Hydraulic Rock Drill Fault Classification Using X−Vectors

• DOI: 10.3390/math11071724
• Publication Date: 2023-04-05T05:39:26Z
• Authors: Huixuan Ling; Tian Gao; Tao Gong; Jiangzhao Wu; Liang Zou
• Keywords: Robustness

Hydraulic rock drills are widely used in drilling, mining, construction, and engineering applications. They typically operate harsh environments with high humidity, large temperature differences, vibration. Under the influence of environmental noise operational patterns, distributions data collected by sensors for different operators equipment differ significantly, which leads to difficulty fault classification hydraulic drills. Therefore, an intelligent robust method is highly desired. In this paper, we propose a technique based on deep learning. First, considering strong robustness x−vectors features extracted from time series, employ end−to−end model realize joint optimization feature extraction classification. Second, overlapping clipping applied during training process, further improves our model. Finally, focal loss focus difficult samples, their accuracy. The proposed obtains accuracy 99.92%, demonstrating its potential drill