTS/IEEE TechnoOcean’04.Kobe:IEEE,2004.

    • [11] ZHANG X Y,HONG J C,XU X M.Fault diagnosis of real-scenario battery systems based on modified entropy algorithms in electric vehicles[J].Journal of Energy Storage,2023,63:107079.

    • [12] PENG C,GAO H N,LIU X Y,et al.A visual vibration characterization method for intelligent fault diagnosis of rotating machinery[J].Mechanical Systems and Signal Processing,2023,192:110229.

    • [13] WANG J B,YANG B,ZENG C Y,et al.Recent advances and summarization of fault diagnosis techniques for proton exchange membrane fuel cell systems:a critical overview[J].Journal of Power Sources,2021,500:229932.

    • [14] KUMAR S S,MAHAPATRA D M,SING H L.Multifaceted Bio-sensing Technology[M].Salt Lake City:Academic Press,2023.

    • [15] WANG S H,CELEBI M E,ZHANGY D.Advances in data preprocessing for biomedical data fusion:an overview of the methods,challenges,and prospects[J].Information Fusion,2021,76:376-421.

    • [16] ZHONG H Y,YU S S,TRINH H,et al.Fine-tuning transfer learning based on DCGAN integrated with self-attention and spectral normalization for bearing fault diagnosis[J].Measurement,2023,210:112421.

    • [17] ALBAHRA S,GORBETT T,ROBERTSON SCOTT,et al.Artificial intelligence and machine learning overview in pathology & laboratory medicine:a general review of data preprocessing and basic supervised concepts[J].Seminars in Diagnostic Pathology,2023,40(2):71-87.

    • [18] DELISLE P L,BENOIT A R,DESROSIERS C,et al.Realistic image normalization for multi-domain segmentation[J].Medical Image Analysis,2021,74:102191.

    • [19] WANG J B,YANG B,ZENG C Y,et al.Recent advances and summarization of fault diagnosis techniques for proton exchange membrane fuel cell systems:a critical overview[J].Journal of Power Sources,2021,500:229932.

    • [20] RUAN W Y,SUN L.Robust latent discriminative adaptive graph preserving learning for image feature extraction[J].Knowledge-based Systems,2023,268:110487.

    • [21] GARCIA-ALVAREZ D,BREGON A,PULIDO B,et al.Integrating PCA and structural model decomposition to improve fault monitoring and diagnosis with varying operation points[J].Engineering Applications of Artificial Intelligence,2023,122:106145.

    • [22] SILVA R,MELO-PINTO P.T-SNE:a study on reducing the dimensionality of hyperspectral data for the regression problem of estimating oenological parameters[J].Artificial Intelligence in Agriculture,2023,7:58-68.

    • [23] ZHOU C,JIA Y,ZHAO S,et al.A mechanical part fault diagnosis method based on improved multiscale weighted permutation entropy and multiclass LSTSVM[J].Measurement,2023:112671.

    • [24] SARKAR S,PRAMANIK A,MAITI J.An integrated approach using rough set theory,ANFIS,and Z-number in occupational risk prediction[J].Engineering Applications of Artificial Intelligence,2023,117:105515.

    • [25] HUANG H,LI Y Y,LI W X.Dynamic reliability analysis of stochastic structures under non-stationary random excitations based on an explicit time-domain method[J].Structural Safety,2023,101:102313.

    • [26] HAN Y J,SONG T C,FENG J,et al.Grayscale-inversion and rotation invariant image description with sorted LBP features[J].Signal Processing:Image Communication,2021,99:116491.

    • [27] HAMDANI H,SEPTIARINI A,SUNYOTO A,et al.Detection of oil palm leaf disease based on color histogram and supervised classifier[J].Optik-International Journal for Light and Electron Optics,2021,245:167753.

    • [28] HUANG X,HAN B,NING Y Q,et al.Edge-based feature extraction module for 3D point cloud shape classification[J].Computers & Graphics,2023,112:31-39.

    • [29] YANG C Z,YU Q.Multiscale Fourier descriptor based on triangular features for shape retrieval[J].Signal Processing:Image Communication,2019,71:110-119.

    • [30] TAHER S A,LI J,FANG H Z.Simultaneous seismic input and state estimation with optimal sensor placement for building structures using incomplete acceleration measurements[J].Mechanical Systems and Signal Processing,2023,188:110047.

    • [31] ZENG C H,LAI W C,LIN H J,et al.Weak information extraction of gamma spectrum based on a two-dimensional wavelet transform[J].Radiation Physics and Chemistry,2023,208:110914.

    • [32] JIAO S B,SHI J Q,WANG Y,et al.A novel image noise reduction method for composite multistable stochastic resonance systems[J].Heliyon,2023,9(3):e14431.

    • [33] WEI D,FENG H D,HAN Q,et al.Fault detection and diagnosis for variable-air-volume systems using combined residual,qualitative and quantitative techniques[J].Energy & Buildings,2022,254:111491.

