AI Gearwheel Surface Profiling for Transmission Vibration Sorting
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for identifying gearwheels that induce vibrations in transmissions are inefficient, leading to prolonged manufacturing times and increased fault analysis times.
Innovation Solution
A computer-implemented method using a generic AI model trained with a data set of reference gearwheels to classify gearwheels into categories based on their surface profiles, determining whether they induce vibrations above or below a threshold, utilizing machine learning techniques such as neural networks and random forests.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If traditional vibration measurement and evaluation methods are used to identify faulty gearwheels, then measurement precision can be maintained, but manufacturing time and fault analysis time increase significantly
Solution Approach 1:
The patent applies preliminary action by training an AI model in advance with a comprehensive data set of vibration signals from multiple gearwheel states. This pre-trained model can then quickly classify new gearwheels without requiring time-consuming traditional vibration analysis, thus reducing manufacturing time while maintaining detection accuracy through the model's learned patterns
Solution Approach 2:
The patent replaces the traditional mechanical vibration measurement and evaluation system with an AI-based classification system. The AI model processes vibration signals and categorizes gearwheels as faulty or non-faulty, substituting complex mechanical analysis with intelligent algorithms that operate faster and require less manual intervention
2Productivity
If traditional fault analysis methods are applied to all transmissions, then reliability can be ensured, but productivity decreases due to increased fault analysis time
Solution Approach 1:
The patent applies local quality by focusing fault detection efforts only where needed - specifically on gearwheels that exhibit vibration patterns consistent with faults. The AI model identifies and flags only those specific gearwheels requiring further attention, rather than performing uniform detailed analysis on all transmissions, thus improving productivity while maintaining reliability for critical components
Solution Approach 2:
The patent changes the parameter of fault detection from continuous traditional vibration analysis to discrete AI-based categorization. By transforming the detection parameter into categorical outputs (faulty/non-faulty classifications), the system achieves faster processing that improves productivity while maintaining sufficient reliability through the AI model's trained classification capabilities
Data Source
AI summary
A computer-implemented method for identifying a gearwheel (5) that induces vibrations in a transmission includes training a generic AI model with a training data set of a group of reference gearwheels (19). At least one reference profile of a surface (21) of the reference gearwheel (19) and a gearwheel category are provided for each reference gearwheel (19) of the group. The method also includes determining a profile of a surface (11) of a gearwheel (5) and assigning one of the gearwheel categories to the gearwheel (5) on the basis of the determined profile by the trained AI model.