Intelligent improved permanent magnet synchronous motor position prediction method and system

Abstract

The invention relates to an intelligent improved permanent magnet synchronous motor position prediction method and system. The method comprises the following steps: injecting a first pulse voltage into a preset injection phase, and determining a closest first target phase of a rotor in a first injection stage by comparing current of each phase; injecting a second pulse voltage into the first target phase, comparing each phase current to determine a closest second target phase of the rotor in a second injection stage, and determining a closest third target phase of the rotor in the first target phase and the second target phase; injecting a reverse third pulse voltage into the third target phase, and determining the polarity position of the rotor by comparing the phase current responded by the third target phase in each injection stage; and a sector corresponding to the third target is divided into a plurality of angles, direct-axis pulse voltage is injected into each angle, and the angle position of the rotor is determined by comparing the direct-axis current of each angle. By adopting the method, a simple and efficient motor initial position prediction function can be realized.

Description

Technical Field

[0001] This application relates to the field of motor control technology, and in particular to an intelligent improved method and system for predicting the position of a permanent magnet synchronous motor. Background Technology

[0002] In the field of motor control technology, there is a need to achieve stable motor start-up by completing initial position identification without position sensors.

[0003] In related position detection methods, the phase current response is usually obtained by rotating high-frequency injection and pulse high-frequency injection to infer the rotor position. However, these methods have the disadvantages of complex algorithms and high implementation costs. In addition, the traditional six-pulse injection method requires six pulses to be sent continuously, which takes a long time to execute, resulting in low startup efficiency and insufficient position determination. Summary of the Invention

[0004] Therefore, it is necessary to provide an intelligent and improved method, system, computer device, and computer-readable storage medium for predicting the position of a permanent magnet synchronous motor, addressing the aforementioned technical problems.

[0005] In a first aspect, this application provides an intelligent improved method for predicting the position of a permanent magnet synchronous motor, comprising: A first pulse voltage is injected into a preset injection phase. By comparing the phase currents of each phase based on the first pulse voltage, the first target phase that is closest to the rotor among the non-injection phases in the first injection stage is determined. A second pulse voltage is injected into the first target phase. By comparing the phase currents of each phase based on the second pulse voltage, the closest second target phase among the non-injected phases of the rotor in the second injection stage is determined. The closest third target phase of the rotor is determined between the first target phase and the second target phase. A reverse third pulse voltage is injected into the third target phase, and the polarity position of the rotor is determined by comparing the phase currents that the third target phase responds to at each injection stage. The sector corresponding to the third target is divided into multiple angles according to a preset angle step size. A direct-axis pulse voltage is injected into each angle. The angular position of the rotor is determined by comparing the direct-axis currents that respond to the direct-axis pulse voltages at each angle.

[0006] Secondly, this application also provides an intelligent improved permanent magnet synchronous motor position prediction system, comprising: The first positioning module is used to inject a first pulse voltage into a preset injection phase, and by comparing the phase currents of each phase based on the first pulse voltage, determine the first target phase that the rotor is closest to among the non-injection phases in the first injection stage. The second positioning module is used to inject a second pulse voltage into the first target phase, and by comparing the phase currents of each phase based on the second pulse voltage, determine the closest second target phase among the non-injected phases of the rotor in the second injection stage, and determine the closest third target phase of the rotor among the first target phase and the second target phase. A polarity prediction module is used to inject a reverse third pulse voltage into the third target phase and determine the polarity position of the rotor by comparing the phase currents that the third target phase responds to at each injection stage. An angle prediction module is used to divide the sector corresponding to the third target into multiple angles according to a preset angle step size, inject a direct-axis pulse voltage into each angle, and determine the angular position of the rotor by comparing the direct-axis currents that respond to the direct-axis pulse voltages at each angle.

[0007] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the above steps.

[0008] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the above steps.

