CN-121984397-A - High-speed permanent magnet motor noninductive control method and system based on fast Fourier transform
Abstract
The invention relates to the technical field of non-inductive control, in particular to a non-inductive control method and a system of a high-speed permanent magnet motor based on fast Fourier transformation. According to the invention, a gradient change sequence is constructed through the instantaneous peak value of the three-phase current of the stator, the dynamic judgment of the electric inertia abrupt change state is carried out, the offset trend of the running state of the channel is effectively perceived, meanwhile, the main response of the disturbance frequency is locked through Fourier transformation, the response gain difference value after current injection is used as a sensitivity reference, the injection configuration and disturbance regulation parameters are adjusted in real time, and the rotor electric angle estimation precision, the current control response sensitivity and the dynamic stability are improved under the condition of lack of a sensor.
Inventors
- Meng Jianmei
Assignees
- 晋中职业技术学院
Dates
- Publication Date
- 20260505
- Application Date
- 20260408
Claims (9)
- 1. The fast Fourier transform-based high-speed permanent magnet motor noninductive control method is characterized by comprising the following steps of: S1, acquiring operation data of a high-speed permanent magnet motor, constructing a gradient change sequence according to a current amplitude difference value between two adjacent periods, judging according to a stator current effective value change limiting threshold value, recording a pulsation enhancement state if the current effective value change limiting threshold value is larger than the threshold value, recording a pulsation convergence state if the current effective value change limiting threshold value is smaller than the threshold value, and generating a current pulsation response state mark; s2, mapping the current pulsation state to an electric inertia change trend interval based on the current pulsation response state mark, judging whether an electric inertia mutation state exists or not, adjusting the disturbance injection reference vector direction, and generating a disturbance estimation correction direction parameter set; S3, calling the disturbance estimation correction direction parameter set, obtaining an instantaneous active power value and an estimated rotating speed value of each channel in a multi-inverter bridge arm channel structure of the permanent magnet motor, adjusting a reference current amplitude instruction of the channel, and generating current injection bias structure information; and S4, based on the current injection bias structure information, performing fast Fourier transform on the power sequence, extracting frequency intervals of the first three of the amplitude values, performing distance calculation with reference frequencies in a preset disturbance frequency table, and selecting a minimum distance frequency value as a matching result to obtain a disturbance frequency main response identifier.
- 2. The fast fourier transform-based high-speed permanent magnet motor non-inductance control method according to claim 1, wherein the current pulsation response state mark comprises a periodic current gradient range judgment result, a stator current amplitude variation trend mark, current dynamic fluctuation boundary identification information and a periodic response window number, the disturbance estimation correction direction parameter set is specifically a rotor position disturbance direction reference angle, an inertia variation trend category mark, a current vector included angle variation amplitude threshold value and a disturbance injection azimuth correction factor, the current injection bias structure information comprises bias current amplitude setting values of all channels, a current channel number mapping table, an integral offset normalization coefficient distribution map and an injection amplitude deviation direction mark, and the disturbance frequency main response mark is specifically a spectrum peak frequency band index, a main response frequency band amplitude level, a frequency band-frequency mapping number and a current power spectrum response level interval.
- 3. The fast fourier transform-based high-speed permanent magnet motor sensorless control method according to claim 2, wherein the step of obtaining the current ripple response status flag specifically comprises: S111, acquiring operation data of the high-speed permanent magnet motor, wherein the operation data comprise instantaneous peaks of stator three-phase currents in a continuous control period in an acceleration stage, calling current peaks in each two adjacent periods, performing subtraction calculation, constructing a corresponding current amplitude change sequence, forming a gradient structure by taking a difference result between each pair of data in the sequence as a data node, and generating a current amplitude gradient sequence; S112, based on the current amplitude gradient sequence, extracting all gradient data nodes in a period window, calling the maximum value and the minimum value in the node sequence, calculating a difference value, taking the difference value as the current amplitude range in the corresponding period, and calculating to obtain a period pulsation intensity value; And S113, judging whether the amplitude limiting threshold value of the stator current effective value is exceeded or not based on the periodic pulse intensity value, marking as pulse enhancement if the amplitude limiting threshold value is exceeded, marking as pulse convergence if the amplitude limiting threshold value is not exceeded, establishing a corresponding state identification marking sequence, numbering the time sequence, and obtaining a current pulse response state marking.
