Thermal Image Error Analysis for Precision Correction in Industrial Devices
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods fail to accurately identify heat-generating areas within industrial devices that significantly affect precision, leading to inadequate correction expressions and reduced precision due to thermal displacement and deformation.
Innovation Solution
An error analysis method using thermal imaging and machine learning to train a model that estimates correction amounts and identifies heat-generating areas affecting precision, employing techniques like Grad-CAM, backpropagation, or deconvolution networks to specify contribution levels.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If temperature sensors are used to measure temperatures of mechanical elements and machine learning is applied to derive correction expressions, then correction precision can be improved at low computational cost, but it remains impossible to identify which specific parts generating heat or deforming significantly contribute to precision errors
Solution Approach 1:
The patent segments the thermal analysis by dividing the industrial device into multiple regions of interest and using thermal imaging to capture temperature distributions across different areas. This segmentation enables identification of specific heat-generating portions by analyzing temperature variations in divided regions, thereby resolving the contradiction between achieving correction precision and identifying specific heat-generating areas.
Solution Approach 2:
The patent introduces thermal imaging technology as an intermediary between temperature measurement and precision error identification. The thermal image serves as a mediator that visualizes temperature distributions and highlights heat-generating areas, enabling the system to identify which specific portions contribute to precision errors while maintaining correction precision through machine learning-based analysis.
2Measurement precision
If thermal imaging and machine learning with contribution level specification are used to identify heat-generating areas, then identification accuracy of precision-affecting portions is improved, but device complexity increases
Solution Approach 1:
The patent applies multi-functionality by using a single machine learning model that performs both correction expression derivation and heat-generating area identification. The model processes thermal image data to simultaneously achieve precision correction and locate problematic areas, reducing the need for separate dedicated systems and thereby managing device complexity while maintaining high identification accuracy.
Solution Approach 2:
The system performs self-service by automatically identifying heat-generating areas and generating correction expressions without requiring manual intervention or additional specialized equipment. The machine learning model autonomously analyzes thermal images, specifies contribution levels of different regions, and determines which portions affect precision, thereby improving identification accuracy while avoiding the complexity of manual analysis systems.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Accurately identifies heat-generating areas that impact precision, enabling precise correction calculations and improved device performance.
Implementation Method 1
obtaining a thermal image taken during an operation of an industrial device
Data Source
AI summary
An error analysis method according to the present disclosure includes: (S1) obtaining a thermal image taken and an error occurring during an operation of an industrial device; and (S2) training a model by using the thermal image and the error in machine learning to estimate an amount of correction for the industrial device from the thermal image, and determining, using a level of contribution specified by a predetermined method, a portion that affects precision out of the industrial device appearing in the thermal image. The obtaining includes obtaining a temperature of the portion determined in the determining to calculate the amount of correction for the industrial device.