[0009] The aforementioned intelligent improved permanent magnet synchronous motor position prediction method, system, computer equipment, and computer-readable storage medium firstly, by applying a first pulse voltage to a preset injection phase and performing a lateral comparison of the currents of each phase, the spatial proximity relationship of the rotor under initial injection conditions is reflected by the difference in phase currents, thereby obtaining a first target phase for coarse sector positioning; secondly, by applying a second pulse voltage to the first target phase and performing another lateral comparison of the currents of each phase, the current sensitivity under the two excitation actions forms a cross-validation relationship, thereby determining a third target phase with higher spatial directivity; on the other hand, by applying a reverse third pulse voltage to the third target phase and performing a longitudinal comparison of the phase currents in the multi-injection stages... By comparing the rotor magnetic pole orientation with the response difference, the polarity position is obtained. On the other hand, by applying multi-angle direct-axis pulse voltages in the sector corresponding to the third target and comparing the direct-axis current, the rotor angle is determined with the response difference, thereby obtaining the angular position. Based on this, in the entire technical solution, under the application of mixed voltage pulses, the rotor position is analyzed layer by layer from coarse positioning to fine positioning, and the corresponding polarity and angular positions are determined. This allows the rotor position to be completely and reliably determined in a short time, thereby greatly shortening the detection delay and improving the stability and fault tolerance of sensorless control. This achieves a simple, efficient, and easy-to-engineer-application initial position prediction function for permanent magnet synchronous motors. Attached Figure Description

[0010] To more clearly illustrate the technical solutions in the embodiments or related technologies of this application, the accompanying drawings used in the description of the embodiments or related technologies will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0011] Figure 1 This is a flowchart illustrating an intelligent improved permanent magnet synchronous motor position prediction method in one embodiment; Figure 2 This is a schematic diagram of the sector positions of a permanent magnet synchronous motor in one embodiment; Figure 3 This is a structural block diagram of a direct-axis pulse voltage injection process in one embodiment; Figure 4 This is a structural block diagram of an intelligent improved permanent magnet synchronous motor position prediction system in one embodiment. Detailed Implementation

[0012] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0013] In one exemplary embodiment, such as Figure 1 As shown, an intelligent improved method for predicting the position of a permanent magnet synchronous motor is provided. This embodiment uses the application of this method to a terminal as an example for illustration, which includes the following steps S101 to S104.

[0014] Step S101: Inject a first pulse voltage into a preset injection phase, and determine the first target phase that is closest to the rotor among the non-injection phases in the first injection stage by comparing the phase currents of each phase based on the first pulse voltage.

[0015] For example, firstly, a first pulse voltage is applied to a preset injection phase, causing that phase to generate a current response characterizing the relative positional relationship of the rotor under excitation. Simultaneously, other phases not injected with the first pulse voltage generate a weaker current response, but which reflects spatial positional differences, due to the coupling effect between the motor windings. The purpose of this entire process is to reveal the relative positional relationship between the rotor poles and each phase winding by using the differences in phase currents generated by the same excitation source.

[0016] Furthermore, the controller synchronously samples the three-phase currents during the first pulse voltage application to ensure that current rise characteristics, steady-state characteristics, or fall characteristics are captured within the same time window. Subsequently, by performing a lateral comparison of the amplitude, rate of change, or steady-state offset characteristics of the three-phase currents, the non-injected phase that is most sensitive to the rotor poles under the corresponding excitation is identified, where a larger phase current amplitude indicates that the phase is closer to the rotor poles in space. When the comparison is complete, the non-injected phase with the larger current value is marked as the first target phase to indicate the location region closest to the rotor poles during the first injection phase.

[0017] Based on this, by using this method of single excitation and cross-phase comparison, the approximate range of the rotor's initial position in phase space is narrowed down to a roughly sector interval corresponding to the first target, providing a basis for subsequent refined orientation identification.

[0018] Furthermore, in permanent magnet synchronous motors, the closer the rotor magnetic poles are to the stator direct axis, the easier it is for the magnetic circuit to saturate, causing a decrease in direct axis inductance. Based on this, under the same pulse voltage, the phase with smaller inductance will generate a larger current. Therefore, by injecting voltage in a fixed direction and comparing the magnitude of the current in each phase, it is possible to determine which phase and sector the rotor is closer to, thus achieving initial position identification.

[0019] Step S102: Inject a second pulse voltage into the first target phase. By comparing the phase currents of each phase based on the second pulse voltage, determine the second target phase that is closest to the rotor among the non-injected phases in the second injection stage. Determine the third target phase that is closest to the rotor between the first target phase and the second target phase.

