水下无人系统学报2018，Vol.26Issue(2)：157-165,184,10.DOI:10.11993/j.issn.2096-3920.2018.02.010
基于RBF与OS-ELM神经网络的AUV传感器在线故障诊断
Online Fault Diagnosis of AUV Sensor Based on RBF and OS-ELM Neural Networks
段杰 1李辉 1陈自立 1龚时华 1赵朝闻1
作者信息
- 1. 中国船舶重工集团公司 第705研究所昆明分部, 云南 昆明, 650118
- 折叠
摘要
Abstract
Sensor is an important component part of an autonomous undersea vehicle(AUV). Real-time and accurate online fault diagnosis of AUV sensors is of great significance to improve the safety of an AUV. This study analyzes the machine learning algorithms, and builds a radial basis function(RBF) neural network-based AUV sensor predictor with high accuracy and real-time performance. Subsequently, the online sequential extreme learning machine(OS-ELM) al-gorithm is applied to the online sensor fault diagnosis to improve the real time performance and accuracy of the pre-dictor. Two kinds of fault diagnosis models are simulated and compared by using the sea trial data of AUV sensor, and the results show that the prediction accuracy and real-time performance of the OS-ELM neural network predictor with RBF neural network algorithm are higher than that of RBF neural network predictor. This research may provide a refer-ence for the design of on-line fault diagnosis scheme of AUV control system.关键词
自主式水下航行器(AUV)/径向基函数(RBF)/在线贯序学习机(OS-ELM)/神经网络/在线故障诊断/传感器Key words
autonomous undersea vehicle(AUV)/radial basis function(RBF)/online sequential extreme learning ma-chine(OS-ELM)/neural network/online fault diagnosis/sensor分类
交通工程引用本文复制引用
段杰,李辉,陈自立,龚时华,赵朝闻..基于RBF与OS-ELM神经网络的AUV传感器在线故障诊断[J].水下无人系统学报,2018,26(2):157-165,184,10.