    • [34] HASSANNAYEBI E,NOURIAN R,MOUSAVI S M,et al.Predictive analytics for fault reasoning in gas flow control facility:a hybrid fuzzy theory and expert system approach[J].Journal of Loss Prevention in the Process Industries,2022,77:104796.

    • [35] LIU Y Q,RAMIN P,FLORES-ALSINA X,et al.Transforming data into actionable knowledge for fault detection,diagnosis and prognosis in urban wastewater systems with AI techniques:a mini-review[J].Process Safety and Environmental Protection,2023,172:501-512.

    • [36] MELLIT A,BENGHANEM M,KALOGIROU S,et al.An embedded system for remote monitoring and fault diagnosis of photovoltaic arrays using machine learning and the internet of things[J].Renewable Energy,2023,208:399-408.

    • [37] CAO H S,SUN P W,ZHAO L.PCA-SVM method with sliding window for online fault diagnosis of a small pressurized water reactor[J].Annals of Nuclear Energy,2022,171:109036.

    • [38] HUANG X,TENG Z S,TANG Q,et al.Fault diagnosis of automobile power seat with acoustic analysis and retrained SVM based on smartphone[J].Measurement,2022,202:111699.

    • [39] ZHANG H X,FENG W,ZHANG W J,et al.High anti-interference fault diagnosis method for PMSG drive system based on feature clustering optimization[J].Energy Reports,2023,9:576-582.

    • [40] WEN S Q,ZHANG W R,SUN Y F,et al.An enhanced principal component analysis method with Savitzky-Golay filter and clustering algorithm for sensor fault detection and diagnosis[J].Applied Energy,2023,337:120862.

    • [41] YUAN C R,SHUAI C G,MA J G,et al.An efficient control allocation algorithm for over-actuated AUVs trajectory tracking with fault-tolerant control[J].Ocean Engineering,2023,273:113976.

    • [42] ZHOU X,HUANG J Q,LU F,et al.A novel compound fault-tolerant method based on online sequential extreme learning machine with cycle reservoir for turbofan engine direct thrust control[J].Aerospace Science and Technology,2023,132:108059.

    • [43] LI Z Q,WANG W W,ZHANG C L,et al.Fault-tolerant control based on fractional sliding mode:crawler plant protection robot[J].Computers and Electrical Engineering,2023,105:108527.

    • [44] CONCHAS R F,SANCHEZ E N,RICALDE L J,et al.Sensor fault-tolerant control for a doubly fed induction generator in a smart grid[J].Engineering Applications of Artificial Intelligence,2023,117:105527.

    • [45] SANWALE J,DAHIYA S,TRIVEDI P,et al.Robust fault-tolerant adaptive integral dynamic sliding mode control using finite-time disturbance observer for coaxial octorotor UAVs[J].Control Engineering Practice,2023,135:105495.

    • [46] CHU Y H,WEI Z Q,YANG Z.New reweighted atomic norm minimization approach for line spectral estimation[J].Signal Processing,2023,206:108897.

    • [47] MENG Q H,ZHU S Y.Anomaly detection for construction vibration signals using unsupervised deep learning and cloud computing[J].Advanced Engineering Informatics,2023,55:101907.

    • [48] YAN Z P,ZHANG M Y,ZHANG C,et al.Decentralized formation trajectory tracking control of multi-AUV system with actuator saturation[J].Ocean Engineering,2022,255:111423.

    • [49] XIA S X,ZHOU X F,SHI H B,et al.A fault diagnosis method with multi-source data fusion based on hierarchical attention for AUV[J].Ocean Engineering,2022,266(1):112595.

    • [50] LI C Y,GUO S X.Characteristic evaluation via multi-sensor information fusion strategy for spherical underwater robots[J].Information Fusion,2023,95:199-214.

    • [51] WANG C,WANG Y H,ZHANG R F,et al.Analysis of roll center compensation method for underwater gliders based on deep learning[J].Ocean Engineering,2022,246:110529.

    • [52] YUAN C R,SHUAI C G,MA J G,et al.An efficient control allocation algorithm for over-actuated AUVs trajectory tracking with fault-tolerant control[J].Ocean Engineering,2023,273:113976.

    • [53] HACIZADE C,VURAL S Y.Active fault tolerant lateral control against actuator faults applied to AUV dynamics[C]//The 4th Conference on Control and Fault Tolerant Systems.Casablanca:IEEE,2019.

    • [54] WU P,HARRIS C A,SALAVASIDIS G,et al.Unsupervised anomaly detection for underwater gliders using generative adversarial networks[J].Engineering Applications of Artificial Intelligence,2021,104:104379.

    • [55] DAS D B,BIRANT D.GASEL:Genetic Algorithm-supported Ensemble Learning for fault detection in autonomous underwater vehicles[J].Ocean Engineering,2023,272:113844.

    • [56] TOPINI E,FANELLI F,TOPINI A,et al.An experimental comparison of Deep Learning strategies for AUV navigation in DVL-denied environments[J].Ocean Engineering,2023,274:114034.

