Fault diagnosis and restoration of ship structure monitoring signals based on machine learning
Image generated by Gemini AI
A new fault diagnosis and restoration method for ship monitoring systems leverages machine learning, incorporating wavelet transforms and a specialized neural network. Achieving over 98% diagnostic accuracy across various positions, it outperforms LSTM and Random Forest in noise resistance. The method significantly reduces RMSE for bias, drift, and impulse faults, enhancing the reliability of structural health monitoring and supporting safer intelligent navigation.
Machine Learning Enhances Ship Structural Health Monitoring
A novel fault diagnosis and restoration method utilizing machine learning has been developed to improve the reliability of ship structural health monitoring systems. This approach aims to mitigate the risk of signal faults that can lead to misjudgments about a ship's structural integrity.
Performance Analysis
To evaluate the effectiveness of this method, an impact analysis was conducted using monitoring data from a model test. Results revealed a diagnostic rate exceeding 98% across various monitoring positions, indicating robust performance in fault detection. The method demonstrated superior noise resistance compared to existing algorithms such as Long Short-Term Memory (LSTM) and Random Forest.
Restoration Efficacy
The method also excels in signal restoration. It achieved significant reductions in root mean square error (RMSE) across different fault types:
- 73.86% reduction for bias faults
- 75.49% reduction for drift faults
- 19.55% reduction for impulse faults
This improvement in restoration efficacy underscores the method's capability to enhance the stability of ship structural health monitoring systems, crucial for safe navigation.
Related Topics:
📰 Original Source: https://doi.org/10.21278/brod77301
All rights and credit belong to the original publisher.