Wavelet Analysis for Surface Feature Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Milling operations often result in surfaces with undesired roughness and waviness due to tool movement patterns, requiring extensive manual rework to achieve desired smoothness, increasing time and expense in manufacturing.
Innovation Solution
A method and apparatus using wavelet analysis to select a range of frequencies based on stepover distances between adjacent peaks, processing surface data to isolate desired features like roughness and waviness, and determining if the surface meets policy specifications, with reworking options such as re-machining, sanding, or burnishing to correct deviations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If manual rework is performed to reduce surface roughness and waviness, then surface quality is improved, but manufacturing time and cost increase
Solution Approach 1:
The patent replaces manual mechanical rework with automated inspection systems that use wavelet analysis to detect surface features. The system automatically measures surface roughness and waviness, identifies problematic areas, and can trigger automated rework processes, eliminating the need for manual inspection and estimation while reducing overall manufacturing time.
Solution Approach 2:
The patent implements a feedback mechanism where surface measurements are continuously monitored during machining, and the results are fed back to control the machining process. Wavelet analysis provides real-time feedback about surface quality, allowing for immediate adjustment of machining parameters to prevent defects rather than requiring post-production rework.
2Measurement precision
If wavelet analysis is used to isolate surface features, then measurement precision is improved, but device complexity increases
Solution Approach 1:
The patent applies parameter changes by transforming surface profile data into the frequency domain using wavelet transforms. This mathematical transformation separates different surface features (roughness, waviness, form) into distinct frequency components, allowing precise identification and measurement of each feature type without requiring complex physical measurement systems.
Solution Approach 2:
The patent uses wavelet analysis as an intermediary mathematical tool that bridges the gap between raw surface measurement data and actionable quality information. This intermediary process converts complex surface profiles into interpretable frequency components, simplifying the inspection system's computational requirements while maintaining high measurement precision.
3Productivity
If automated inspection is implemented to reduce manual rework, then productivity is improved, but device complexity increases
Solution Approach 1:
The patent creates a universal inspection system that can detect multiple surface features (roughness, waviness, form errors) using a single integrated wavelet analysis algorithm. This multi-functional approach consolidates what would otherwise require multiple separate measurement devices and manual inspection procedures into one automated system, improving productivity without proportionally increasing complexity.
Solution Approach 2:
The patent performs preliminary surface feature identification during the machining process itself rather than requiring separate post-production inspection steps. By using wavelet analysis to detect surface features in real-time during machining, the system enables preventive rework or process adjustment before defects are created, streamlining the overall manufacturing workflow and improving productivity.
Data Source
Figure 1~2
Figure 3
Figure 4
AI summary
A method and apparatus for inspecting a surface of an object. Data from measuring the surface of the object is obtained to form surface data for the object. A range of frequencies for features on the object is selected based on a range of distances between adjacent peaks for the features. The features are formed by a tool moving along a number of paths. Desired surface data for the features is obtained from the surface data using the range of frequencies selected. A determination is made as to whether the desired surface data for the features meets a policy specifying a desired surface for the object. In response to an absence of a determination that the desired surface data for the features meets the policy, the object is reworked.