Piecewise Integrated Composite Bumper Beam Design Method with Machine Learning Technique
Keywords:
Machine learning, Composite materials, PIC (piecewise integrated composite), Bumper beamAbstract
In this study, the Piecewise Integrated Composite (PIC) design method with machine learning that automatically assigns different stacking sequences according to loading types was applied to bumper beam. The input value and labels of the training data for applying machine learning were defined as coordinates and loading types of reference elements that are part of the total elements, respectively. In order to compare the two-dimensional (2D) and three-dimensional (3D) implementation method, which are methods of representing coordinate value, training data were generated, and machine learning models were trained with each method. The 2D implementation method divided finite element (FE) model into each face, and generating learning data and training machine learning models accordingly. The 3D implementation method involved training one machine learning model by generating training data from the entire finite element model. The hyperparameter were tuned to optimal values through the Bayesian algorithm, and the k-nearest neighbors (k-NN) classification method showed the highest prediction rate and area under the curve-receiver operation characteristic (AUC-ROC) among the tuned models. The 3D implementation method revealed higher performance than the 2D implementation method. The loading type data predicted through the machine learning model were mapped to the finite element model and comparatively verified through FE analysis. It was found that 3D implementation PIC bumper beam was superior to 2D implementation and uni-stacking sequence composite bumper.
References
Na HJ, Chun JS, Cho KS, 2018, Development of CFRP Tubes for the Light-Weight Propeller Shaft of 4WD SUV Vehicles. Journal of the Korean Society of Manufacturing Process Engineers, 17(4): 32–38.
Chun DM, Ahn SH, 2013, 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, 37(2): 199–204.
Cheon SS, Choi JH, Lee DG, 1995, Development of the Composite Bumper Beam for Passenger Cars. Composite Structures, 32(1–4): 491–499.
Belingardi G, Beyene AT, Koricho EG, 2013, Geometrical Optimization of Bumper Beam Profile Made of Pultruded Composite by Numerical Simulation. Composite Structures, 2013(102): 217–225.
Kim DH, Kim HG, Kim HS, 2015, Design Optimization and Manufacture of Hybrid Glass/Carbon Fiber Reinforced Composite Bumper Beam for Automobile Vehicle. Composite Structures, 131(1): 742–752.
Jeong CH, Ham SW, Kim GS, et al., 2018, Development of the Piecewisely-Integrated Composite Bumper Beam Based on the IIHS Crash Analysis. Composites Research, 31(1): 37–41.
Ham SW, Cheon SS, Jeong KY, 2019, Strength Optimization of Piecewise Integrated Composite Beam Through Machine Learning. Transactions of the KSME, A, 43(8): 521–528.
Ji SM, Ham SW, Choi JK, et al., 2021, Stiffness E Enhancement of Piecewise Integrated Composite Beam using 3D Training Data Set. Composites Research, 34(6): 394–399.
Li D, Gu M, Liu S, et al., 2022, Continual Learning Classification Method with the Weighted K-Nearest Neighbor Rule for Time-Varying Data Space Based on the Artificial Immune System. Knowledge-Based Systems, 240(15): 108145.
Ham SW, Cheon SS, 2020, Load Fidelity Improvement of Piecewise Integrated Composite Beam by Construction Training Data of k-NN Classification Model. Composites Research, 33(3): 108–114.
Shahhosseini M, Hu G, Pham H, 2022, Optimizing Ensemble Weights and Hyperparameters of Machine Learning Models for Regression Problems. Machine Learning with Applications, 7(15): 100251.
Liu X, Liu TQ, Feng P, 2022, Long-Term Performance Prediction Framework Based on XGBoost Decision Tree for Pultruded FRP Composites Exposed to Water, Humidity and Alkaline Solution. Composite Structures, 284(15): 115184.
Fayed HA, Atiya AF, 2021, Decision Boundary Clustering for Efficient Local SVM. Applied Soft Computing, 2021(110): 107628.
Lee SH, Mazumder J, Park JW, et al., 2020, Ranked Feature-Based Laser Material Processing Monitoring and Defect Diagnosis Using k-NN and SVM. Journal of Manufacturing Processes, 2020(55): 307–316.
Fakhrmoosavi F, Kamjoo E, Kavianipour M, et al., 2022, A Stochastic Framework Using Bayesian Optimization Algorithm to Assess the Network-Level Societal Impacts of Connected and Autonomous Vehicles. Transportation Research Part C: Emerging Technologies, 2022(139): 103663.
Tsai SW, Melo JDD, 2016, A Unit Circle Failure Criterion for Carbon Fiber Reinforced Polymer Composites. Composites Science and Technology, 123(8): 71–78.
Ham SW, Cho JU, Cheon SS, 2019, Load Fidelity Improvement of Piecewise Integrated Composite Beam by Irregular Arrangement of Reference Points. Composites Research, 23(5): 216–221.