- 4. The fast fourier transform-based high-speed permanent magnet motor sensorless control method according to claim 3, wherein the obtaining step of the disturbance estimation correction direction parameter set specifically includes: S211, calling an electric angle estimation value of a current period based on the current pulsation response state mark, extracting a corresponding d-axis current component and a corresponding q-axis current component in the period, respectively extracting a projection result of a current vector under a corresponding static coordinate system in a rotating coordinate system, and carrying out space matching mapping on a space vector formed by the current components and the current electric angle to obtain space vector projection offset information; S212, acquiring electric angular velocity estimated values in a current period and the first two periods according to the space vector projection offset information, constructing an electric angular velocity change amplitude sequence in a three-period sequence, extracting corresponding current space vector included angles in the three periods, calculating an included angle change rate, establishing a disturbance intensity judgment reference interval, calculating an inertia change trend response value, judging the inertia change type of the inertia change trend response value in the reference interval, and establishing a mapping label to obtain an inertia mutation trend identifier; S213, extracting disturbance injection angle offset parameters in the corresponding inertia state mapping relation and comparing the disturbance injection angle offset parameters with the reference vector direction of the current period according to the inertia abrupt change trend identification, and adjusting vector rotation amplitude and amplitude range of the target direction in the disturbance injection path to obtain a disturbance estimation correction direction parameter set.
- 5. The fast fourier transform-based high-speed permanent magnet motor sensorless control method according to claim 4, wherein the current injection bias structure information obtaining step specifically includes: S311, calling the disturbance estimation correction direction parameter set, acquiring an instantaneous active power value and a period estimation rotating speed value in a corresponding period of each channel under a multi-inversion bridge arm channel structure of the permanent magnet motor according to the channel direction parameter, establishing a power time sequence according to the channel number, calculating a power change difference value of the channel at a corresponding sampling moment, and generating a channel power difference sequence; S312, according to the channel power differential sequence, performing integral operation in the period on differential data of the channel in three continuous periods, extracting integral total quantity of the sequence, and normalizing an integral result by using a unit current deviation ratio coefficient to obtain offset value distribution data of the channel, and obtaining a channel offset return value sequence; S313, judging with the current amplitude bias judging threshold value based on the channel deviation normalization value sequence, screening channels with normalization values exceeding the threshold value, adjusting reference current amplitude command values of the corresponding channels, establishing an amplitude adjustment record table of the modulated channels, and generating current injection bias structure information.
- 6. The fast fourier transform-based high-speed permanent magnet motor sensorless control method according to claim 5, wherein the step of obtaining the disturbance frequency main response identifier specifically comprises: S411, based on the current injection bias structure information, obtaining the direct current bus voltage at the input end of the permanent magnet motor and the three-phase current at the output end of the inversion unit, and performing product calculation on the voltage and the current according to the position corresponding to the sampling time point to obtain the instantaneous power value of the sampling point, and constructing an instantaneous power time sequence in a complete period; s412, performing fast Fourier transform on the sequence according to the instantaneous power time sequence, extracting amplitude information of corresponding frequency points in a frequency spectrum structure, screening frequency points with top three amplitude ranks, recording frequency corresponding values, and generating a power spectrum main frequency interval group; S413, according to the power spectrum main frequency interval group, calling frequency items in a set disturbance frequency reference table, calculating absolute differences among frequency points one by one, and screening the frequency with the smallest difference as a matching result to obtain a disturbance frequency main response identifier.
- 7. The fast fourier transform-based high speed permanent magnet motor sensorless control method of claim 6, further comprising the steps of: S5, calling the disturbance frequency main response identifier and current injection bias structure information, applying a disturbance signal corresponding to the frequency to a selected channel, collecting the current response amplitude of the channel in an injection period, calculating a disturbance response gain coefficient, executing difference judgment with a previous period value, and recording injection configuration if the difference is larger than a disturbance adjustment sensitivity reference value to generate a disturbance adjustment parameter group table item; The disturbance regulation and control parameter group table entry comprises an injection frequency and channel pairing entry, a periodic current response amplitude recording value, a disturbance response gain change value and a sensitivity reference matching result.
- 8. The fast fourier transform-based high-speed permanent magnet motor sensorless control method according to claim 7, wherein the step of obtaining the disturbance regulation parameter group table entry specifically comprises: s511, calling the disturbance frequency main response identification and current injection bias structure information, judging whether a frequency injection channel is matched with the bias current channel according to the corresponding relation between a frequency point and the current channel number, marking an injectable channel number, recording a frequency channel comparison item according to the injection direction, and generating a disturbance frequency channel matching structure; S512, according to the disturbance frequency channel matching structure, applying the selected frequency injection signal to the corresponding channel, collecting the current response peak value of the channel during injection, and calculating to obtain a disturbance response gain coefficient; S513, according to the disturbance response gain coefficient, carrying out item-by-item comparison with a set disturbance regulation sensitivity reference value, if the disturbance response gain coefficient exceeds a base criterion, recording the current injection frequency, the channel number and response data, constructing a corresponding recording field, and obtaining a disturbance regulation parameter group table item.