    • [57] HUANG Z Y,SU Z Y,HUANG B.Quaternion-based finite-time fault-tolerant trajectory tracking control for autonomous underwater vehicle without unwinding[J].ISA Transactions,2022,131:15-30.

    • [58] YAN Z P,ZHANG M Y,ZHANG C,et al.Decentralized formation trajectory tracking control of multi-AUV system with actuator saturation[J].Ocean Engineering,2022,255:111423.

    • [59] LIU X X,JU Y T,LIU X H,et al.An IMU fault diagnosis and information reconstruction method based on analytical redundancy for autonomous underwater vehicle[J].IEEE Sensors Journal,2022,22(12):12127-12138.

    • [60] LI Z F,WANG M,MA G,et al.Adaptive reinforcement learning fault-tolerant control for AUVs with thruster faults based on the integral extended state observer[J].Ocean Engineering,2023,271:113722.

    • [61] WANG Q Y,LIU K Y,CAO Z Y.System noise variance matrix adaptive Kalman filter method for AUV INS/DVL navigation system[J].Ocean Engineering,2023,267:113269.

    • [62] SONG J H,HE X.Robust state estimation and fault detection for autonomous underwater vehicles considering hydrodynamic effects[J].Control Engineering Practice,2023,135:105497.

    • [63] LI X J,QIN H D,LI L Y.Fixed-time formation control for AUVs with unknown actuator faults based on lumped disturbance observer[J].Ocean Engineering,2023,269:113495.

    • [64] HAO L Y,ZHANG Y Q,LI H.Fault-tolerant control via integral sliding mode output feedback for unmanned marine vehicles[J].Applied Mathematics and Computation,2021,401:126078.

    • [65] ZHU C,HUANG B,ZHOU B,et al.Adaptive model-parameter-free fault-tolerant trajectory tracking control for autonomous underwater vehicles[J].ISA Transactions,2021,114:57-71.

    • [66] HAO L Y,ZHANG H,YUE W,et al.Fault-tolerant compensation control based on sliding mode technique of unmanned marine vehicles subject to unknown persistent ocean disturbances[J].International Journal of Control,Automation and Systems,2020,18(3):739-752.

    • [67] CHU Z Z,XIANG X B,ZHU D Q,et al.Adaptive fuzzy sliding mode diving control for autonomous underwater vehicle with input constraint[J].International Journal of Fuzzy Systems,2018,20(5):1460-1469.

    • [68] LIU X,ZHANG M,YAO F.Adaptive fault tolerant control and thruster fault reconstruction for autonomous underwater vehicle[J].Ocean Engineering,2018,155:10-23.

    • [69] BALDINI A,FASANO A,FELICETTI R,et al.A model-based active fault tolerant control scheme for a remotely operated vehicle[J].IFAC PapersOnLine,2018,51(24):798-805.

    • [70] CORRADINI M L,LONGHI S,MONTERIU A,et al.Observer-based fault tolerant sliding mode control for remotely operated vehicles[J].IFAC PapersOnLine,2010,43(20):173-178.

    • [71] ZHANG M J,LIU X,YIN B J,et al.Adaptive terminal sliding mode based thruster fault tolerant control for underwater vehicle in time-varying ocean currents[J].Journal of the Franklin Institute,2015,352(11):4935-4961.

    • [72] CHU Z Z,ZHANG M J.Fault reconstruction of thruster for autonomous underwater vehicle based on terminal sliding mode observer[J].Ocean Engineering,2014,88:426-434.

    • [73] SHUMSKY A.Data driven method for fault detection and isolation in nonlinear uncertain systems[J].IFAC PapersOnLinec,2007,7(1):283-286.

    • [74] FILARETOV V,ZHIRABOK A,ZUEV A,et al.The new approach for synthesis of diagnostic system for navigation sensors of underwater vehicles[J].Procedia Engineering,2014,69:822-829.

    • [75] ALVAREZ M D,LANE D M.A Hidden Markov Model application with Gaussian Mixture emissions for fault detection and diagnosis on a simulated AUV platform[C]//OCEANS 2016 MTS/IEEE Monterey.Monterey:IEEE,2016.

    • [76] FREDDI A,LONGHI S,MONTERIU A.Actuator fault detection system for a remotely operated vehicle[J].IFAC PapersOnLine,2013,46(33):356-361.

    • [77] CORRADINI M L,A MONTERIÙ,ORLANDO G.Actuator failure tolerant control design for underwater remotely operated vehicles[J].IFAC PapersOnLine,2010,43(16):25-30.

    • [78] BIAN X Q,MOU C H,YAN Z P,et al.Reliability analysis of AUV based on fuzzy fault tree[C]//2009 International Conference on Mechatronics and Automation.Changchun:IEEE,2009.

    • [79] MADSEN A L,KJÆRULFF U B.Applications of HUGIN to diagnosis and control of autonomous vehicles[J].Advances in Probabilistic Graphical Models,2007,213:313-332.

    • [80] LIU W X,WANG Y J,YIN B J,et al.Thruster fault identification based on fractal feature and multiresolution wavelet decomposition for autonomous underwater vehicle[J].Proceedings of the Institution of Mechanical Engineers Part C:Journal of Mechanical Engineering Science,2017,231(13):2528-2539.

