Abstract:In this paper, an efficient method for performing minimum weight optimization of composite laminates using artificial neural network (ANN) based surrogate models is proposed. By predicting the buckling loads of the laminates using ANN the use of time-consuming buckling evaluations during the iterative optimization process are avoided. Using for the first time lamination parameters, laminate thickness and other dimensional parameters as inputs for these ANN models significantly reduces the number of models required and therefore computational cost of considering laminates with many different numbers of layers and total thickness. Besides, as the stacking sequences are represented by lamination parameters, the number of inputs of the ANN models is also significantly reduced, avoiding the curse of dimensionality. Finite element analysis (FEA) is employed together with the Latin hypercube sampling (LHS) method to generate the database for the training and testing of the ANN models. The trained ANN models are then employed within a genetic algorithm (GA) to optimize the stacking sequences and structural dimensions to minimize the weight of the composite laminates. The advantages of using ANN in predicting buckling load is proved by comparison with other machine learning methods, and the effectiveness and efficiency of the proposed optimization method is demonstrated through the optimization of flat, blade-stiffened and hat-stiffened laminates.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is the minimum - weight optimal design of composite laminated structures. Specifically, the author proposes an effective method based on artificial neural networks (ANN) and genetic algorithms (GA) to predict the buckling load of composite laminated structures and minimize the structural weight by optimizing lamination parameters, lamination thickness and other dimensional parameters. This method aims to avoid using time - consuming buckling evaluations in the iterative optimization process, thereby significantly reducing the computational cost, especially when dealing with laminates with different numbers of layers and total thicknesses. ### Main problems 1. **Minimum - weight optimization**: How can the minimum - weight design of composite laminated structures be achieved by optimizing lamination parameters, lamination thickness and other structural dimensions under the condition of satisfying buckling constraints? 2. **Computational efficiency**: How can the ANN model be used to predict the buckling load of laminated structures to reduce the computational cost in the optimization process? 3. **Multi - layer number optimization**: How can laminates with different numbers of layers and total thicknesses be processed in the optimization process to avoid the curse of dimensionality and training difficulties? ### Solutions - **ANN model**: Use the finite element analysis (FEA) and Latin hypercube sampling (LHS) methods to generate a database and train the ANN model to predict the buckling load of laminated structures. - **Genetic algorithm**: Combine the trained ANN model and use GA to optimize lamination parameters, lamination thickness and other structural dimensions to minimize the structural weight. - **Lamination parameters**: Introduce lamination parameters as inputs to the ANN model to reduce the number of inputs, avoid the curse of dimensionality and improve the generalization ability of the model. ### Method steps 1. **Data generation**: Use ABAQUS for buckling analysis to generate a data set containing different lamination parameters, lamination thickness and other dimensional parameters. 2. **ANN model development**: Use Keras to develop the ANN model and optimize the model parameters through the training data set. 3. **GA optimization**: Combine the trained ANN model and use GA for optimization to minimize the structural weight and satisfy the buckling constraints. ### Advantages - **High computational efficiency**: By predicting the buckling load through the ANN model, the computational time in the optimization process is significantly reduced. - **Strong generalization ability**: Using lamination parameters as inputs enables the model to handle laminates with different numbers of layers and total thicknesses. - **Good optimization effect**: Through GA optimization, the minimum - weight design of the structure can be effectively achieved under the condition of satisfying the buckling constraints. ### Application examples The paper verifies the effectiveness and efficiency of the proposed method by optimizing the laminated structures of planes, blade reinforcements and hat - shaped reinforcements. These applications show the potential value of this method in practical engineering.

Design optimization of laminated composite structures using artificial neural network and genetic algorithm

Multiple-scale Uncertainty Optimization Design of Hybrid Composite Structures Based on Neural Network and Genetic Algorithm

Enhancing composite laminate structures with tailored neural networks

An Efficient Optimization Method for Stacking Sequence of Composite Pressure Vessels Based on Artificial Neural Network and Genetic Algorithm

Aerodynamic Optimization Design of Airfoil Configurations Based on Cascade Feedforward Neural Network

Optimization Design for Laminated Composite Structure Based on Kriging Model

An even-load-distribution design for composite bolted joints using a novel circuit model and artificial neural networks

Integrated optimization of ply number, layer thickness, and fiber angle for variable-stiffness composites using dynamic multi-fidelity surrogate model

Optimal Design of Composite Grid/skin Structures Based on Deep Learning and Double-Double Layup Strategy

Buckling Performance Evaluation of Double-Double Laminates with Cutouts Using Artificial Neural Network and Genetic Algorithm

An Even-Load-distribution Design for Composite Bolted Joints Using a Novel Circuit Model and Neural Network

Enhanced Optimization of Composite Laminates: Multi-Objective Genetic Algorithms with Improved Ply-Stacking Sequences

Simultaneous optimization of fiber paths and geometric dimensions for fiber composite laminates using double neural network-surrogate model and genetic algorithm

A Deep Learning-Based Composite Design Strategy for Efficient Selection of Material and Layup Sequences from a Given Database

Strength prediction and progressive damage analysis of carbon fiber reinforced polymer-laminate with circular holes by an efficient Artificial Neural Network

Neural networks modeling of strain energy density and Tsai-Wu index in laminated composites

Composite Lamina Model Design with the Use of Heuristic Optimization

Machine learning based topology optimization of fiber orientation for variable stiffness composite structures

A Optimization Technique for the Composite Strut Using Genetic Algorithms

Prediction and Optimization of Process Parameters for Composite Thermoforming Using a Machine Learning Approach

Optimal design of laminates orientation in a composite material by genetics algorithm and simulated annealing