DOI QR코드

DOI QR Code

Detects abnormal behavior using motor power consumption

  • Kim, KiHwan (Dept. of Computer Science, Dongseo University) ;
  • Ryu, Su-Mi (Dept. of Computer Science, Dongseo University) ;
  • Kim, Min-Kyu (Dept. of Computer Science, Dongseo University) ;
  • Kang, Young-Jin (Dept. of Ubiquitous IT, Dongseo University) ;
  • Kim, HyunHo (Dept. of Ubiquitous IT, Dongseo University) ;
  • Lee, HoonJae (Dept of Information and Communication Engineering, Dongseo University) ;
  • Lee, Jin-Heung (Daun Information & Communication Co)
  • Received : 2018.10.01
  • Accepted : 2018.10.22
  • Published : 2018.10.31

Abstract

In this paper, we used LSTM as a method to detect abnormal behavior of motors. We fixed the high layout size to 1 and changed the range of the input values and the neural network structure to see what change in power consumption prediction. Now, as the fourth industrial revolution era, smart factories are attracting attention. All the physical actions of smart factories are done using motors. Continuous monitoring of motor malfunctions helps to detect malfunctions and efficient operation. However, it is difficult to acquire the power consumption constantly due to the influence of the noise. We have experimented with a simple experimental environment, a method of predicting similarity to input data by adjusting the range of the input data or by changing the neural network structure.

Keywords

References

  1. Jo Seongjun, Gang Seokho, “Industrial Applications of Machine Learning (Artificial Intelligence),” Industrial Engineering Magazine, Vol. 23, No. 2, pp. 34-38, June. 2016.
  2. Gong Mansik, Chae Hongjun, Yu Bohyeon, “Trends and Prospects of Internet (IoT) Technology,” Journal of the KSME, Vol. 56, No. 2, pp. 32-36, Feb. 2016.
  3. Choe Yerim, Gim Gwanho, “Artificial intelligence overview and application examples,” Industrial Engineering Magazine, Korean Institute of Electrical Engineers, Vol. 23, No. 2, pp. 23-29, May. 2016.
  4. Da Silva, I. N., Spatti, D. H., Flauzino, R. A., Liboni, L. H. B., & dos Reis Alves, S. F. "Artificial neural networks," Cham: Springer International Publishing, 2017.
  5. Gu Hyeongil, “Artificial Intelligence and Deep Running Trends,” The Korean Institute of Electrical Engineers, Vol. 67, No. 7, pp. 7-12, July. 2018.
  6. Caterini, A. L., & Chang, D. E. "Deep Neural Networks in a Mathematical Framework," Springer International Publishing, 2018.
  7. Recurrent Neural Networks Tutorial, Part 1 - Introduction to RNNs, http://www.wildml.com/2015/09/recurrent-neural-networks-tutorial-part-1-introduction-to-rnns/
  8. Yann LeCun, Yoshua Bengio, Geoffrey Hin ton, "Deep learning", nature, Vol. 521, pp. 436-444, 28, May. 2015. https://doi.org/10.1038/nature14539
  9. Goodfellow, I., Bengio, Y., Courville, A., & Bengio, Y. "Deep learning (Vol. 1)," Cambridge: MIT press, 2016.
  10. Ian Goodfellow, Yoshua Bengio, Aaaron Courville, "Deep Learning," MIT Press, pp. 196, 2016.
  11. Tallec, C., & Ollivier, Y. "Unbiasing Truncated Backpropagation Through Time," arXiv preprint arXiv:1705.08209, 2017.
  12. Pascanu, Razvan, Tomas Mikolov, and Yoshua Bengio. "On the difficulty of training recurrent neural networks," International Conference on Machine Learning, 2013.
  13. Greff, K., Srivastava, R. K., Koutnik, J., Steunebrink, B. R., & Schmidhuber, J. "LSTM: A search space odyssey," IEEE transactions on neural networks and learning systems, Vol. 28, No. 10, pp. 2222-2232, 2017. https://doi.org/10.1109/TNNLS.2016.2582924
  14. Understanding LSTM Networks, http://colah.github.io/posts/2015-08-Understanding-LSTMs/