    • [81] YIN B J,LIN X,TANG W X,et al.Thruster fault identification for autonomous underwater vehicle based on time-domain energy and time-frequency entropy of fusion signal[J].Intelligent Robotics and Applications,2019,11742:264-275.

    • [82] ZHANG M J,WU J,CHU Z Z.Multi-fault diagnosis for autonomous underwater vehicle based on fuzzy weighted support vector domain description[J].China Ocean Engineering Society,2014,28(5):599-616.

    • [83] NASCIMENTO S,VALDENEGRO-TORO M.Modeling and soft-fault diagnosis of underwater thrusters with Recurrent Neural Networks[J].IFAC PapersOnLine,2018,51(29):80-85.

    • [84] WANG Y J,ZHANG M J,WU J,et al.Research of the fault diagnosis method for the thruster of AUV based on information fusion[C]//The 3rd International Conference on Intelligent Computing.Qingdao:IEEE,2007.

    • [85] ZHANG Y J,ZHANG M J.Study of fuzzy neural networks model for system condition monitoring of AUV[J].Journal of Marine Science and Application,2002,1(2):42-45.

    • [86] LIU Q,ZHU D Q,YANG S X.Unmanned underwater vehicles fault identification and fault-tolerant control method based on FCA-CMAC neural networks,applied on an actuated vehicle[J].Journal of Intelligent & Robotic Systems,2012,66(4):463-475.

    • [87] SUN Y S,LI Y M,ZHANG G C,et al.Actuator fault diagnosis of autonomous underwater vehicle based on improved Elman neural network[J].Journal of Central South University,2016,23(4):808-816.

    • [88] LI X,SONG Y,GUO J,et al.Sensor fault diagnosis of autonomous underwater vehicle based on extreme learning machine[C]//2017 IEEE Underwater Technology.Busan:IEEE,2017.

    • [89] SUBHA T D,SUBASH T,JANE K,et al.Autonomous under water vehicle based on extreme learning machine for sensor fault diagnosistics[J].Materials Today:Proceedings,2020,24(2):2394-2402.

    • [90] WANG Y J,ZHAO J,ZHANG M J,et al.Research on the sensors condition monitoring method for AUV[J].Intelligent Robotics and Applications,2008,5314:427-436.

    • [91] TANG T,GANG Y.An intelligence fusion system for ship fault-tolerant control[J].IFAC PapersOnLine,2004,37(10):83-87.

    • [92] QIN X C,ZHANG W,GAO S,et al.Sensor fault diagnosis of autonomous underwater vehicle based on LSTM[C]//The 37th Chinese Control Conference.Wuhan:IEEE,2018.

    • [93] SHI C T,ZHANG R B.Software fault diagnosis model of AUV based on Bayesian Networks and its simplified method[C]//The 9th World Congress on Intelligent Control and Automation.Taipei:IEEE,2011.

    • [94] OMERDIC E,ROBERTS G.Thruster fault diagnosis and accommodation for open-frame underwater vehicles[J].Control Engineering Practice,2004,12(12):1575-1598.

    • [95] ZHENG X,FENG C,LI T,et al.Analysis of autonomous underwater vehicle navigational states based on complex networks[J].Ocean Engineering,2019,187:106141.

    • [96] LIU W X,WANG Y J,LIU X,et al.Weak thruster fault detection for AUV based on stochastic resonance and wavelet reconstruction[J].Journal of Central South University,2016,23(11):2883-2895.

    • [97] WANG Y J,ZHANG M J,WILSON P A,et al.Adaptive neural network-based backstepping fault tolerant control for underwater vehicles with thruster fault[J].Ocean Engineering,2015,110:15-24.

    • [98] ZHANG M J,WANG Y J,XU J A,et al.Thruster fault diagnosis in autonomous underwater vehicle based on grey qualitative simulation[J].Ocean Engineering,2015,105:247-255.

  • 参考文献

    • [1] WANG T Q,YUAN P.Technological economic study for ocean energy development in China[J].IEEE Technology Management Council Singapore Chapter,2011:610-614.

    • [2] 习近平.中国共产党第二十次全国代表大会上的报告[N].人民日报,2022-10-17(12).

    • [3] XU Q,TAN Z Y.Building a maritime community with a shared future:scholarly reflections on China’s new ocean vision[J].Marine Policy,2023,149:105508.

    • [4] CHEN Y H,XIAO S,LI D J.Stability analysis model for multi-node undersea observation networks[J].Simulation Modelling Practice and Theory,2019,97:101971.

    • [5] YUAN C R,SHUAI C G,MA J G,et al.An efficient control allocation algorithm for over-actuated AUVs trajectory tracking with fault-tolerant control[J].Ocean Engineering,2023,273:113976.

    • [6] YUAN C R,SHUAI C G,MA J G,et al.Adaptive optimal 3D nonlinear compound line-of-sight trajectory tracking control for over-actuated AUVs in attitude space[J].Ocean Engineering,2023,274:114056.

