Burnishing State Prediction for Thin-Walled Hollow Workpieces
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In burnishing processes, especially for hollow workpieces with thin wall thickness, it is challenging to adjust the amount of rolling compaction to match the workpiece material hardness and wall thickness, leading to waviness or deformation and requiring frequent readjustment of machining conditions.
Innovation Solution
A machine learning device is developed to generate a learned model that predicts the machined state of a workpiece after burnishing by inputting machining information, including workpiece details and machining conditions, allowing for accurate prediction without actual burnishing or simulation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If burnishing is performed on hollow workpieces with thin wall thickness, then surface roughness is improved, but waviness or deformation occurs due to excessive rolling compaction
Solution Approach 1:
The system performs preliminary prediction of the machined state using a learned model before actual burnishing. By inputting workpiece information (material, hardness, wall thickness) and machining conditions (rolling compaction amount, rotation speed, feedrate), the model predicts surface roughness and dimensional accuracy, allowing operators to adjust conditions in advance to avoid waviness or deformation
Solution Approach 2:
The system uses feedback from the learned model's predictions to adjust machining conditions. The model learns from training data the relationship between machining parameters and outcomes, then provides feedback on optimal settings to achieve desired surface roughness without causing deformation in thin-walled hollow workpieces
2Manufacturing precision
If machining conditions are adjusted to match workpiece material and wall thickness, then manufacturing precision is improved, but time and effort are required for readjustment and reperformance
Solution Approach 1:
The learned model performs preliminary prediction of machining outcomes before actual burnishing operations. This allows the system to identify optimal machining conditions in advance based on workpiece characteristics, eliminating the need for time-consuming trial-and-error adjustments and reperformance of burnishing operations
Solution Approach 2:
The system uses the learned model to automatically determine optimal machining conditions without requiring expert intervention. The model self-services by taking workpiece information and desired outcomes as input and automatically providing recommended machining parameters, reducing the time and effort previously required for manual readjustment
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
The learned model effectively outputs the machined state and surface roughness after burnishing, enabling accurate prediction and reducing the time and effort required for readjusting machining conditions and reperforming burnishing.
Implementation Method 1
contact pressure exceeds the yield point of the workpiece W, and plastic deformation occurs
Implementation Method 2
the tool T separates from the burnished surface, and the metal of the workpiece W elastically recovers only a minute distance Dr
Implementation Method 3
burnishing a method of plastic working for smooth finishing a surface by pressing a tool T having a roller (or spherical) shape onto a surface to be burnished of a metal workpiece W while causing rolling compaction
Implementation Method 4
the vicinity of a burnished portion of the workpiece W hardens and compressive residual stress is applied
Data Source
AI summary
A machine learning device is provided with: an input data obtaining unit that, in burnishing process in which a processing surface of an arbitrary workpiece is surface-treated with an arbitrary tool, obtains as input data processing information including information of the workpiece prior to the burnishing process and information of a processing condition; a label obtaining unit that obtains label data indicating processed state information including a processed state of the workpiece after the burnishing process and surface roughness of the workpiece when the processed state is normal; and a learning unit that carries out supervised learning using the input data and the label data thus obtained to generate a learned model to which processing information of an upcoming burnishing process is input and which outputs processed state information for the burnishing process.