- 9. A fast fourier transform-based high speed permanent magnet motor sensorless control system for implementing the fast fourier transform-based high speed permanent magnet motor sensorless control method of any one of claims 1-8, the system comprising: The current response analysis module is used for acquiring the running data of the high-speed permanent magnet motor, constructing a gradient change sequence according to the current amplitude difference value between two adjacent periods, judging according to the stator current effective value change limiting threshold value, recording a pulsation enhancement state if the current limit value is larger than the threshold value, recording a pulsation convergence state if the current limit value is smaller than the threshold value, and generating a current pulsation response state mark; The disturbance correction analysis module maps the current pulsation state to an electric inertia change trend interval based on the current pulsation response state mark, judges whether an electric inertia mutation state exists, adjusts the disturbance injection reference vector direction, and generates a disturbance estimation correction direction parameter set; the current bias identification module is used for calling the disturbance estimation correction direction parameter set, obtaining the instantaneous active power value and the estimated rotating speed value of each channel in the multi-inversion bridge arm channel structure of the permanent magnet motor, adjusting the reference current amplitude instruction of the channel and generating current injection bias structure information; The disturbance frequency identification module is used for carrying out fast Fourier transform on the power sequence based on the current injection bias structure information, extracting frequency intervals of the first three of the amplitude values, carrying out distance calculation with reference frequencies in a preset disturbance frequency table, and selecting a minimum distance frequency value as a matching result to obtain a disturbance frequency main response identification; And the disturbance regulation and control processing module is used for calling the disturbance frequency main response identifier and the current injection bias structure information, applying a disturbance signal corresponding to the frequency to a selected channel, collecting the current response amplitude of the channel in an injection period, calculating a disturbance response gain coefficient, executing difference judgment with a previous period value, and recording injection configuration if the difference is larger than a disturbance regulation sensitivity reference value to generate a disturbance regulation and control parameter group table item.
Description
High-speed permanent magnet motor noninductive control method and system based on fast Fourier transform Technical Field The invention relates to the technical field of non-inductive control, in particular to a high-speed permanent magnet motor non-inductive control method and system based on fast Fourier transform. Background The technical field of non-inductive control is a control method for estimating rotor position information according to motor types such as permanent magnet synchronous motors and the like without depending on a position sensor in the operation process. The technology mainly relies on motor body parameters and electrical response characteristics, adopts control algorithms such as a back electromotive force observation method, a high-frequency signal injection method, a model reference self-adaptive system method and the like through real-time sampling data of electrical quantities such as voltage, current and the like, dynamically calculates motor rotor position information in a controller, and realizes closed-loop vector control or direct torque control. The non-inductive control technology is widely applied to scenes with strict requirements on volume, cost and structural complexity, such as electric automobiles, industrial automation systems, electric tools, aerospace equipment and the like, and has the advantages of high reliability, low cost, easiness in maintenance and the like. The high-speed permanent magnet motor noninductive control method is a control method for a high-speed running permanent magnet synchronous motor, and aims to accurately estimate the position of a rotor magnetic pole and realize high-efficiency running control under the condition that the motor lacks a position feedback sensor. The method generally combines the electrical parameter change characteristics of the high-speed motor, and adopts means such as potential counter electromotive force estimation, rotation coordinate transformation, speed self-adaptive control mechanism and the like based on a mathematical model to realize stable start-stop, high-speed operation and torque output adjustment. The application field of the method comprises scenes such as a high-speed centrifugal compressor, an aviation electric propulsion system, a high-speed numerical control main shaft, new energy vehicle driving and the like. The traditional control method relies on stator current change trend and model back electromotive force estimation to infer the rotor position, when the current change is severe in the initial stage of high-speed operation and the back electromotive force amplitude is insufficient, the electric angle is difficult to accurately model and track, especially under the condition of frequent starting or dynamic load switching, the current vector is easy to generate disturbance reverse change to cause estimation deviation, so that rotational speed adjustment lag and torque output fluctuation are caused, meanwhile, a compensation mechanism is not established for the dynamic power difference relation among channels in the traditional method, so that the channel mismatch rate under a multi-bridge arm structure is increased, the problems of high-frequency oscillation and uneven energy distribution are typically presented, and long-term operation can cause operation risks such as overheating of an inverter, reduction of response efficiency and the like. Disclosure of Invention The invention aims to solve the defects in the prior art, and provides a fast Fourier transform-based high-speed permanent magnet motor noninductive control method and a fast Fourier transform-based high-speed permanent magnet motor noninductive control system. In order to achieve the aim, the invention adopts the following technical scheme that the high-speed permanent magnet motor noninductive control method based on the fast Fourier transform comprises the following steps: S1, acquiring operation data of a high-speed permanent magnet motor, constructing a gradient change sequence according to a current amplitude difference value between two adjacent periods, judging according to a stator current effective value change limiting threshold value, recording a pulsation enhancement state if the current effective value change limiting threshold value is larger than the threshold value, recording a pulsation convergence state if the current effective value change limiting threshold value is smaller than the threshold value, and generating a current pulsation response state mark; s2, mapping the current pulsation state to an electric inertia change trend interval based on the current pulsation response state mark, judging whether an electric inertia mutation state exists or not, adjusting the disturbance injection reference vector direction, and generating a disturbance estimation correction direction parameter set; S3, calling the disturbance estimation correction direction parameter set, obtaining an instantaneous active power value and