    • [7] GAO T Y,YANG J L,JIANG S D,et al.A novel fault diagnosis method for analog circuits based on conditional variational neural networks[J].Circuits,Systems,and Signal Processing,2021,40(6):2609-2633.

    • [8] LIU C Z,ANDRZEJ C,GRZEGORZ K,et al.Technology development and commercial applications of industrial fault diagnosis system:a review[J].The International Journal of Advanced Manufacturing Technology,2021,118(11-12):3497-3529.

    • [9] ABDULLAH A M,MAHATHIR M B,AYANTHA S M,et al.Multi-channel sensor fusion for real-time bearing fault diagnosis by frequency-domain multilinear principal component analysis[J].International Journal of Advanced Manufacturing Technology,2023,124(3-4):1321-1334.

    • [10] HASHIMOTO K,WATANABE M,TASHIRO S,et al.Missing of the ROV Kaiko vehicle-problem on the secondary cable[C]//Oceans’04 MTS/IEEE TechnoOcean’04.Kobe:IEEE,2004.

    • [11] ZHANG X Y,HONG J C,XU X M.Fault diagnosis of real-scenario battery systems based on modified entropy algorithms in electric vehicles[J].Journal of Energy Storage,2023,63:107079.

    • [12] PENG C,GAO H N,LIU X Y,et al.A visual vibration characterization method for intelligent fault diagnosis of rotating machinery[J].Mechanical Systems and Signal Processing,2023,192:110229.

    • [13] WANG J B,YANG B,ZENG C Y,et al.Recent advances and summarization of fault diagnosis techniques for proton exchange membrane fuel cell systems:a critical overview[J].Journal of Power Sources,2021,500:229932.

    • [14] KUMAR S S,MAHAPATRA D M,SING H L.Multifaceted Bio-sensing Technology[M].Salt Lake City:Academic Press,2023.

    • [15] WANG S H,CELEBI M E,ZHANGY D.Advances in data preprocessing for biomedical data fusion:an overview of the methods,challenges,and prospects[J].Information Fusion,2021,76:376-421.

    • [16] ZHONG H Y,YU S S,TRINH H,et al.Fine-tuning transfer learning based on DCGAN integrated with self-attention and spectral normalization for bearing fault diagnosis[J].Measurement,2023,210:112421.

    • [17] ALBAHRA S,GORBETT T,ROBERTSON SCOTT,et al.Artificial intelligence and machine learning overview in pathology & laboratory medicine:a general review of data preprocessing and basic supervised concepts[J].Seminars in Diagnostic Pathology,2023,40(2):71-87.

    • [18] DELISLE P L,BENOIT A R,DESROSIERS C,et al.Realistic image normalization for multi-domain segmentation[J].Medical Image Analysis,2021,74:102191.

    • [19] WANG J B,YANG B,ZENG C Y,et al.Recent advances and summarization of fault diagnosis techniques for proton exchange membrane fuel cell systems:a critical overview[J].Journal of Power Sources,2021,500:229932.

    • [20] RUAN W Y,SUN L.Robust latent discriminative adaptive graph preserving learning for image feature extraction[J].Knowledge-based Systems,2023,268:110487.

    • [21] GARCIA-ALVAREZ D,BREGON A,PULIDO B,et al.Integrating PCA and structural model decomposition to improve fault monitoring and diagnosis with varying operation points[J].Engineering Applications of Artificial Intelligence,2023,122:106145.

    • [22] SILVA R,MELO-PINTO P.T-SNE:a study on reducing the dimensionality of hyperspectral data for the regression problem of estimating oenological parameters[J].Artificial Intelligence in Agriculture,2023,7:58-68.

    • [23] ZHOU C,JIA Y,ZHAO S,et al.A mechanical part fault diagnosis method based on improved multiscale weighted permutation entropy and multiclass LSTSVM[J].Measurement,2023:112671.

    • [24] SARKAR S,PRAMANIK A,MAITI J.An integrated approach using rough set theory,ANFIS,and Z-number in occupational risk prediction[J].Engineering Applications of Artificial Intelligence,2023,117:105515.

    • [25] HUANG H,LI Y Y,LI W X.Dynamic reliability analysis of stochastic structures under non-stationary random excitations based on an explicit time-domain method[J].Structural Safety,2023,101:102313.

    • [26] HAN Y J,SONG T C,FENG J,et al.Grayscale-inversion and rotation invariant image description with sorted LBP features[J].Signal Processing:Image Communication,2021,99:116491.

    • [27] HAMDANI H,SEPTIARINI A,SUNYOTO A,et al.Detection of oil palm leaf disease based on color histogram and supervised classifier[J].Optik-International Journal for Light and Electron Optics,2021,245:167753.

    • [28] HUANG X,HAN B,NING Y Q,et al.Edge-based feature extraction module for 3D point cloud shape classification[J].Computers & Graphics,2023,112:31-39.

