Gradient Vector Optimization for Composite Sandwich Structures
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The complex design of fuselages, composed of numerous composite sandwich structures, poses a computationally intractable combinatorial problem when selecting optimal structures for thousands of analysis locations, making existing optimization techniques impractical for large structural elements.
Innovation Solution
A method involving the calculation of a gradient vector to identify optimal composite sandwich structures by weighting components of vectors representing available structures at different locations, allowing for the selection of structures that satisfy design criteria through gradient descent techniques and providing output via a user interface.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing optimization techniques are applied to choose composite sandwich structures for thousands of analysis locations, then design objectives can be met, but computational complexity becomes intractable
Solution Approach 1:
The fuselage is divided into multiple discrete analysis locations (thousands of locations), and the optimization problem is segmented into independent sub-problems at each location. This allows the global optimization problem to be broken down into manageable local decisions, reducing overall computational complexity while still achieving global design objectives.
Solution Approach 2:
The patent transforms the discrete optimization problem into a continuous optimization problem by parameterizing the composite sandwich structure selection. Instead of evaluating discrete structure types at each location, the method uses continuous parameters (such as thickness, ply orientation, material composition) that can be optimized using gradient-based methods, making the problem computationally tractable.
2Use of energy by moving object
If a lightweight fuselage is designed to reduce fuel consumption, then energy efficiency improves, but structural safety constraints may be compromised
Solution Approach 1:
The patent applies local quality optimization by allowing different composite sandwich structures with different properties (thickness, material composition, ply orientation) at different analysis locations along the fuselage. This enables the structure to be lightweight where possible while maintaining necessary strength and stiffness where structural demands require it, thus reducing overall weight and fuel consumption without compromising safety.
Solution Approach 2:
The optimization method uses gradient vectors that provide feedback on how changes in composite sandwich structure parameters affect both weight and structural performance. This feedback mechanism allows the optimization algorithm to iteratively adjust the structure design, reducing weight while ensuring that structural safety constraints are satisfied at each iteration.
Data Source
AI summary
A method includes calculating a gradient vector of a property of a framework with respect to first components of a first vector and second components of a second vector. The first components correspond to first structures available for a first location of the framework and the second components correspond to second structures available for a second location of the framework. The method includes identifying, based on the gradient vector, first values for the first components and second values for the second components that yield a third value of the property that satisfies a criterion, selecting a first selected composite sandwich structure of the first composite sandwich structures and a second selected composite sandwich structure of the second composite sandwich structures, and providing output indicating (a) the first values and the second values or (b) the first selected composite sandwich structure and the second selected composite sandwich structure.