DOI QR코드

DOI QR Code

Study on Improving Learning Speed of Artificial Neural Network Model for Ammunition Stockpile Reliability Classification

저장탄약 신뢰성분류 인공신경망모델의 학습속도 향상에 관한 연구

  • Received : 2020.03.13
  • Accepted : 2020.06.05
  • Published : 2020.06.30

Abstract

The purpose of this study is to improve the learning speed of an ammunition stockpile reliability classification artificial neural network model by proposing a normalization method that reduces the number of input variables based on the characteristic of Ammunition Stockpile Reliability Program (ASRP) data without loss of classification performance. Ammunition's performance requirements are specified in the Korea Defense Specification (KDS) and Ammunition Stockpile reliability Test Procedure (ASTP). Based on the characteristic of the ASRP data, input variables can be normalized to estimate the lot percent nonconforming or failure rate. To maintain the unitary hypercube condition of the input variables, min-max normalization method is also used. Area Under the ROC Curve (AUC) of general min-max normalization and proposed 2-step normalization is over 0.95 and speed-up for marching learning based on ASRP field data is improved 1.74 ~ 1.99 times depending on the numbers of training data and of hidden layer's node.

본 연구에서 저장탄약 신뢰성평가(ASRP: Ammunition Stockpile Reliability Program)의 데이터 특성을 고려하여 입력변수를 줄이는 정규화기법을 제안함으로써 분류성능의 저하 없이 저장탄약 신뢰성분류 인경신경망모델의 학습 속도향상을 목표로 하였다. 탄약의 성능에 대한 기준은 국방규격(KDS: Korea Defense Specification)과 저장탄약 시험절차서(ASTP: Ammunition Stockpile reliability Test Procedure)에 규정되어 있으며, 평가결과 데이터는 이산형과 연속형 데이터가 복합적으로 구성되어 있다. 이러한 저장탄약 신뢰성평가의 데이터 특성을 고려하여 입력변수는 로트 추정 불량률(estimated lot percent nonconforming) 또는 고장률로 정규화 하였다. 또한 입력변수의 unitary hypercube를 유지하기 위하여 최소-최대 정규화를 2차로 수행하는 2단계 정규화 기법을 제안하였다. 제안된 2단계 정규화 기법은 저장탄약 신뢰성평가 데이터를 이용하여 비교한 결과 최소-최대 정규화와 유사하게 AUC(Area Under the ROC Curve)는 0.95 이상이었으며 학습속도는 학습 데이터 수와 은닉 계층의 노드 수에 따라 1.74 ~ 1.99 배 향상되었다.

Keywords

References

  1. K.S. Yoon, J.C, Lee, "A Case Study on the Reliability Assessment of Stockpile Ammunition", Journal of the Korean Society for Quality Management , Vol.40, No.3, pp.259-269, 2012. DOI: https://doi.org/10.7469/JKSQM.2012.40.3.259
  2. K.S. Yoon, S.W. Park, "A Study on the Estimation of Shelf-Life for 155mm propelling charge KM4A2 using ASRP's data" Journal of the Korean Society for Quality Management, Vol.42, No.3, pp.291-300, 2014. DOI: https://doi.org/10.7469/JKSQM.2014.42.3.291
  3. D.N. Lee, K.S. Yoon, "A Study on the Estimation of Shelf Life for Fuze MTSQ KM577A1 from ASRP Data" Journal of Applied Reliability , Vol.18, No.1, pp.56-65, 2018. DOI: https://doi.org/10.33162/JAR.2018.03.18.1.56
  4. Y.C. Noh, D.N. Lee, N.W. Cho, "A Study on Determinants of Stockpile Ammunition Using Data Mining", Proceedings of The Korean Society for Quality Management, p.20, October 2018.
  5. S. Fritsch, F. Guenther, M. N. Wright, "neuralnet: Training of Neural Networks", R package version 1.44.2, 2019. https://CRAN.R-project.org/package=neuralnet
  6. K.H. Cho, "A Study on the Self-life Estimation of the Propellant KM10 by using High Temperature Acceleration Aging Tests", Journal of the Korea Academia-Industrial Cooperation Society, Vol.11, No.5, pp.1735-1740, 2010. DOI: https://doi.org/10.5762/KAIS.2010.11.5.1735
  7. F. Liu, H. Gong, L. Cai, K. Xu, "Prediction of Ammunition Storage Reliability Based on Improved Ant Colony Algorithm and BP Neural Network", Complexity, Vol.2019, Article ID 5039097, pp.13, 2019. DOI: https://doi.org/10.1155/2019/5039097
  8. B. Lantz, "Machine Learning with R : Korean Edition," Packt Packt Publishing Ltd., p.306, p.447, 2017.
  9. K.A. Mason, P.H. Raven, S.R. Singer, "Biology 11th," McGraw Hill Education, p.890, 2017.
  10. "Sampling Procedures and Tables for Inspection by Variables for Percent Nonconforming," American Society for Quality, ANSI/ASQ Z1.9, 2008.
  11. T. Fawcett, "ROC Graphs : Notes and Practical Considerations for Data Mining Researchers," Technical Report, HP Laboratories, US, p.3, 2003.