Failure Estimation of the Composite Laminates in Layup Optimization Using Finite Element Analysis and Deep Learning
- PDF / 589,177 Bytes
- 13 Pages / 595.276 x 790.866 pts Page_size
- 79 Downloads / 261 Views
TECHNICAL ARTICLE—PEER-REVIEWED
Failure Estimation of the Composite Laminates in Layup Optimization Using Finite Element Analysis and Deep Learning Alexandru S¸ erban
Submitted: 21 April 2020 / Accepted: 8 July 2020 ASM International 2020
Abstract Layup optimization of the composite laminates is a very complex problem due to the convoluted multidimensional solution space which is usually explored by addressing different heuristic methods from which the most reliable are the genetic algorithms (GA). The optimization process converges by evaluating a lot of layup configurations which imply that the evaluation should be not only robust but also very fast. The most accurate numerical tool used to simulate the mechanical behavior of the composite laminates is the finite element analysis (FEA) which unfortunately is a computational intensive method. Some studies proposed very fast FEA models specially designed for the layup optimization with the lower bound of the execution time determined by the global linear system solving. Other studies pushed this bound even lower using classical machine learning techniques trained with prior observations (layup configurations) evaluated with FEA. It has been shown that the trained models can successfully replace the computational intensive FEA. The results are very important because the optimization time is dramatically reduced, while the estimation errors induced by the statistical models are acceptable. In this paper, we propose different deep neural network architectures such as multilayer perceptron (MLP) and convolutional and recurrent neural networks (CNN and RNN) that significantly reduce the estimation errors. For example, the classification error reduces from 2% to zero compared to previous studies, for the same numerical example. Also, we use different sets of predictors which allow the failure estimation for each layer in the composite laminate opposite to the previous studies A. S¸ erban (&) Faculty of Mechanical Engineering, University of Gheorghe Asachi, 700050 Iasi, Romania e-mail: [email protected]
which model the failure response only for the whole structure. Keywords Layup optimization Failure estimation Multilayer perceptron Convolutional neural networks Recurrent neural networks
Introduction The failure mechanism of layered composite materials is highly influenced by the fiber orientation due to the orthotropic behavior exhibited by the reinforced layers. From the design point of view, it is desirable to find a stacking sequence of the constituent layers such that the composite laminate has a minimal weight and/or cost while it is also able to carry the loads for which it is designed. The problem of finding such optimal stacking sequences (configurations) is very complex and has a multidimensional solution space explored with heuristic computational techniques [1–8] from which the most reliable are the genetic algorithms. In order to converge to an optimal solution during the optimization process, a lot of candidate configurations are explore
Data Loading...