[0020] For example, the first target phase determined in the aforementioned steps is used as the injection phase in the second injection stage. By applying a second pulse voltage to this phase, the rotor forms another set of distinguishable response currents under the new excitation direction. The sampling process at this time is consistent with the previous stage, that is, the three-phase current is monitored synchronously. However, due to the change in the injection phase, the coupling effect of the non-injection phases changes, thus causing the current distribution in the current injection stage to present a new contrast relationship. Then, the response currents formed by each non-injection phase based on the second pulse voltage are compared to determine the non-injection phase that is most sensitive to the rotor poles under the new excitation direction, and this phase is marked as the second target phase.

[0021] Furthermore, the combination of the first target phase and the second target phase reflects a more precise position range of the rotor magnetic poles relative to the three-phase spatial coordinate system, and a final selection needs to be made between the two to determine the third target phase. This selection process is based on the magnitude relationship of the current of each phase in the first injection stage and the second injection stage, that is, by comparing the current characteristics of the target phases corresponding to the two injection stages, it is determined which target phase the rotor is closer to.

[0022] Specifically, a combined judgment is made based on the difference in response current caused by the change in injection direction. That is, by comparing the amplitude, rate of change, or steady-state offset characteristics of the first target phase and the second target phase in two injection stages, the phase that better reflects spatial sensitivity under different excitation is identified. If the second target phase shows a more prominent response characteristic in the cross-stage comparison, it is determined to be the third target phase; otherwise, the first target phase is selected. After the above comparison is completed, the third target phase serves as the spatial reference direction for subsequent position identification, so that the rotor polarity position and angular position can be determined under the same spatial reference.

[0023] Step S103: Inject a reverse third pulse voltage into the third target phase, and determine the polarity position of the rotor by comparing the phase currents that the third target phase responds to at each injection stage.

[0024] For example, the controller applies a third pulse voltage in the opposite direction to the third target phase determined in the preceding steps, causing the phase to generate a response current with directional discrimination under reverse excitation. That is, since the third target phase is injected in the reverse phase during the third injection stage, if the third target phase is also an injection phase during the first or second injection stage, then it is injected in the positive phase during the first or second injection stage. Therefore, the change in the excitation direction of this phase will cause the response current between the third injection stage and the corresponding injection stage to form differences in amplitude, rate of change, or steady-state offset characteristics that can be used for polarity identification.

[0025] To ensure the validity of the comparison, the current data generated by the third target phase at each injection stage are uniformly sampled and normalized to ensure consistent reference across time scales, amplitude scales, and sampling windows for multiple excitation actions. Based on this, the current response trends under different excitation directions are compared, and the spatial positional relationship of the magnetic poles relative to the excitation direction is determined by observing whether the current exhibits a higher response amplitude or a stronger change amplitude in reverse excitation.

[0026] If the current characteristic obtained from reverse excitation shows a stronger response, it indicates that the rotor's N pole orientation is closer to the opposite direction of the third target phase; if the current characteristic obtained from forward excitation shows a stronger response, it indicates that the rotor's N pole orientation is closer to the positive direction of the third target phase. Based on this, through this longitudinal comparison method based on the same phase but different excitation directions, the polarity determination does not depend on the absolute current magnitude, but on the response difference under direction switching; thus, the polarity position is clarified by the response law of the third target phase, so that the third target phase not only indicates the sector position, but also carries polarity information indicating the magnetic pole orientation.

[0027] Step S104: Divide the sector corresponding to the third target into multiple angles according to a preset angle step size, inject direct-axis pulse voltage into each angle, and determine the angular position of the rotor by comparing the direct-axis currents that respond to the direct-axis pulse voltages at each angle.

[0028] For example, based on the sector corresponding to the third target determined in the preceding steps, the controller first divides the sector into several discrete angles according to a preset angular step size. Then, at each discrete angle within the sector, a pulse voltage in the direct axis direction is sequentially applied to the permanent magnet synchronous motor. By acquiring the direct axis current generated by the pulse voltage excitation within a fixed time window, each angular position corresponds to a set of current data with direction mapping characteristics. To ensure the comparability of the currents obtained at different angles, the injection amplitude, sampling frequency, and sampling timing are kept consistent throughout the process, so that the excitation method and data acquisition process are under the same reference system.

