针对水下机器人推进系统的在线监测，提出一种具有在线学习能力的推进系统故障诊断方法。通过分析相关性的变化趋势，在线估计推进系统的时延。利用作业过程中采集的数据，对控制量与转速之间的关系进行在线建模。为提高建模精度，采用粒子群算法，对模型阶次和建模数据量进行在线优化。为适应作业过程中环境和系统自身状态的变化，设计了模型在线更新机制。基于该在线更新机制，提出一种不依赖传统阈值的自适应故障检测方法。通过海上试验数据和水池测试，验证了所提出算法的有效性。
For the online monitoring of underwater vehicle thruster system, a thruster system fault diagnosis method with online learning ability is proposed. Firstly, time delay of the thruster system is estimated online by analyzing the changing trend of correlationship. After that, online modeling of the relationship between control voltage and rotating speed is implemented use the data acquired during operation. In order to improve the modeling accuracy, the particle swarm optimization algorithm is utilized to optimize the model order and modeling data volume. To adapt to the changes of environment and system status during operation, an online update mechanism of the model is designed. Based on the online update mechanism, an adaptive fault detection method that does not rely on traditional thresholds is proposed. Finally, the effectiveness of the proposed algorithm is verified by sea trial data and pool tests.
2020,42(6): 95-100 收稿日期：2019-05-13
DOI：10.3404/j.issn.1672-7649.2020.06.019
分类号：TP242
基金项目：国家重点研发计划(2016YFC0300604)；海洋公益性行业科研专项(201505017-3)
作者简介：徐高飞(1987-),男,博士研究生,主要研究方向为水下机器人故障诊断
参考文献：
[1] FREDDI A, LONGHI S, MONTERIÙ A. Actuator fault detection system for a remotely operated vehicle[J]. IFAC Proceedings Volumes, 2013, 46(33):356-361
[2] 刘富樯, 徐德民, 高剑, 等. 水下航行器执行机构的故障诊断与容错控制[J]. 控制理论与应用, 2014, 31(9):1143-1150 LIU F Q, XU D M, GAO J, et al. Fault diagnosis and fault tolerant control for actuators of underwater vehicles[J]. Control Theory & Applications, 2014, 31(9):1143-1150
[3] ZHAO B, BLANKE M, SKJETNE R. Particle filter for fault diagnosis and robust navigation of underwater robot[J]. IEEE Transactions on Control Systems Technology, 2014, 22(6):2399-2407
[4] 程大军, 刘开周. 一种基于SUKF的广义行为环境建模及在远程AUV推进系统的应用研究[J]. 机械工程学报, 2011, 47(19):14-21 CHENG D J, LIU K Z. SUKF-based modeling for general behavior of the environment and its application in long-range AUV propulsion system[J]. Journal of Mechanical Engineering, 2011, 47(19):14-21
[5] SUN Y S, RAN X R, LI Y M, et al. Thruster fault diagnosis method based on Gaussian particle filter for autonomous underwater vehicles[J]. International Journal of Naval Architecture & Ocean Engineering, 2016, 8(3):243-251
[6] YU S S, YUE M L, GUO C Z, 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
[7] 万磊, 王建国, 姜春萌, 等. 基于神经网络的水下机器人推进器故障诊断[J]. 中国造船, 2011, 52(4):139-146 WAN L, WANG J G, JIANG C M, et al. Research on thruster fault diagnosis of underwater vehicles based on NN[J]. Shipbuilding of China, 2011, 52(4):139-146
[8] ANTONELLI G, CACCAVALE F, SANSONE C, et al. Fault diagnosis for AUVs using support vector machines[C]//ICRA 2004, New Orleans, LA, USA:IEEE, 2004:4486-4491.
[9] CAROLIS V D, MAURELLI F, BROWN K E, et al. Energy-aware fault-mitigation architecture for underwater vehicles[J]. Autonomous Robots, 2016, 41(5):1-23
[10] 袁芳, 朱大奇, 叶银忠. 基于降阶卡尔曼滤波器的水下机器人滑模容错控制[J]. 控制与决策, 2011, 26(7):1031-1035 YUAN F, ZHU D Q, YE Y Z. Sliding-mode fault-tolerant control method of underwater vehicle based on reduced-order Kalman filter[J]. Control and Decision, 2011, 26(7):1031-1035
[11] RAANAN B Y, BELLINGHAM J, ZHANG Y, et al. Detection of unanticipated faults for autonomous underwater vehicles using online topic models[J]. Journal of Field Robotics, 2018, 35:705-716
[12] 刘鑫宇, 李一平, 封锡盛. 万米级水下机器人浮力实时测量方法[J]. 机器人, 2018, 40(2):216-221 LIU X Y, LI Y P, FENG X S. Real-time measurement method of buoyancy of a full-ocean-depth underwater robot[J]. Robot, 2018, 40(2):216-221
[13] GARCIA-TEJEDA Y V, BARRERA-FIGUEROA V. Least squares fitting-polynomials for determining inflection points in adsorption isotherms of spray-dried acai juice (Euterpe oleracea Mart.) and soy sauce powders[J]. Powder Technology, 2019, 342:829-839
