Neural Network Strength Prediction for Ceramic Honeycomb Substrates
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for inspecting honeycomb cellular structures are time-consuming and destructive, failing to accurately predict the strength of these structures under high pressure or force without causing damage, and do not account for minute structural differences.
Innovation Solution
A method utilizing a neural network that inspects cellular structures using an inspecting device to capture and analyze cell parameters, which are then used to predict the structural strength by learning from a database of sample structures and applying force measurements to determine the relationship between cell parameters and strength.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional destructive testing methods are used to determine pressure withstand capability, then strength measurement accuracy is improved, but the structure is destroyed and cannot be reused
Solution Approach 1:
The patent uses optical copying techniques to create digital representations of the honeycomb cellular structure. An inspecting device captures images of the structure's geometry, cell dimensions, and wall thicknesses, then processes these images into numerical data that serves as a virtual model. This copying approach allows strength prediction without physical destruction, resolving the contradiction between accurate measurement and part preservation.
Solution Approach 2:
The patent replaces the mechanical destructive testing system with a neural network-based prediction system. Instead of physically applying destructive forces to measure strength, the system uses optical inspection to gather geometric data, then employs a neural network trained on strength parameters to predict the structure's pressure withstand capability. This substitution eliminates the need for destructive mechanical testing while providing strength assessment.
2Measurement precision
If manual inspection methods are used to examine cellular structures, then inspection thoroughness is improved, but inspection time increases significantly
Solution Approach 1:
The patent replaces manual inspection methods with an automated optical inspection system. The inspecting device captures images of the cellular structure, and a processing system automatically analyzes the images to extract geometric parameters such as cell dimensions, wall thicknesses, and structural features. This automation maintains thoroughness by systematically examining all visible features while dramatically reducing inspection time compared to manual methods.
Solution Approach 2:
The patent performs preliminary optical capture and image processing to extract all relevant geometric information before strength prediction is needed. The system pre-processes the structural data, creating a comprehensive digital representation that can be quickly analyzed by the neural network. This preliminary action ensures thorough inspection is completed upfront, enabling rapid subsequent strength assessment without repeated detailed examinations.
3Measurement precision
If conventional inspection methods are used to detect structural differences, then defect detection capability is improved, but the ability to predict strength of structures with minute differences deteriorates
Solution Approach 1:
The patent transforms the inspection approach by changing from qualitative defect detection to quantitative parameter measurement. The system measures specific geometric parameters such as cell dimensions, wall thicknesses, and structural geometries with high precision. These numerical parameters are then input to the neural network, which has been trained to correlate these parameters with strength outcomes. This parameter transformation enables the system to reliably predict strength differences even for minute structural variations that conventional defect detection might miss.
Solution Approach 2:
The patent replaces conventional defect detection methods with a neural network-based prediction system. Instead of relying on inspectors to identify and categorize defects, the system uses optical measurement to capture precise geometric parameters and employs a trained neural network to predict strength based on these parameters. This substitution improves reliability for predicting strength of structures with minute differences, as the neural network can detect subtle parameter variations and their impact on strength that human inspectors might overlook.
Data Source
AI summary
A method of examining a cellular structure includes the steps of providing an inspecting device, a neural network and a target cellular structure that includes a plurality of target cells extending therethrough and further includes a target face exposing an arrangement of the target cells; inspecting the arrangement of cells on the face of the target cellular structure using the inspecting device; representing the arrangement of cells with numerically defined target cell parameters; inputting the target cell parameters into the neural network; and generating an output from the neural network based on the target cell parameters, the output being indicative of a strength of the target cellular structure.