    • [29] YANG C Z,YU Q.Multiscale Fourier descriptor based on triangular features for shape retrieval[J].Signal Processing:Image Communication,2019,71:110-119.

    • [30] TAHER S A,LI J,FANG H Z.Simultaneous seismic input and state estimation with optimal sensor placement for building structures using incomplete acceleration measurements[J].Mechanical Systems and Signal Processing,2023,188:110047.

    • [31] ZENG C H,LAI W C,LIN H J,et al.Weak information extraction of gamma spectrum based on a two-dimensional wavelet transform[J].Radiation Physics and Chemistry,2023,208:110914.

    • [32] JIAO S B,SHI J Q,WANG Y,et al.A novel image noise reduction method for composite multistable stochastic resonance systems[J].Heliyon,2023,9(3):e14431.

    • [33] WEI D,FENG H D,HAN Q,et al.Fault detection and diagnosis for variable-air-volume systems using combined residual,qualitative and quantitative techniques[J].Energy & Buildings,2022,254:111491.

    • [34] HASSANNAYEBI E,NOURIAN R,MOUSAVI S M,et al.Predictive analytics for fault reasoning in gas flow control facility:a hybrid fuzzy theory and expert system approach[J].Journal of Loss Prevention in the Process Industries,2022,77:104796.

    • [35] LIU Y Q,RAMIN P,FLORES-ALSINA X,et al.Transforming data into actionable knowledge for fault detection,diagnosis and prognosis in urban wastewater systems with AI techniques:a mini-review[J].Process Safety and Environmental Protection,2023,172:501-512.

    • [36] MELLIT A,BENGHANEM M,KALOGIROU S,et al.An embedded system for remote monitoring and fault diagnosis of photovoltaic arrays using machine learning and the internet of things[J].Renewable Energy,2023,208:399-408.

    • [37] CAO H S,SUN P W,ZHAO L.PCA-SVM method with sliding window for online fault diagnosis of a small pressurized water reactor[J].Annals of Nuclear Energy,2022,171:109036.

    • [38] HUANG X,TENG Z S,TANG Q,et al.Fault diagnosis of automobile power seat with acoustic analysis and retrained SVM based on smartphone[J].Measurement,2022,202:111699.

    • [39] ZHANG H X,FENG W,ZHANG W J,et al.High anti-interference fault diagnosis method for PMSG drive system based on feature clustering optimization[J].Energy Reports,2023,9:576-582.

    • [40] WEN S Q,ZHANG W R,SUN Y F,et al.An enhanced principal component analysis method with Savitzky-Golay filter and clustering algorithm for sensor fault detection and diagnosis[J].Applied Energy,2023,337:120862.

    • [41] YUAN C R,SHUAI C G,MA J G,et al.An efficient control allocation algorithm for over-actuated AUVs trajectory tracking with fault-tolerant control[J].Ocean Engineering,2023,273:113976.

    • [42] ZHOU X,HUANG J Q,LU F,et al.A novel compound fault-tolerant method based on online sequential extreme learning machine with cycle reservoir for turbofan engine direct thrust control[J].Aerospace Science and Technology,2023,132:108059.

    • [43] LI Z Q,WANG W W,ZHANG C L,et al.Fault-tolerant control based on fractional sliding mode:crawler plant protection robot[J].Computers and Electrical Engineering,2023,105:108527.

    • [44] CONCHAS R F,SANCHEZ E N,RICALDE L J,et al.Sensor fault-tolerant control for a doubly fed induction generator in a smart grid[J].Engineering Applications of Artificial Intelligence,2023,117:105527.

    • [45] SANWALE J,DAHIYA S,TRIVEDI P,et al.Robust fault-tolerant adaptive integral dynamic sliding mode control using finite-time disturbance observer for coaxial octorotor UAVs[J].Control Engineering Practice,2023,135:105495.

    • [46] CHU Y H,WEI Z Q,YANG Z.New reweighted atomic norm minimization approach for line spectral estimation[J].Signal Processing,2023,206:108897.

    • [47] MENG Q H,ZHU S Y.Anomaly detection for construction vibration signals using unsupervised deep learning and cloud computing[J].Advanced Engineering Informatics,2023,55:101907.

    • [48] YAN Z P,ZHANG M Y,ZHANG C,et al.Decentralized formation trajectory tracking control of multi-AUV system with actuator saturation[J].Ocean Engineering,2022,255:111423.

    • [49] XIA S X,ZHOU X F,SHI H B,et al.A fault diagnosis method with multi-source data fusion based on hierarchical attention for AUV[J].Ocean Engineering,2022,266(1):112595.

    • [50] LI C Y,GUO S X.Characteristic evaluation via multi-sensor information fusion strategy for spherical underwater robots[J].Information Fusion,2023,95:199-214.

    • [51] WANG C,WANG Y H,ZHANG R F,et al.Analysis of roll center compensation method for underwater gliders based on deep learning[J].Ocean Engineering,2022,246:110529.

    • [52] YUAN C R,SHUAI C G,MA J G,et al.An efficient control allocation algorithm for over-actuated AUVs trajectory tracking with fault-tolerant control[J].Ocean Engineering,2023,273:113976.

