Virtual Balance Prediction for Unfinished Rotating Workpieces
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Identifying and remedying rotational imbalances in vehicle components like crankshafts is time-consuming and resource-intensive, requiring operators to determine the imbalance source and corrective actions.
Innovation Solution
A method and system for predicting rotational balance using image data to generate a virtual representation, segment into target meshes, determine parameters, and apply a balance prediction model to identify and correct imbalances.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional physical manufacturing and testing methods are used to identify rotational imbalances, then measurement accuracy is improved, but time consumption and resource usage increase
Solution Approach 1:
The patent creates a virtual copy (digital twin) of the physical workpiece by generating a point cloud from images and converting it into a mesh model. This virtual representation allows for rotational balance analysis without physically manufacturing or testing the actual component, thereby maintaining measurement accuracy while dramatically reducing time and resource consumption.
Solution Approach 2:
The patent replaces physical mechanical measurement systems with computational image processing and virtual modeling. Instead of using physical balancers and test equipment, the system uses image data processing, point cloud generation, and virtual mesh analysis to determine rotational balance characteristics, eliminating the need for physical manufacturing and testing.
2Reliability
If physical manufacturing and testing processes are used, then measurement reliability is improved, but resource consumption and cost increase
Solution Approach 1:
The system creates a faithful virtual replica of the physical workpiece that preserves all geometric and mass distribution characteristics needed for rotational balance analysis. This digital twin enables reliable measurement without consuming physical resources for manufacturing test pieces or operating physical testing equipment.
Solution Approach 2:
The system uses the workpiece's own images (taken during normal manufacturing or inspection processes) to generate the virtual model, eliminating the need for separate physical sampling or test piece creation. The existing visual data is repurposed for balance analysis, reducing additional resource consumption.
3Manufacturing precision
If detailed physical inspection and correction processes are used, then manufacturing precision is improved, but process complexity and time increase
Solution Approach 1:
The patent replaces complex physical inspection and correction procedures with automated image processing and virtual modeling. The system automatically generates point clouds from images, converts them to mesh models, calculates mass distribution, and identifies balance issues without requiring manual measurement or physical adjustment processes.
Solution Approach 2:
The system performs rotational balance analysis on the virtual model before physical manufacturing or assembly is completed. By identifying balance issues in the virtual stage, corrections can be planned and executed more efficiently, reducing the need for iterative physical adjustments and complex on-site correction processes.
Data Source
AI summary
A method for predicting an incoming rotational balance of an unfinished workpiece includes obtaining image data of the unfinished workpiece, generating a virtual representation of the unfinished workpiece based on the image data, segmenting the virtual representation into one or more target meshes, determining one or more parameters of the one or more target meshes, where the one or more parameters comprise a volume, a center of gravity, or a combination thereof, and determining a balance metric of the unfinished workpiece based on the one or more parameters and a balance prediction model.


