A Hybrid Approach of Data-Driven and Signal Processing-Based Methods for Fault Diagnosis of Hydraulic Rock Drill

##plugins.themes.bootstrap3.article.main##

##plugins.themes.bootstrap3.article.sidebar##

Published Oct 28, 2022
Hye Jun Oh
Jinoh Yoo Sangkyung Lee Minseok Chae Jongmin Park Byeng D Youn

Abstract

This study presents a method for fault diagnosis of the hydrostatic rock drill. Hydraulic rock drill suffers from domain discrepancy due to harsh environment and indivisible difference, which leads to difficulty in diagnosing fault. To overcome these problems, we propose a fusion method of data-driven-based method and signal process-based method. In the case of the data-driven based method, the overall fault classification was performed using domain adaptation, metric learning, and pseudo label-based deep learning methods, and the signal process-based method was diagnosed for a specific fault by generating a reference signal. As a result, the fault diagnosis performance was 100%, and it was able to perform well even in domain discrepancy.

How to Cite

Oh, H. J., Yoo, J., Lee, S., Chae, M., Park, J., & Youn, B. D. (2022). A Hybrid Approach of Data-Driven and Signal Processing-Based Methods for Fault Diagnosis of Hydraulic Rock Drill. Annual Conference of the PHM Society, 14(1). Retrieved from https://papers.phmsociety.org/index.php/phmconf/article/view/3408
Abstract 599 |

##plugins.themes.bootstrap3.article.details##

Keywords

Fault Diagnosis, Hydraulic Rock Drill, Hybrid Approach, Domain Adaptation

Section
Data Challenge Papers