Component Deformation Modeling via Statistical Distribution Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for calculating component deformation, such as in combustion systems, are complex, time-consuming, and inaccurate due to part-to-part variation, making them unsuitable for precise deformation modeling.
Innovation Solution
A system and computer program product that model deformation in manufactured components by forming pre-exposure and post-exposure statistical distributions of measured coordinates, calculating differences between these distributions, and adjusting an expected deformation model to minimize errors from part-to-part variation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional methods take measurements of reference points and manufacture multiple components to develop statistically significant comparisons, then measurement precision may be improved, but device complexity and time consumption increase significantly
Solution Approach 1:
The patent extracts and removes part-to-part variation from the deformation measurement process by using statistical distribution analysis. Instead of measuring multiple individual components to achieve statistical significance, the method extracts the common deformation pattern from a single component by separating it from manufacturing variations through coordinate mapping and statistical distribution comparison between as-manufactured and as-delivered states
Solution Approach 2:
The patent creates a statistical distribution model (a virtual copy) of the component's coordinate variations from the as-manufactured state, then compares this model against the as-delivered state. This copying approach eliminates the need to physically manufacture and measure multiple components, achieving statistical significance through computational modeling rather than physical replication
2Reliability
If conventional methods manufacture multiple components for statistically significant comparisons, then reliability of deformation data improves, but loss of time and productivity decrease
Solution Approach 1:
The patent performs preliminary statistical distribution analysis on the as-manufactured component coordinates before exposure to deformation conditions. By establishing the baseline statistical distribution in advance, the method enables direct comparison with the as-delivered state, eliminating the need to wait for multiple components to be manufactured and measured over time
Solution Approach 2:
The patent transforms the physical requirement of manufacturing multiple components into a computational parameter change - comparing statistical distributions of coordinates. This parameter transformation allows deformation analysis to be performed on a single component's coordinate data, converting a time-consuming physical process into a rapid computational comparison
3Manufacturing precision
If conventional approaches use multiple components for deformation analysis, then accuracy of deformation modeling improves, but device complexity and cost increase
Solution Approach 1:
The patent introduces statistical distribution analysis as an intermediary between the physical component and the deformation measurement. Instead of directly comparing multiple physical components, the method uses statistical distributions of coordinates as an intermediary representation, which simplifies the measurement process while maintaining or improving deformation modeling accuracy
Solution Approach 2:
The patent replaces the mechanical approach of manufacturing and measuring multiple physical components with a computational system that analyzes coordinate data. This substitution uses information processing and statistical analysis to replace the physical replication and measurement process, reducing manufacturing complexity and cost while maintaining measurement accuracy
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Various embodiments include a system (100) having: a computing device (126) configured to model deformation (184) in a set of manufactured components (170) by: forming a pre-exposure statistical distribution (175) of measured coordinates describing the set of manufactured components (170) from a pre-exposure three-dimensional (3D) depiction of a first sample of the manufactured component (170), and forming a post-exposure statistical distribution of measured coordinates describing the set of manufactured components (170) from a post-exposure 3D depiction of a second sample of the manufactured component (170); calculating a difference between parameters of the pre-exposure statistical distribution (175) and parameters of the post-exposure statistical distribution (175); and adjusting an expected deformation model (180) for the set of manufactured components (170) based upon the difference between parameters of the pre-exposure statistical distribution (175) and the post-exposure statistical distribution (175) to model the deformation of the manufactured component (170).