Domain Adaptation based Fault Diagnosis under Variable Operating Conditions of a Rock Drill



Published Oct 28, 2022
Yong Chae Kim
Taehun Kim
Jin Uk Ko
Jinwook Lee
Keon Kim


Fault diagnosis using a data-driven approach is an essential technology for the safety and maintenance of a rock drill. However, since the signals acquired from the rock drill have different distributions due to variable operating conditions, the classification performance of the data-driven method decreases; this is called the domain shift issue. This paper proposes a new domain adaptation-based fault diagnosis scheme to solve the problem. The proposed method introduces a data cropping technique to mitigate the difference in the length of the data measured from the rock drill for each impact cycle. To extract invariant features for all operating conditions, the proposed method combines two methods: the domain adversarial neural network and minimization of the maximum mean discrepancy (MMD) between the features from different domains. In addition, a soft voting ensemble is used to reduce the model uncertainty. The proposed method shows superior performance when validated with the rock drill dataset and ranked 2nd place in the 2022 PHM Conference Data Challenge.

How to Cite

Kim, Y. C., Kim, T., Ko, J. U., Lee, J., & Kim, K. (2022). Domain Adaptation based Fault Diagnosis under Variable Operating Conditions of a Rock Drill. Annual Conference of the PHM Society, 14(1). Retrieved from
Abstract 537 |



Rock Drill, Fault Diagnosis, Domain Adaptation, Deep-learning

Erik, J, Erik, F., Mattias, K., & Robert, P. (2021). Fault Identification in Hydraulic Rock Drills from Indirect Measurement During Operation. IFAC-PapersOnLine, vol. 54(11), pp. 73-78.
Li, B., Chow, MY., Tipsuwan, Y., & Hung, J.C. (2000) Neural-network-based motor rolling bearing fault diagnosis. IEEE transactions on industrial electronics, vol.47(5), pp. 1060-1069.
Guo, X., Chen, L., & Shen, C. (2016). Hierarchical adaptive deep convolution neural network and its application to bearing fault diagnosis. Measurement, vol. 93, pp. 490-502.
Ganin, Y., Ustinova, E., Ajakan, H., Germain, P., Larochelle, H., Laviolette, F., Marchand, M., & Lempitsky, V. (2016). Domain-adversarial training of neural networks. The journal of machine learning research, vol.17(1), pp. 2096-2030.
Jiao, J., Zhao, M., & Lin, J. (2019). Unsupervised adversarial adaptation network for intelligent fault diagnosis. IEEE Transactions on Industrial Electronics, vol. 67(11), pp. 9904-9913.
Guo, L., Lei, Y., Xing, S., Tan, T., & Li, N. (2018). Deep convolutional transfer learning network: A new method for intelligent fault diagnosis of machines with unlabeled data. IEEE Transactions on Industrial Electronics, vol.66(9) pp. 7316-7325.
Borgwardt, KM., Gretton, A., Rasch, MJ., Kriegel, HP., Schölkopf, B., Karsten M., & Smola, AJ. (2006) Integrating structured biological data by kernel maximum mean discrepancy. Bioinformatics, vol.22(14) pp. 49-57.
Data Challenge Papers