[0029] As each discrete angle is excited one by one, the direct-axis current at different angle positions will exhibit differences in amplitude or response trend due to the spatial relationship between the rotor magnetic poles and the excitation direction. Therefore, by comparing the direct-axis currents corresponding to all angles laterally, the angle position that best reflects the magnetic flux alignment direction can be identified. This angle position can not only indicate the specific position of the rotor within the sector, but also form a correspondence with the polarity position identified in the previous step, so as to construct a complete initial rotor position in space.

[0030] In the aforementioned intelligent improved permanent magnet synchronous motor position prediction method, in step S101, by applying a first pulse voltage to a preset injection phase and comparing the currents of each phase laterally, the spatial proximity relationship of the rotor under initial injection conditions is reflected by the difference in phase currents, thereby obtaining a first target phase for sector coarse positioning; in step S102, by applying a second pulse voltage to the first target phase and comparing the currents of each phase laterally again, the current sensitivity under the two excitation actions forms a cross-validation relationship, thereby determining a third target phase with higher spatial directivity; in step S103, by applying a reverse third pulse voltage to the third target phase and comparing the phase currents of the multi-injection stages longitudinally, The rotor magnetic pole orientation is determined by the response difference, thereby obtaining the polarity position; in step S104, by applying multi-angle direct-axis pulse voltages in the sector corresponding to the third target and comparing the direct-axis current, the rotor angle is determined by the response difference, thereby obtaining the angular position; based on this, in the entire technical solution, under the application of mixed voltage pulses, the rotor position is analyzed layer by layer from coarse positioning to fine positioning, and the corresponding polarity position and angular position are determined, so that the rotor position can be completely and reliably determined in a short time, thereby greatly shortening the detection delay, improving the stability and fault tolerance of sensorless control, and thus realizing a simple, efficient and easy-to-engineer permanent magnet synchronous motor initial position prediction function.

[0031] In an exemplary embodiment, a first pulse voltage is injected into a preset injection phase, and the first target phase closest to the rotor among the non-injection phases in the first injection stage is determined by comparing the phase currents of each phase based on the first pulse voltage, including steps S201 to S203.

[0032] Step S201: Inject a first pulse voltage into a preset injection phase, and compare the phase currents of each non-injection phase in the first injection stage based on the first pulse voltage.

[0033] For example, phase A is used as the preset injection phase. A first pulse voltage is applied to phase A, causing it to generate a current response that can be used for position identification within a short period of time. Since the pulse voltage only acts on phase A, the other phases B and C, which are not directly excited, will generate a current response with a smaller amplitude but related to the rotor's spatial position under the winding coupling effect. Furthermore, to ensure data comparability, the controller synchronously samples the phase currents of the three phases during the pulse application and extracts their respective response amplitudes within a fixed sampling window, so that the collected currents can reflect the flux linkage coupling state within the same time period.

[0034] Step S202: If the absolute difference between the phase currents of each non-injected phase is greater than a preset threshold, then the non-injected phase corresponding to the phase current with the maximum value is taken as the first target phase.

[0035] For example, based on the phase current amplitudes of phases B and C (which are non-injected phases), the absolute difference between them is calculated, and this absolute difference is compared with a preset threshold to determine whether the first target phase can be directly identified. Essentially, the coupling difference caused by the rotor position manifests as significant current asymmetry in a specific direction. Therefore, when the current difference between the two non-injected phases significantly exceeds the threshold, it indicates that a certain phase is closer to the rotor magnetic pole, and its phase current amplitude better reflects the directional characteristics. In this case, highly reliable phase-level direction determination can be achieved by selecting the phase with the larger phase current amplitude as the first target phase.

[0036] For example, if the calculation results show that the amplitude of the current in phase B is greater than that in phase C, and the absolute difference between the two exceeds a preset threshold, it indicates that under the excitation condition of phase A, the spatial direction of phase B is closer to the rotor magnetic pole, so phase B is directly determined as the first target phase.

[0037] Furthermore, the preset threshold serves to filter out insignificant differences caused by noise, minor disturbances, or small rotor offsets, ensuring sufficient stability and fault tolerance in the decision-making process. When the absolute difference is greater than the preset threshold, it indicates that the current difference has significant directional indicative value and can directly support the selection of the first target phase without additional injection or further comparison.

