Aircraft Composite Laminate Stacking for Fast Buckling Optimization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for manufacturing aeronautical composite structures face high computational costs and inefficiencies in optimizing stacking sequences due to the mismatch between continuous and discrete parameters, leading to increased time and material waste.
Innovation Solution
A method utilizing artificial neural networks (ANNs) to predict buckling loads and optimize laminate stacking sequences, combined with a greedy algorithm, to generate a database of optimal laminates that meet specified constraints, thereby reducing computational burden and material usage.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If genetic algorithms are used for direct optimization of stacking orientations, then the design can directly optimize discrete stacking sequences, but the computational time required increases significantly
Solution Approach 1:
The patent pre-calculates and stores buckling loads for various laminate configurations in a lookup table before the actual optimization process. This preliminary action allows the genetic algorithm to quickly retrieve pre-computed values instead of performing time-consuming finite element analyses during optimization, thereby maintaining discrete stacking sequence optimization while dramatically reducing computational time.
Solution Approach 2:
The patent creates a simplified computational model by storing buckling load data in a lookup table that replicates the results of complex finite element analyses. This copied data structure allows the optimization algorithm to work with approximate but sufficient information, avoiding the need to run full FEM simulations during the optimization process and thus reducing computational burden.
2Measurement precision
If finite element models are used for structural analysis during optimization, then accurate buckling loads can be computed, but the computational cost increases
Solution Approach 1:
The patent performs finite element analyses in advance to compute buckling loads for a comprehensive set of laminate configurations, storing these results in a lookup table. This preliminary computation phase separates the expensive FEM analyses from the iterative optimization process, allowing accurate buckling load data to be available without incurring repeated computational costs during optimization iterations.
Solution Approach 2:
The patent replaces expensive real-time finite element analyses with a pre-computed lookup table that contains copied buckling load data. This approach maintains measurement precision by using FEM-derived data while eliminating the need to run FEM simulations during each optimization iteration, thus significantly reducing computational energy consumption.
3Ease of manufacture
If continuous parameters are used for optimization, then the optimization process is simpler, but the target stacking sequence becomes discrete requiring additional computational steps
Solution Approach 1:
The patent pre-generates a lookup table containing buckling loads for discrete stacking sequences before optimization begins. This preliminary preparation allows the genetic algorithm to work directly with discrete stacking parameters without needing to map from continuous to discrete space, simplifying the optimization process while maintaining discrete output requirements.
Data Source
AI summary
A method intended for manufacturing a composite structure for an aircraft. The method includes training a set of artificial neural networks (3.1, 3.2, 3.3) to obtain a composite structure which fulfills certain specification requirements (material properties, shear loads, compression loads, buckling constraints). The neural networks are trained with data obtained from analytical equations and from a finite element model.