Citation: Kim, H.; Lee, H.; Kim, S.W.
Current Only-Based Fault Diagnosis
Method for Industrial Robot Control
Cables. Sensors 2022, 22, 1917.
https://doi.org/10.3390/s22051917
Academic Editors: Kim Phuc Tran,
Athanasios Rakitzis
and Khanh T. P. Nguyen
Received: 2 February 2022
Accepted: 26 February 2022
Published: 1 March 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
Current Only-Based Fault Diagnosis Method for Industrial
Robot Control Cables
Heonkook Kim
1,2
, Hojin Lee
1
and Sang Woo Kim
1,
*
1
Department of Electrical Engineering, Pohang University of Science and Technology, 77 Cheongam-Ro,
Nam-Gu, Pohang 37673, Korea; kimhk85@postech.ac.kr (H.K.); suvvus@postech.edu (H.L.)
2
Hyundai Robotics Co., Ltd., 50, Techno Sunhwan-ro 3-gil, Yuga, Dalseong-gun, Daegu 43022, Korea
* Correspondence: swkim@postech.edu; Tel.: +82-54-279-2237; Fax: +82-54-279-2903
Abstract:
With the growth of factory automation, deep learning-based methods have become popular
diagnostic tools because they can extract features automatically and diagnose faults under various
fault conditions. Among these methods, a novelty detection approach is useful if the fault dataset
is imbalanced and impossible reproduce perfectly in a laboratory. This study proposes a novelty
detection-based soft fault-diagnosis method for control cables using only currents flowing through
the cables. The proposed algorithm uses three-phase currents to calculate the sum and ratios of
currents, which are used as inputs to the diagnosis network to detect novelties caused by soft faults.
Autoencoder architecture is adopted to detect novelties and calculate anomaly scores for the inputs.
Applying a moving average filter to anomaly scores, a threshold is defined, by which soft faults can
be properly diagnosed under environmental disturbances. The proposed method is evaluated in
11 fault scenarios. The datasets for each scenario are collected when an industrial robot is working.
To induce soft fault conditions, the conductor and its insulator in the cable are damaged gradually
according to the scenarios. Experiments demonstrate that the proposed method accurately diagnoses
soft faults under various operating conditions and degrees of fault severity.
Keywords:
autoencoder; control and instrumentation cable; fault diagnosis; industrial robot; novelty
detection; soft fault
1. Introduction
Control and instrumentation (C&I) cables are utilized in a wide range of industrial
applications, including nuclear power plants [
1
], ship power systems [
2
], vehicles [
3
], and
factory automation [
4
], owing to their vital role in the control of motors and instrumen-
tation of sensors. To maintain the stability of automated systems in various applications,
monitoring the health status of the C&I cable is crucial. Early and accurate diagnosis of
faults reduces unwanted system downtime and improves the system reliability, resulting in
the maximization of productivity. In modern automated manufacturing processes, a single
fault in an industrial machine could cause an entire production line stoppage because the
safety signals of machines are shared and controlled by a process control unit. Researchers
have studied the faults from industrial machines. A diagnosis method based on an artificial
neural network using most salient features has shown reliability in diagnosing [
5
]. In the
other hand, the modern robust control studies have shown that certain types of robot faults
can be overcome with compensation of perturbations [6].
Cable faults comprise hard faults and soft faults. Hard faults include open circuit
and short circuit, and soft faults are characterized by small impedance changes. A local
modification of the cable (soft fault) due to a harsh environment could be transformed into
hard fault by subsequent partial damage of the components, such as the cable conductors,
coatings, and shield [
7
]. The worst possible case occurs when the machine is still working
without any indication of cable faults, while the cable is partially damaged. This is a
common practical soft-fault problem that occurs in automated factories. Therefore, timely
Sensors 2022, 22, 1917. https://doi.org/10.3390/s22051917 https://www.mdpi.com/journal/sensors