[0038] Step S203: If the absolute difference of the phase current between each non-injection phase is less than or equal to a preset threshold, then a new injection phase is determined among each non-injection phase and a preset pulse voltage is injected. Based on the phase current of each phase in the current injection stage that is responded to by the pulse voltage, the first target phase is determined.

[0039] For example, when the absolute difference between the amplitudes of phase B and phase C currents does not exceed a preset threshold, i.e., when their amplitudes are close, it indicates that the current excitation conditions are insufficient to form an effective direction distinction. Therefore, to further improve the identification accuracy, a new injection phase needs to be selected from phases B and C and a pulse voltage needs to be applied so that the new excitation direction forms a more sensitive response relationship with the rotor magnetic poles.

[0040] For example, in a scenario where phase A is the initial injection phase, if the absolute difference in phase current amplitude between phase B and phase C is small, any uninjected phase can be selected as the new injection phase, such as phase B, and a new pulse voltage can be applied to that phase. At this time, phase B generates the main current response, while phase A and phase C form a new current response under the new winding coupling.

[0041] Based on this, since the new excitation direction changes the magnetic flux coupling effect, the spatial sensitivity observed from the excitation direction is improved, thus enabling the acquisition of a more discriminative current difference than in the first injection stage. If the absolute difference in phase current amplitude between the non-injected phases formed in the new injection stage increases significantly, the first target phase can be directly determined based on this absolute difference. If the absolute difference is still insufficient for judgment, the injection phase can be reselected and the excitation process can be repeated until the difference exceeds a preset threshold.

[0042] Furthermore, if phase A is taken as the injection phase and a pulse voltage is applied, phase currents will be generated in the three phases of the permanent magnet synchronous motor, and the expressions for the three-phase currents are as follows: (1) In equation (1), , , These represent the phase currents of phases A, B, and C respectively under excitation in the direction of phase A. Represents the fundamental DC component. Represents the amplitude component of the second harmonic. The rotor electrical angle is represented by . Based on this, a functional relationship between the three-phase current and the rotor electrical angle is established in equation (1), which describes how the current of each phase changes with the rotor position in a specific amplitude pattern after a single pulse injection, thereby assisting in the determination of the rotor position.

[0043] In this embodiment, in step S201, by applying a first pulse voltage to the preset injection phase and comparing the current responses of each non-injection phase, the initial injection stage can obtain distinguishable current characteristics based on the same excitation conditions. In step S202, by determining whether the current difference between each non-injection phase exceeds a preset threshold, the corresponding non-injection phase can be quickly identified as the first target phase when the current difference is significant, avoiding detection delay caused by repeated excitation. In step S203, when the current difference does not exceed the preset threshold, a new injection phase is switched and re-excited, so that the positioning process can still obtain current characteristics that can be used for judgment under weak difference conditions, thereby improving the reliability of the first target phase determination. Based on this, in the entire technical solution, the acquisition process of the first target phase has both speed and robustness, making the phase-level positioning of the initial position more accurate and adaptable.

[0044] In an exemplary embodiment, a second pulse voltage is injected into the first target phase, and the closest second target phase among the non-injected phases of the rotor in the second injection stage is determined by comparing the phase currents of each phase based on the second pulse voltage, including steps S301 to S303.

[0045] Step S301: Inject a second pulse voltage into the first target phase, and compare the phase currents of each non-injected phase in the second injection stage based on the second pulse voltage.

[0046] For example, when phase C is determined to be the first target phase, a second pulse voltage is applied to phase C, causing it to generate a current response that can be used for position identification within a short period of time. Since the pulse voltage only acts on phase C, the other phases A and B, which are not directly excited, will generate a current response with a smaller amplitude but related to the rotor's spatial position under the winding coupling effect. Furthermore, to ensure data comparability, the controller synchronously samples the phase currents of the three phases during the pulse application and extracts their respective response amplitudes within a fixed sampling window, so that the collected currents can reflect the flux linkage coupling state within the same time period.

[0047] Step S302: If the absolute difference between the phase currents of each non-injected phase is greater than a preset threshold, then the non-injected phase corresponding to the phase current with the maximum value is taken as the second target phase.

