Average Acceleration Optimization for CAD Design Noise Reduction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing computer-aided design (CAD) and computer-aided engineering (CAE) systems face challenges in computationally expensive optimization methods for designing real-world objects, which often result in noisy acceleration and sensitivity data due to high-localized oscillations, leading to suboptimal designs and increased computational costs.
Innovation Solution
The implementation of a method that calculates continuous average accelerations using velocities at the start and end points of time windows, reducing noise and oscillations by directly applying transient responses and sensitivities in optimization processes, without linear static approximations, and using these average accelerations to iteratively optimize CAD models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing optimization methods are used to design real-world objects, then design optimization can be achieved, but computational cost increases significantly and noise and oscillations occur in acceleration and sensitivity data
Solution Approach 1:
The patent changes the parameter representation from discrete acceleration values to continuous average acceleration over time windows. This parameter transformation smooths the data, eliminating high-frequency noise and oscillations while preserving the essential optimization information, thereby reducing computational requirements for sensitivity calculations
Solution Approach 2:
The patent segments the time domain into multiple time windows and calculates average acceleration for each window. This segmentation approach allows the optimization to consider acceleration behavior over meaningful time intervals rather than at discrete points, reducing the impact of localized oscillations and improving computational efficiency
2Manufacturing precision
If existing optimization methods are used, then design improvements can be identified, but noisy acceleration data leads to suboptimal designs
Solution Approach 1:
The patent introduces average acceleration as an intermediary parameter between the raw acceleration data and the optimization objective. This intermediary smooths the noisy acceleration measurements by averaging over time windows, providing a more reliable input for sensitivity calculations while maintaining the connection to the original acceleration data
Solution Approach 2:
The patent transforms the acceleration parameter from point-in-time values to time-averaged values over windows. This parameter change filters out high-frequency noise and oscillations that would otherwise corrupt the optimization, yielding more accurate and reliable design results
3Measurement precision
If high-frequency acceleration data is used in optimization, then detailed acceleration behavior is captured, but computational complexity and noise increase
Solution Approach 1:
The patent segments the acceleration data into time windows and computes averages for each segment. This segmentation reduces the complexity of the optimization process by transforming high-frequency detailed data into a smaller set of representative average values, maintaining essential acceleration behavior information while reducing computational burden
Solution Approach 2:
The patent extracts the essential acceleration information by calculating average acceleration over time windows, filtering out the high-frequency noise and oscillations. This extraction approach retains the meaningful acceleration behavior patterns while eliminating the computational complexity associated with processing every high-frequency data point
Data Source
AI summary
Embodiments automatically determine optimized designs of real-world objects. Using a computer-based model representing a real-world object, an embodiment determines equilibriums of the real-world object across a plurality of time steps. Determining said equilibriums determines velocities of the real-world object across the plurality of time steps. Average acceleration of the real-world object is determined for each of a plurality of time windows (defined across the plurality of time steps) using the determined velocities. Sensitivity of each determined average acceleration is calculated. The determined average accelerations are used to define at least one of a constraint and an objective function. The computer-based model representing the real-world object is iteratively optimized, using the calculated sensitivity of each determined acceleration, with respect to at least one of the constraint and the objective function. The iterative optimization results in an updated computer-based model, representing the optimized design of the real-world object.