    • [53] HACIZADE C,VURAL S Y.Active fault tolerant lateral control against actuator faults applied to AUV dynamics[C]//The 4th Conference on Control and Fault Tolerant Systems.Casablanca:IEEE,2019.

    • [54] WU P,HARRIS C A,SALAVASIDIS G,et al.Unsupervised anomaly detection for underwater gliders using generative adversarial networks[J].Engineering Applications of Artificial Intelligence,2021,104:104379.

    • [55] DAS D B,BIRANT D.GASEL:Genetic Algorithm-supported Ensemble Learning for fault detection in autonomous underwater vehicles[J].Ocean Engineering,2023,272:113844.

    • [56] TOPINI E,FANELLI F,TOPINI A,et al.An experimental comparison of Deep Learning strategies for AUV navigation in DVL-denied environments[J].Ocean Engineering,2023,274:114034.

    • [57] HUANG Z Y,SU Z Y,HUANG B.Quaternion-based finite-time fault-tolerant trajectory tracking control for autonomous underwater vehicle without unwinding[J].ISA Transactions,2022,131:15-30.

    • [58] YAN Z P,ZHANG M Y,ZHANG C,et al.Decentralized formation trajectory tracking control of multi-AUV system with actuator saturation[J].Ocean Engineering,2022,255:111423.

    • [59] LIU X X,JU Y T,LIU X H,et al.An IMU fault diagnosis and information reconstruction method based on analytical redundancy for autonomous underwater vehicle[J].IEEE Sensors Journal,2022,22(12):12127-12138.

    • [60] LI Z F,WANG M,MA G,et al.Adaptive reinforcement learning fault-tolerant control for AUVs with thruster faults based on the integral extended state observer[J].Ocean Engineering,2023,271:113722.

    • [61] WANG Q Y,LIU K Y,CAO Z Y.System noise variance matrix adaptive Kalman filter method for AUV INS/DVL navigation system[J].Ocean Engineering,2023,267:113269.

    • [62] SONG J H,HE X.Robust state estimation and fault detection for autonomous underwater vehicles considering hydrodynamic effects[J].Control Engineering Practice,2023,135:105497.

    • [63] LI X J,QIN H D,LI L Y.Fixed-time formation control for AUVs with unknown actuator faults based on lumped disturbance observer[J].Ocean Engineering,2023,269:113495.

    • [64] HAO L Y,ZHANG Y Q,LI H.Fault-tolerant control via integral sliding mode output feedback for unmanned marine vehicles[J].Applied Mathematics and Computation,2021,401:126078.

    • [65] ZHU C,HUANG B,ZHOU B,et al.Adaptive model-parameter-free fault-tolerant trajectory tracking control for autonomous underwater vehicles[J].ISA Transactions,2021,114:57-71.

    • [66] HAO L Y,ZHANG H,YUE W,et al.Fault-tolerant compensation control based on sliding mode technique of unmanned marine vehicles subject to unknown persistent ocean disturbances[J].International Journal of Control,Automation and Systems,2020,18(3):739-752.

    • [67] CHU Z Z,XIANG X B,ZHU D Q,et al.Adaptive fuzzy sliding mode diving control for autonomous underwater vehicle with input constraint[J].International Journal of Fuzzy Systems,2018,20(5):1460-1469.

    • [68] LIU X,ZHANG M,YAO F.Adaptive fault tolerant control and thruster fault reconstruction for autonomous underwater vehicle[J].Ocean Engineering,2018,155:10-23.

    • [69] BALDINI A,FASANO A,FELICETTI R,et al.A model-based active fault tolerant control scheme for a remotely operated vehicle[J].IFAC PapersOnLine,2018,51(24):798-805.

    • [70] CORRADINI M L,LONGHI S,MONTERIU A,et al.Observer-based fault tolerant sliding mode control for remotely operated vehicles[J].IFAC PapersOnLine,2010,43(20):173-178.

    • [71] ZHANG M J,LIU X,YIN B J,et al.Adaptive terminal sliding mode based thruster fault tolerant control for underwater vehicle in time-varying ocean currents[J].Journal of the Franklin Institute,2015,352(11):4935-4961.

    • [72] CHU Z Z,ZHANG M J.Fault reconstruction of thruster for autonomous underwater vehicle based on terminal sliding mode observer[J].Ocean Engineering,2014,88:426-434.

    • [73] SHUMSKY A.Data driven method for fault detection and isolation in nonlinear uncertain systems[J].IFAC PapersOnLinec,2007,7(1):283-286.

    • [74] FILARETOV V,ZHIRABOK A,ZUEV A,et al.The new approach for synthesis of diagnostic system for navigation sensors of underwater vehicles[J].Procedia Engineering,2014,69:822-829.

    • [75] ALVAREZ M D,LANE D M.A Hidden Markov Model application with Gaussian Mixture emissions for fault detection and diagnosis on a simulated AUV platform[C]//OCEANS 2016 MTS/IEEE Monterey.Monterey:IEEE,2016.

