Multi-objective optimization for 3D printed origami crash box cell based on artificial neural networks and NSGA-II

• DOI: 10.23977/jmpd.2023.070401
• Publication Date: 2023-10-19T13:40:15Z
• Authors: Jiahao Li; Minjie Qu; Zhiyu Jiang
• Keywords: Specific energy

Composite structures are increasingly used in the automotive industry due to their lightweight and specific energy absorption capabilities, while 3D printing is also widely because of high efficiency precision. Recently, Origami crash box (OCB) has been proposed as an absorber for automobiles low initial peak load average load. Experiments theory have shown that energy-absorbing effect OCB will change significantly with size. Since composed multiple cells, therefore, it necessary develop a model can predict according scale cell. And this utilized optimize size maximize its capability reducing force. This paper explores printed OCB, which made carbon fiber-reinforced nylon same wight thickness stable surface area 14400mm2. The Artificial Neural Network (ANN) Mean Squared Error (MSE) measure accuracy established non-linear behavior cell at different then Non-dominated Sorting Genetic Algorithm (NSGA-II) Peak Crush Force (PCF) Energy Absorption (EA) optimization metrics, applied complete multi-objective optimization. ANN precisely predicts variation displacement MSE 0.046kN², error 5.97J PCF 0.17kN. A configuration generated by NSGA-II shows superior performance than standard In terms prediction, there 13.5% decrease PCF, from 2.75 kN 2.38 kN, EA experiences 7.8% increase, rising 34.5 J 37.2 J. experimental results, exhibits 14% reduction, decreasing 3.08 2.65 14.3% climbing 30.61 35