DOI QR코드

DOI QR Code

참조점의 불규칙적 배치를 통한 PIC보의 하중 충실도 향상에 관한 연구

Load Fidelity Improvement of Piecewise Integrated Composite Beam by Irregular Arrangement of Reference Points

  • Ham, Seok Woo (Department of Mechanical Engineering, Graduated School, Kongju National University) ;
  • Cho, Jae Ung (Department of Mechanical Engineering, Graduated School, Kongju National University) ;
  • Cheon, Seong S. (Department of Mechanical Engineering, Graduated School, Kongju National University)
  • 투고 : 2019.10.07
  • 심사 : 2019.10.31
  • 발행 : 2019.10.31

초록

Piecewise Integrated Composite (PIC) 보는 하중 유형에 따라 구간을 나누어, 각 구간마다 하중 유형에 강한 복합재료의 적층 순서를 배열한 보이다. 본 연구에서는 보의 거동을 고려하여 PIC 보의 구간을 머신 러닝을 통해 나누어 기존에 제시되었던 PIC 보에 비해 우수한 굽힘 특성을 갖게 하는 것이 목적이다. FE 모델의 240개 요소가 참조점으로 선택되었다. 선행 유한요소해석은 머신 러닝의 학습데이터 생성을 위하여 규칙적으로 분포된 참조점에서 3축 특성 값(Triaxiality)으로 나타냈다. 3축 특성 값은 인장, 압축 그리고 전단의 하중유형을 나타낸다. 머신 러닝 모델은 하이퍼파라미터(Hyperparameter)와 학습데이터로 구성되었으며, 하이퍼파라미터 튜닝을 통해 적절한 하중 충실도를 도출하였지만, 거동이 큰 보의 옆면에서는 적절하지 않은 하중 충실도가 도출되었다. 이를 해결하기 위하여 고르게 배치한 참조점을 보의 거동에 따라 배치하여 학습 데이터를 얻었고, 머신 러닝 모델이 생성되었다. 앞서 생성된 머신 러닝 모델을 통하여 보가 매핑 되었고, PIC 보에 대하여 유한요소 해석을 진행한 결과, 기존에 제시되었던 PIC 보에 비해 최대하중과 흡수 에너지가 커지는 특성이 나타났다.

Piecewise integrated composite (PIC) beam has different stacking sequences for several regions with respect to their superior load-resisting capabilities. On the interest of current research is to improve bending characteristics of PIC beam, with assigning specific stacking sequence to a specific region with the help of machine learning techniques. 240 elements of from the FE model were chosen to be reference points. Preliminary FE analysis revealed triaxialities at those regularly distributed reference points to obtain learning data creation of machine learning. Triaxiality values catagorise the type of loading i.e. tension, compression or shear. Machine learning model was formulated by learning data as well as hyperparameters and proper load fidelity was suggested by tuned values of hyperparameters, however, comparatively higher nonlinearity intensive region, such as side face of the beam showed poor load fidelity. Therefore, irregular distribution of reference points, i.e., dense reference points were distributed in the severe changes of loading, on the contrary, coarse distribution for rare changes of loading, was prepared for machine learning model. FE model with irregularly distributed reference points showed better load fidelity compared to the results from the model with regular distribution of reference points.

키워드

참고문헌

  1. Jeoung, S.K., and Ha, J.U., "Lightweighting Technology Development and Trends in Automotive," Auto Journal, Vol. 40, No. 8, 2018, pp. 30-34.
  2. Jeoung, S.K., and Lee, P.C., "Trends of Lightweight Automotive using Polymer Based Materials," Auto Journal, Vol. 36, No. 3, 2014, pp. 27-32.
  3. Lee, H.W., Ha, J.W., Ko, K.Y., and Lee, P.C., "Trends of Polymer based Lightweight Materials Technology in Automotive," Polymer Science and Technology, Vol. 29, No. 9, 2018, pp. 505-509.
  4. Kim, D.Y., Cheon, S.S., Kim, B.K., and Jong, C.H., "Absorption Energy Analysis due to the Cross-Section Changes in the Lightweight Seat Frame," Proceedings of KSPE 2015 Autumn Conference, Jeongseon, Korea, 2015, pp. 98-99.
  5. Na, H.J., Chun, J.S., and Cho, K.S., "Development of CFRP Tubes for the Light-Weight Propeller Shaft of 4WD SUV Vehicles," Journal of the Korean Society of Manufacturing Process Engineers, Vol. 17, No. 4, 2018, pp. 32-38. https://doi.org/10.14775/ksmpe.2018.17.4.032
  6. Chun, D.M. and Ahn, S.H., "Change of Mechanical Properties of Injection-Molded Glass-Fiber-Reinforced Plastic (GFRP) According to Temperature and Water Absorption for Vehicle Weight Reduction," Transactions of the Korean Society of Mechanical Engineers - A, Vol. 37, No. 2, 2013, pp. 199-204. https://doi.org/10.3795/KSME-A.2013.37.2.199
  7. Belingardi, G., Beyene, A.T., and Koricho, E.G. "Geometrical Optimization of Bumper Beam Profile Made of Pultruded Composite by Numerical Simulation," Composite Structures, Vol. 102, 2013, pp. 217-225. https://doi.org/10.1016/j.compstruct.2013.02.013
  8. Kim, D.H., Kim, H.G., and Kim, H.S., "Design Optimization and Manufacture of Hybrid Glass/carbon Fiber Reinforced Composite Bumper Beam for Automobile Vehicle," Composite Structures, Vol. 131, 2015, pp. 742-752. https://doi.org/10.1016/j.compstruct.2015.06.028
  9. Jeong, C.H., Ham, S.W., Kim, G.S. and Cheon, S.S. "Development of the Piecewisely-integrated Composite Bumper Beam Based on the IIHS Crash Analysis," Composite Research, Vol. 31, No. 1, 2018, pp. 37-41. https://doi.org/10.7234/composres.2018.31.1.037
  10. Ham, S.W., Cheon, S.S., and Jeong, K.Y., "Strength Optimization of Piecewise Integrated Composite Beam through Machine Learning," Transactions of the Korean Society of Mechanical Engineers - A, Vol. 43, No. 8, 2019, pp. 521-528. https://doi.org/10.3795/ksme-a.2019.43.8.521
  11. Lantz, B., Machine Learning with R, 2nd ed., Packt Publishing, UK, 2015.
  12. Bai, Y., and Wierzbicki, T., "A New Model of Metal Plasticity and Fracture with Pressure and Lode Dependence," International Journal of Plasticity, Vol. 24, 2008, pp. 1071-1096. https://doi.org/10.1016/j.ijplas.2007.09.004