Motor Evaluation Estimation Using Gap Flux Feature Extraction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The high calculation load during learning of estimation models for rotary electric machines, particularly due to the large dimension of explanatory variables, poses a challenge in existing methods like CNN-based approaches.
Innovation Solution
An estimation model is created by learning the relationship between magnetic flux density distribution in the gap between the rotor and stator, using a simplified stator model to reduce the calculation load and improve efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a CNN-based method using grayscale images representing magnetic flux density distribution is employed to estimate rotary electric machine characteristics, then the estimation capability is improved, but the calculation load during learning increases due to the enormous dimension of explanatory variables
Solution Approach 1:
The patent extracts only the essential explanatory variables needed for accurate estimation from the complete magnetic flux density distribution. Instead of using all pixel values from the grayscale image, the invention identifies and selects specific extraction positions where magnetic flux density values are most relevant for estimating torque characteristics, thereby reducing the dimension of explanatory variables while preserving estimation accuracy.
Solution Approach 2:
The patent segments the magnetic flux density distribution into discrete extraction positions around the rotor perimeter. By dividing the continuous distribution into specific sampling points, the invention transforms the high-dimensional image data into a manageable set of discrete features that capture the essential information needed for torque estimation.
2Power
If the dimension of explanatory variables is reduced by using only magnetic flux density distribution in the gap, then the calculation load is reduced, but the estimation accuracy may be compromised
Solution Approach 1:
The patent applies local quality by strategically selecting extraction positions where magnetic flux density values have the greatest impact on torque characteristics. Rather than uniformly sampling the entire rotor perimeter, the invention places extraction points at locations with higher informational value for torque estimation, such as regions where magnetic flux density gradients are most significant.
Solution Approach 2:
The patent changes the parameter representation from complete spatial distribution to a reduced set of discrete magnetic flux density values at specific positions. This parameter transformation maintains the essential physical relationships while reducing computational complexity, enabling accurate torque estimation with fewer input variables.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An estimation model creation device (110) acquires a magnetic flux density distribution in a gap (G) between a rotor (210) and a stator (320). The estimation model creation device (110) calculates an estimated value of a motor evaluation index corresponding to the magnetic flux density distribution by using an estimation model created by learning a relationship between the magnetic flux density distribution in the gap (G) and the motor evaluation index.