一种基于物理信息神经网络的永磁同步电机多参数辨识方法及系统
By constructing a physical information neural network model and combining a composite loss function of data fitting and physical constraints, the problems of insufficient accuracy and poor noise resistance in the parameter identification of traditional permanent magnet synchronous motors are solved, achieving high-precision and robust multi-parameter identification, which is applicable to fields such as electric vehicles, industrial servo motors, and aerospace.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- TONGJI UNIV
- Filing Date
- 2025-11-18
- Publication Date
- 2026-07-17
AI Technical Summary
Traditional methods for identifying parameters of permanent magnet synchronous motors are inaccurate and have poor noise resistance. They rely on specific excitation conditions and are difficult to achieve high-precision and robust online identification of multiple parameters.
A physical information neural network model is constructed, which combines a composite loss function of data fitting terms and physical constraint terms. Using motor operating data and voltage equation constraints, multiple parameters of the motor are updated through a backpropagation algorithm to ensure that the identification process follows the physical laws of the motor.
It achieves high-precision and robust multi-parameter identification, can perform online or offline identification under normal operating conditions, has a wide range of applications, strong anti-noise capability, and meets the needs of high-performance motor control.
Smart Images
Figure CN121618895B_ABST