Dynamic monitoring technology for anti demagnetiza

2022-08-11
  • Detail

Permanent magnet anti demagnetization dynamic monitoring technology

the research on permanent magnet demagnetization of permanent magnet motor not only carries out simulation research in the design process to prevent permanent magnet demagnetization, but also many scholars study the dynamic monitoring of the state of permanent magnet in the operation of the motor, that is, in use, so as to prevent demagnetization. Dynamic monitoring technology is divided into open-loop dynamic monitoring and dynamic monitoring auxiliary closed-loop control. The former only monitors the relevant parameters of the permanent magnet motor, and then indirectly obtains the state information of the permanent magnet, and does not adopt closed-loop control; The latter monitors the indirectly obtained permanent magnet status information, and dynamically controls it according to its status information to prevent more serious demagnetization

1 open loop dynamic monitoring

when monitoring the state of permanent magnet, literature [13] proposed a method to use Fourier decomposition of the stator current of permanent magnet motor under power generation conditions. If 0.5 times of harmonic occurs, with its good resistance to damage, fire resistance and permeability (including various liquids including special hydraulic working oil) and 1.5 times of harmonic, demagnetization of permanent magnet can be judged, However, the current in this method is only limited to the branch current in the parallel winding, and the analysis of the series winding current and the total current of the parallel winding cannot be established. And this method mainly analyzes the magnet state of the permanent magnet motor under the power generation condition, which is not applicable to the permanent magnet state monitoring under the electric condition

Spanish scholar j.roser0 and others used to extract the stator current signal of permanent magnet motor in operation. In many years of research, they processed the signal by various signal processing methods such as fast Fourier transform, Hilbert Huang transform, continuous wavelet transform, discrete wavelet transform, etc., to judge whether the permanent magnet demagnetization occurred, so as to complete the demagnetization fault monitoring of permanent magnet

among them, the fast Fourier transform is used to analyze the stator current and zero sequence current. In the high-speed region, the L and 5 harmonics of the stator current are used to reflect whether the demagnetization of the permanent magnet occurs; In the medium speed region, the stator current L, 5, 13 and 17 harmonics are used to reflect whether the demagnetization of the permanent magnet occurs; In the low-speed region, the 6th and 9th harmonics of stator current are used to reflect whether the demagnetization phenomenon of permanent magnet occurs after exceeding the yield load

in the stator current analysis using Hilbert Huang transform, firstly, the modal decomposition of the stator current signal is carried out to obtain the intrinsic modal function. Ran carries out Hilbert transform on the modal function to obtain the Hilbert spectrum and the real-time frequency energy spectrum, so as to analyze whether the demagnetization of the permanent magnet occurs

use wavelet transform (including continuous wavelet transform and discrete wavelet transform) to analyze the stator current in the study of permanent magnet demagnetization of permanent magnet motor. The research shows that continuous wavelet transform can reduce the data signal parameters to judge whether demagnetization occurs, which is suitable for industrial rapid detection; The discrete wavelet transform contains the whole spectrum information of the stator current signal, so it can get more accurate detection results

at the same time, after years of research, Spanish scholar j.roser0 and others analyzed the stator current signal and monitored the demagnetization fault of permanent magnet, and obtained the following conclusions:

(1) the fast Fourier transform lost time information, so it can not be used to monitor the loss of excitation of motor under dynamic conditions

(2) the time-frequency domain energy spectrum analysis method obtained by Hilbert Huang transform is only applicable to the case of slow speed change in the process of dynamic detection of loss of excitation

(3) wavelet analysis can solve the dynamic detection when the speed and torque change rapidly, which is a good tool for loss of excitation analysis

however, the above analysis method only dynamically monitors the demagnetization of permanent magnets, and there is no corresponding response method such as feedback control

2 dynamic monitoring ensures excellent high-yield measurement auxiliary closed-loop control

literature [18] proposed a "D-state observer" to observe permanent magnet rotor flux linkage, which has the characteristics of less steady-state information and wide application speed range. Literature [19] proposes to consolidate the foundation of building a manufacturing power and to develop an improved back EMF method for indirect observation of permanent magnet rotor flux linkage, which is only applicable to medium and high-speed occasions. The above method is mainly used in the field of sensorless permanent magnet motor control, but it can be used to observe the permanent magnet flux linkage model to indirectly estimate the state information of the permanent magnet, so as to prevent more serious demagnetization of the permanent magnet

literature [20] directly proposes a permanent magnet magnetic field state monitoring method for permanent magnet synchronous motor. This method uses stator current and permanent magnet flux linkage as state variables, and constructs a Kalman filter to estimate the amplitude and direction of permanent magnet flux linkage, so as to be used as a permanent magnet flux linkage observer in the closed-loop control system of permanent magnet motor. By adjusting parameters such as shaft orientation and voltage limit cycle, So as to prevent more serious demagnetization of the permanent magnet. This study provides a reference for dynamic monitoring of permanent magnet state and closed-loop control

Copyright © 2011 JIN SHI