    • [76] FREDDI A,LONGHI S,MONTERIU A.Actuator fault detection system for a remotely operated vehicle[J].IFAC PapersOnLine,2013,46(33):356-361.

    • [77] CORRADINI M L,A MONTERIÙ,ORLANDO G.Actuator failure tolerant control design for underwater remotely operated vehicles[J].IFAC PapersOnLine,2010,43(16):25-30.

    • [78] BIAN X Q,MOU C H,YAN Z P,et al.Reliability analysis of AUV based on fuzzy fault tree[C]//2009 International Conference on Mechatronics and Automation.Changchun:IEEE,2009.

    • [79] MADSEN A L,KJÆRULFF U B.Applications of HUGIN to diagnosis and control of autonomous vehicles[J].Advances in Probabilistic Graphical Models,2007,213:313-332.

    • [80] LIU W X,WANG Y J,YIN B J,et al.Thruster fault identification based on fractal feature and multiresolution wavelet decomposition for autonomous underwater vehicle[J].Proceedings of the Institution of Mechanical Engineers Part C:Journal of Mechanical Engineering Science,2017,231(13):2528-2539.

    • [81] YIN B J,LIN X,TANG W X,et al.Thruster fault identification for autonomous underwater vehicle based on time-domain energy and time-frequency entropy of fusion signal[J].Intelligent Robotics and Applications,2019,11742:264-275.

    • [82] ZHANG M J,WU J,CHU Z Z.Multi-fault diagnosis for autonomous underwater vehicle based on fuzzy weighted support vector domain description[J].China Ocean Engineering Society,2014,28(5):599-616.

    • [83] NASCIMENTO S,VALDENEGRO-TORO M.Modeling and soft-fault diagnosis of underwater thrusters with Recurrent Neural Networks[J].IFAC PapersOnLine,2018,51(29):80-85.

    • [84] WANG Y J,ZHANG M J,WU J,et al.Research of the fault diagnosis method for the thruster of AUV based on information fusion[C]//The 3rd International Conference on Intelligent Computing.Qingdao:IEEE,2007.

    • [85] ZHANG Y J,ZHANG M J.Study of fuzzy neural networks model for system condition monitoring of AUV[J].Journal of Marine Science and Application,2002,1(2):42-45.

    • [86] LIU Q,ZHU D Q,YANG S X.Unmanned underwater vehicles fault identification and fault-tolerant control method based on FCA-CMAC neural networks,applied on an actuated vehicle[J].Journal of Intelligent & Robotic Systems,2012,66(4):463-475.

    • [87] SUN Y S,LI Y M,ZHANG G C,et al.Actuator fault diagnosis of autonomous underwater vehicle based on improved Elman neural network[J].Journal of Central South University,2016,23(4):808-816.

    • [88] LI X,SONG Y,GUO J,et al.Sensor fault diagnosis of autonomous underwater vehicle based on extreme learning machine[C]//2017 IEEE Underwater Technology.Busan:IEEE,2017.

    • [89] SUBHA T D,SUBASH T,JANE K,et al.Autonomous under water vehicle based on extreme learning machine for sensor fault diagnosistics[J].Materials Today:Proceedings,2020,24(2):2394-2402.

    • [90] WANG Y J,ZHAO J,ZHANG M J,et al.Research on the sensors condition monitoring method for AUV[J].Intelligent Robotics and Applications,2008,5314:427-436.

    • [91] TANG T,GANG Y.An intelligence fusion system for ship fault-tolerant control[J].IFAC PapersOnLine,2004,37(10):83-87.

    • [92] QIN X C,ZHANG W,GAO S,et al.Sensor fault diagnosis of autonomous underwater vehicle based on LSTM[C]//The 37th Chinese Control Conference.Wuhan:IEEE,2018.

    • [93] SHI C T,ZHANG R B.Software fault diagnosis model of AUV based on Bayesian Networks and its simplified method[C]//The 9th World Congress on Intelligent Control and Automation.Taipei:IEEE,2011.

    • [94] OMERDIC E,ROBERTS G.Thruster fault diagnosis and accommodation for open-frame underwater vehicles[J].Control Engineering Practice,2004,12(12):1575-1598.

    • [95] ZHENG X,FENG C,LI T,et al.Analysis of autonomous underwater vehicle navigational states based on complex networks[J].Ocean Engineering,2019,187:106141.

    • [96] LIU W X,WANG Y J,LIU X,et al.Weak thruster fault detection for AUV based on stochastic resonance and wavelet reconstruction[J].Journal of Central South University,2016,23(11):2883-2895.

    • [97] WANG Y J,ZHANG M J,WILSON P A,et al.Adaptive neural network-based backstepping fault tolerant control for underwater vehicles with thruster fault[J].Ocean Engineering,2015,110:15-24.

    • [98] ZHANG M J,WANG Y J,XU J A,et al.Thruster fault diagnosis in autonomous underwater vehicle based on grey qualitative simulation[J].Ocean Engineering,2015,105:247-255.