Iranian Journal of Manufacturing Engineering

Iranian Journal of Manufacturing Engineering

Energy Absorption Analysis in an Auxetic Lattice Structure Using Artificial Neural Network Machine Learning and Genetic Algorithm

Document Type : Original Article

Authors
Noushin Ghannadi 1
Amin Farrokhabadi 2
Shahram Hosseini 3
1 PhD Student, Department of Mechanical Engineering, Tarbiat Modares University, Tehran, Iran
2 Associate Professor, Department of Mechanical Engineering, Tarbiat Modares University, Tehran, Iran
3 PhD Graduate, Department of Mechanical Engineering, Tarbiat Modares University, Tehran, Iran
10.22034/ijme.2024.470912.1993
Abstract
Today, with the advancement of technology, improvement and quality of life, artificial intelligence has found a special place in the world. In this research, the optimization of the curved mesh structure made of polylactic acid has been discussed. Mesh structures are widely used in various industries due to their advantages such as high energy absorption. The geometric parameters of this structure can have a noticeable effect on the amount of energy absorption of these structures. In this regard, optimization of geometrical parameters has been done using genetic algorithm. In this research, the parameters of radius of curvature R1, angle of curvature Ɵ1, radius of curvature R2, angle of curvature Ɵ2, length L relative to the energy absorbed by the structure, maximum force and modulus of elasticity have been optimized. A large number of optimization parameters confirm the use of genetic algorithm for the optimization process of this structure. Optimization using genetic algorithm requires the existence of an objective function to which the geometric parameters are optimized. This objective function should be a continuous function that can produce the necessary outputs for each geometric parameter. In this research, artificial neural network has been used to construct the objective function. The artificial neural network makes it possible to create a continuous function with a limited number of inputs and outputs that can produce suitable outputs for receiving different inputs. After extracting the optimal geometric parameters, the optimal structure was made using a 3D printer and subjected to quasi-static pressure testing.
Keywords
Auxetic Structure
Artificial Neural Network
Genetic Algorithm
Energy Absorption
Artificial Intelligence
[1] El Kadi H, Al-Assaf Y. Prediction of the fatigue life of unidirectional glass fiber/epoxy composite laminae using different neural network paradigms. Composite Structures. 2002 Feb 1;55(2):239-46. doi: 10.1016/S0263-8223(01)00152-0
[2] Zhang Z, Friedrich K. Artificial neural networks applied to polymer composites: a review. Composites Science and technology. 2003 Nov 1;63(14):2029-44. doi: 10.1016/S0266-3538(03)00106-4
[3] Thankachan T, Prakash KS, Jothi S. Artificial neural network modeling to evaluate and predict the mechanical strength of duplex stainless steel during casting. Sādhanā. 2021 Dec;46:1-2. doi: 10.1007/s12046-021-01742-w
[4] Lee S, Park S, Kim T, Lieu QX, Lee J. Damage quantification in truss structures by limited sensor-based surrogate model. Applied Acoustics. 2021 Jan 15;172:107547.doi: 10.1016/j.apacoust.2020.107547
[5] Balokas G, Czichon S, Rolfes R. Neural network assisted multiscale analysis for the elastic properties prediction of 3D braided composites under uncertainty. Composite Structures. 2018 Jan 1;183:550-62.doi: 10.1016/j.compstruct.2017.06.037
[6] Ang JY, Majid MA, Nor AM, Yaacob S, Ridzuan MJ. First-ply failure prediction of glass/epoxy composite pipes using an artificial neural network model. Composite Structures. 2018 Sep 15;200:579-88.doi: 10.1016/j.compstruct.2018.05.139
[7] Ghanadi N, Farrokhabadi A, Hosseini S. Optimizing the failure analysis of a network structure using artificial neural network and genetic algorithm. Iranian Journal of Manufacturing Engineering. 2023 Jan 21;9(11):35-44. doi: 10.22034/ijme.2023.400071.1789 [In Persian]
[8] Yan S, Zou X, Ilkhani M, Jones A. An efficient multiscale surrogate modelling framework for composite materials considering progressive damage based on artificial neural networks. Composites Part B: Engineering. 2020 Aug 1;194:108014.doi: 10.1016/j.compositesb.2020.108014
[9] Serjouei A, Yousefi A, Jenaki A, Bodaghi M, Mehrpouya MJ. 4D printed shape memory sandwich structures: experimental analysis and numerical modeling. Smart Materials and Structures. 2022 Apr 8;31(5):055014.doi: 10.1088/1361-665X/ac60b5
[10] Zeinolabedin-Beygi A, Naeini HM, Talebi-Ghadikolaee H, Rabiee AH, Hajiahmadi S. Predictive modeling of spring-back in pre-punched sheet roll forming using machine learning. The Journal of Strain Analysis for Engineering Design. 2024 Oct;59(7):463-74. doi: 10.1177/03093247241263685
[11] Akhoundi B, Khosravian E, Modanloo V. Deposition of continuous glass fibers on a curved surface by 3D printer based on fused filament fabrication technology. Iranian Journal of Manufacturing Engineering. 2024 Jan 21;10(11):16-23. doi: 10.22034/ijme.2024.429126.1885 [In Persian]
[12] Imani L, Rahmani Hanzaki A, Hamzeloo SR, Davoodi B. Modeling and optimization of cutting force and surface roughness in the milling process of Inconel 738 by Neural Network and Genetic Algorithm. Iranian Journal of Manufacturing Engineering. 2019 Oct 23;6(5):25-38. doi: 10.1177/0954405419889204 [In Persian]
[13] Deilami Azodi H, Badparva H, Zeinolabedin Beygi A. Optimizing AA3105-St12 two-layer sheet in incremental sheet forming process using neural network and multi-objective genetic algorithm. Modares Mechanical Engineering. 2022 Jan 10;22(2):121-32. doi: 20.1001.1.10275940.1400.22.2.4.7
[14] Siahvashi A, Shahbazi M, Niknam SA. The use of a neuro-fuzzy network coupled with meta-heuristic learning methods to predict surface roughness in the machining of aluminum alloys. Iranian Journal of Manufacturing Engineering. 2024 Jan 21;10(11):49-60. doi: 10.22034/ijme.2024.412021.1816 [In Persian]