하이브리드 LSTM기반 제조로봇 고장예지 시스템의 설계 및 구현 = Design and Implementation of Fault Prognostics System for Manufacturing Facility Robot based on Hybrid LSTM

다국어 초록 (Multilingual Abstract)

The data collected by the sensor in the manufacturing facility may include missing values due to the malfunction of the sensor and measurement information. These difficulties may cause suffering from obtaining fault data. In this paper, we propose a method for forecasting the failure of a robot in a manufacturing facility under the limited conditions in which it is difficult to secure failure data generated by the manufacturing facility. In this proposed method, outlier and missing values were refined to utilize the noise-free data, Z values and thresholds were set. And then data preprocessing was performed using linear interpolation. The proposed model learns based on the steady-state data of manufacturing facilities. Then, the input vector of preprocessed data was sampled using a hybrid long short term memory (H-LSTM) circulatory neural network model and used for learning. In order to verify the proposed method, data were collected based on two fault conditions and the experiments were performed based on the two fault conditions. The degree of abnormality is expressed by measuring the root mean square error(RMSE) between the output of each state data and the prediction result. The experiments verified the accuracy of the proposed failure prediction technique.

참고문헌 (Reference)

1 최성원, "시험장비의 예방보전을 위한 RCM 적용 방안에 대한 연구" 대한설비관리학회 24 (24): 5-15, 2019

2 "Thingsboard"

3 H, Hejazi, "Survey of Platforms for Massive IoT" 1-8, 2018

4 A. Graves, "Speech Recognition with Deep Recurrent Neural Networks" 1-5, 2013

5 I. Sutskever, "Sequence to Sequence Learning with Neural Networks" 3104-3112, 2014

6 A. Gruslys, "Memory-Efficient Backpropagation Through Time" 4132-4140, 2016

7 J. Zhang, "Long Short-term Memory for Machine Remaining Life Prediction" 48 (48): 78-86, 2018

8 S. Hochreiter, "LSTM can solve hard long time lag problems" 9 : 473-479, 1997

9 A.A. Jaber, "Fault Diagnosis of Industrial Robot Bearings Based on Discrete Wavelet Transform and Artificaial Neural Networrk" 7 (7): 1-13, 2016

10 E. Khalastchi, "Fault Detection and Diagnosis in Multi-Robot Systems: A Survey" 19 (19): 1-19, 2019

11 L. Sidhom, "Fault Actuators detection Based on Inputs-Outputs Data Mapping: Application to a Robot Manipulator" 7 (7): 328-336, 2017

12 I. Eski, "Falut Detection on Robot Manipulators using Artificial Neural Networks" 27 (27): 115-123, 2011

13 J. Chung, "Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling" 1-9, 2014

14 Helbing G, "Deep Learning for Fault Detection in Wind Turbines" 98 (98): 189-198, 2018

15 J. Jiang, "Bibliographical review on reconfigurable fault-tolerant control systems" 32 (32): 229-252, 2008

16 J. Bergstra, "Algorithms for Hyper-Parameter Optimization" 2546-2554, 2011

17 E. Khalastchi, "A Senor-based Approach for Fault Detection and Diagnosis for Robotic Systems" 42 (42): 1231-1248, 2018

18 F. Jia, "A Neural Network Constructed by Deep Learning Technique and its Application to Intelligent Fault Diagnosis of Machines" 272 (272): 619-628, 2018