Magnetic core loss prediction method based on data driving and computer program product

By combining the modified empirical compensation model and machine learning prediction model, using neural networks and genetic algorithms to optimize weight parameters, the problem of low core loss prediction accuracy in the existing technology is solved, and higher accuracy and widely applicable core loss prediction are achieved.

CN119646482AActive Publication Date: 2025-03-18XIAN INST OF OPTICS & PRECISION MECHANICS CHINESE ACAD OF SCI
4 Cites 1 Cited by

Patent Information

Application Number
CN202411705668.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-11-26
Publication Date
2025-03-18
Estimated Expiration
2044-11-26

AI Technical Summary

Technical Problem

It is difficult for the prior art to accurately predict core loss, especially under high frequency conditions. The fitting coefficient of the traditional Steinmetz empirical equation is greatly affected by temperature and material factors, resulting in low prediction accuracy.

Method used

Using a data-driven method, combined with the modified empirical compensation model and the machine learning prediction model, the compensation terms of the two prediction results are obtained through the neural network, and the weight parameters are optimized through the genetic algorithm to finally calculate the predicted value of the core loss.

Benefits of technology

This greatly reduces the mean square error between the prediction results and the real loss, and improves the accuracy and wide applicability of core loss prediction.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a magnetic core loss prediction method based on data driving and a computer program product. The method comprises the following steps: S1, constructing an experience compensation model and a machine learning prediction model of magnetic core loss; s2, correcting the experience compensation model to obtain a prediction result, and correcting the machine learning prediction model to obtain a prediction result; s3, training the two prediction results through the constructed neural network to obtain compensation values of the prediction results; s4, weight parameters are adapted to the prediction result and the compensation value; and S5, calculating a predicted value of the magnetic core loss through the prediction result and the compensation value in combination with a weight parameter, and obtaining an optimal parameter of the magnetic core loss through genetic algorithm iteration. According to the magnetic core loss prediction method based on data driving and the computer program product, the mean square error between the prediction result and the real loss is greatly reduced.
Need to check novelty before this filing date? Find Prior Art