[0048] For example, based on the phase current amplitudes of phases A and B (which are non-injected phases), the absolute difference between them is calculated, and this absolute difference is compared with a preset threshold to determine whether the second target phase can be directly identified. Essentially, the coupling difference caused by the rotor position manifests as significant current asymmetry in a specific direction. Therefore, when the current difference between the two non-injected phases significantly exceeds the threshold, it indicates that one phase is closer to the rotor magnetic pole, and its phase current amplitude better reflects the directional characteristics. In this case, highly reliable phase-level direction determination can be achieved by selecting the phase with the larger phase current amplitude as the second target phase.

[0049] For example, if the calculation results show that the amplitude of the current in phase A is greater than that in phase B, and the absolute difference between the two exceeds a preset threshold, it indicates that under the excitation condition of phase C, the spatial direction of phase A is closer to the rotor magnetic pole, so phase A is directly determined as the second target phase.

[0050] Furthermore, the preset threshold is used to determine whether the current difference obtained in the second injection stage has a stable directionality that can be directly used to determine the second target phase. When the absolute difference is greater than the preset threshold, it indicates that the excitation direction of the second injection stage has made the current distribution of each non-injected phase show a sufficiently clear difference, so that the second target phase that is closer to the rotor position can be directly selected.

[0051] Step S303: If the absolute difference of the phase current between each non-injected phase is less than or equal to a preset threshold, then the first target phase is taken as the second target phase.

[0052] For example, when the absolute difference between the current amplitude of phase A and the current amplitude of phase B does not exceed a preset threshold, that is, when the amplitudes of the two are close, it means that the current characteristics of the two under the current pulse excitation do not show an effective amplitude difference sufficient to distinguish the spatial position difference. Therefore, a reliable second target phase cannot be obtained by comparing the phases in the current injection stage. In this case, it is necessary to maintain the continuity of the identification direction based on the first target phase determined in the previous steps.

[0053] Specifically, in this embodiment, the first target phase is set as phase C, and phase C has already been verified as a phase relatively close to the rotor position through a comparison process in the previous injection stage. Furthermore, in the second injection stage, when the amplitudes of non-injected phases are close to each other, their current responses often exhibit a spatial symmetry along the three-phase structure. This spatial symmetry makes phase C, corresponding to phases A and B, more spatially consistent with the distribution characteristics of the rotor's actual position. Therefore, continuing to use phase C as the second target phase maintains the consistency and rationality of the position judgment. In addition, phase C, as the closest phase in the previous injection stage, already carries effective information about the rotor's spatial position. If the current injection stage lacks sufficient distinguishability, continuing to use phase C not only avoids the risk of misjudgment but also maintains the continuity of the identification direction across stages.

[0054] In this embodiment, in step S301, by applying a second pulse voltage to the first target phase and comparing the current responses of each non-injection phase, the second injection stage can obtain distinguishable current characteristics based on the same excitation conditions. In step S302, by determining whether the current difference between each non-injection phase exceeds a preset threshold, the corresponding non-injection phase can be quickly identified as the second target phase when the current difference is significant, thus achieving independent direction determination based on the current injection stage. In step S303, when the current difference does not exceed the preset threshold, the first target phase is used as the second target phase, thereby maintaining the continuity of the identification direction across stages. Based on this, in the entire technical solution, the determination of the second target phase has both independence within the stage and continuity across stages, making the initial position identification process both reliable and consistent in direction.

[0055] In an exemplary embodiment, determining the third target phase closest to the rotor between the first target phase and the second target phase includes steps S401 to S403.

[0056] Step S401: Based on the phase relationship of the voltage vectors corresponding to each injection stage, align and compare the phase current corresponding to the first target phase in the first injection stage with the phase current corresponding to the second target phase in the second injection stage.

[0057] For example, in order to further clarify which target phase is closer to the actual rotor position given that the first target phase and the second target phase have been obtained, it is necessary to perform a unified reference processing on the current response formed in different injection stages so that the comparison process has a directly measurable basis.

