Research on fast identification method of tunnel cable state based on sparse online hybrid Gaussian multi-classification algorithm
Open Access
- 15 January 2025
- conference paper
- Published by SPIE-Intl Soc Optical Eng
- Vol. 13513, 61-365
Abstract
In this paper, a fast tunnel cable state identification method based on a sparse online hybrid Gaussian multi-classification algorithm is proposed for real-time monitoring of the state of tunnel cables and their changes. The method first reduces the size of the training set using the data subset approximation method, then uses the hybrid Gaussian multi-classification algorithm to train the tunnel cable state fast identification model, and finally performs the parameter update online when necessary. The proposed method takes less time in the model training phase, and the trained tunnel cable state fast identification model can maintain good recognition accuracy under complex operating conditions and can be updated quickly.Keywords
This publication has 19 references indexed in Scilit:
- Enhanced artificial intelligence technique for soft fault localization and identification in complex aircraft microgridsEngineering Applications of Artificial Intelligence, 2023
- Efficient finite element modelling of helical strand cables utilising periodicityInternational Journal of Mechanical Sciences, 2023
- Knowledge-informed Variational Bayesian Gaussian mixture regression model for predicting mixed oil lengthEnergy, 2023
- A mixed-categorical correlation kernel for Gaussian processNeurocomputing, 2023
- Health Assessment of HV Cable Based on Data Correlation Analysis and Fuzzy Rules Joint Algorithm2022
- Online multi-parameter sensing and condition assessment technology for power cables: A reviewElectric Power Systems Research, 2022
- Analysis on switching overvoltage and suppression method of cable joint in 500 kV cable lineEnergy Reports, 2021
- A CNN recognition method for early stage of 10 kV single core cable based on sheath currentElectric Power Systems Research, 2020
- Advanced Smart Grid Monitoring: Intelligent Cable Diagnostics using Neural Networks2020
- Overcurrent Tests and Numerical Simulations on a 66-kV-Class RE123 High-Temperature Superconducting Model CableIEEE Transactions on Applied Superconductivity, 2011