Modeling and Soft-fault Diagnosis of Underwater Thrusters with Recurrent\n Neural Networks

2018/07/11 by Samy Nascimento, Nascimento, Samy, Matías Valdenegro-Toro +1
Engineering · #Fault Detection and Control Systems #Oil and Gas Production Techniques #Machine Fault Diagnosis Techniques

paper · pdf · doi:10.48550/arxiv.1807.04109

Abstract

Noncritical soft-faults and model deviations are a challenge for Fault\nDetection and Diagnosis (FDD) of resident Autonomous Underwater Vehicles\n(AUVs). Such systems may have a faster performance degradation due to the\npermanent exposure to the marine environment, and constant monitoring of\ncomponent conditions is required to ensure their reliability. This works\npresents an evaluation of Recurrent Neural Networks (RNNs) for a data-driven\nfault detection and diagnosis scheme for underwater thrusters with empirical\ndata. The nominal behavior of the thruster was modeled using the measured\ncontrol input, voltage, rotational speed and current signals. We evaluated the\nperformance of fault classification using all the measured signals compared to\nusing the computed residuals from the nominal model as features.\n

Related

• A Deep Learning Approach To Dead-Reckoning Navigation For Autonomous Underwater Vehicles With Limited Sensor Payloads
• Multiple faults diagnosis for ocean-going marine diesel engines based on different neural network algorithms
• MCE-based Direct FTC Method for Dynamic Positioning of Underwater Vehicles with Thruster Redundancy
• Overview of Fault Tolerant Techniques in Underwater Sensor Networks
• Adaptive Fault-tolerant Control of Underwater Vehicles with Thruster Failures