[0058] Specifically, the phase currents corresponding to the first target phase in the first injection stage and the phase currents corresponding to the second target phase in the second injection stage are phase-aligned according to the voltage vector directions of each injection stage, so that the two current quantities express their respective magnetic flux coupling degrees under the same reference direction. That is, the pulse directions used in different injection stages are not completely consistent, and the resulting current responses are direction-dependent. Without phase synchronization processing, the comparison of phase current amplitudes will deviate due to different excitation directions. Therefore, it is necessary to map the current quantities of the two injection stages to the same direction reference based on the directional relationship of the voltage vectors, so that the phase current amplitude comparison process is established in a consistent spatial coordinate system.

[0059] After phase alignment is completed, the two sets of current quantities after alignment can reflect the actual proximity between each winding and the rotor magnetic pole in its respective injection stage. If a target phase generates a stronger current in the corresponding injection stage, it means that the target phase is closer to the rotor magnetic pole in its excitation direction.

[0060] For example, when a first pulse voltage is applied to phase A in the first injection stage, and phase C is determined as the first target phase accordingly; and a second pulse voltage is applied to phase C in the second injection stage, and phase A is determined as the second target phase accordingly, then after phase alignment, the current characteristics of phase C in the first injection stage and the current characteristics of phase A in the second injection stage need to be compared in amplitude to determine which direction is closer to the actual rotor position.

[0061] Step S402: If the amplitude of the phase current corresponding to the second target phase in the second injection stage is greater than the amplitude of the phase current corresponding to the first target phase in the first injection stage, then the second target phase is taken as the third target phase.

[0062] Step S403: If the amplitude of the phase current corresponding to the first target phase in the first injection stage is greater than or equal to the amplitude of the phase current corresponding to the second target phase in the second injection stage, then the first target phase is taken as the third target phase.

[0063] For example, after completing the phase alignment and unified reference processing in the preceding steps, the phase current of the second target phase corresponding to the second injection stage is compared with the phase current of the first target phase corresponding to the first injection stage. In this process, since the phases have been aligned, the phase current amplitude can directly express the flux linkage response intensity of each target phase in its excitation direction. The larger the amplitude, the closer the phase direction is to the location of the rotor magnetic pole.

[0064] On the one hand, when the comparison results show that the phase current amplitude of the second target phase in the second injection stage is greater than that of the first target phase in the first injection stage, it indicates that under the excitation direction of the second injection stage, the spatial distance between the rotor magnetic pole and the second target phase is smaller, and the magnetic flux linkage effect is more obvious. Therefore, the response of the second injection stage has more directional indication significance.

[0065] On the other hand, when the comparison results show that the phase current amplitude of the second target phase in the second injection stage is less than that of the first target phase in the first injection stage, it indicates that under the excitation direction of the first injection stage, the spatial distance between the rotor magnetic pole and the first target phase is smaller, the magnetic flux linkage effect is more obvious, and therefore the response of the first injection stage has more directional indication significance.

[0066] On the other hand, when the comparison results show that the phase current amplitude of the second target phase in the second injection stage is equal to the phase current amplitude of the first target phase in the first injection stage, since the first target phase, as the closest phase in the previous injection stage, already carries effective information about the rotor's spatial position, in the boundary scenario where the phase current amplitudes are equal, using the first target phase as the third target phase can ensure the continuity and stability of the identification process.

[0067] In this embodiment, in step S401, the phase currents of the first target phase and the second target phase are aligned and compared according to the phase relationship of the voltage vectors corresponding to the two injection stages, so that the current from different excitation directions is established under a unified reference, ensuring that the responses of the two phases are comparable; in steps S402 and S403, based on the comparison result between the phase current amplitude of the second target phase in the second injection stage and the phase current amplitude of the first target phase in the first injection stage, a third target phase that is closer to the rotor position is determined between the first target phase and the second target phase; based on this, in the entire technical solution, the third target phase can be stably and accurately determined on the basis of the comparison of cross-stage current trends, thereby providing a reliable spatial reference for subsequent polarity identification and angle identification.

[0068] In an exemplary embodiment, a reverse third pulse voltage is injected into the third target phase, and the polarity position of the rotor is determined by comparing the phase currents that the third target phase responds to at each injection stage, including steps S501 to S503.

[0069] Step S501: Inject a reversed third pulse voltage into the third target phase, and compare the phase currents of the third target phase in each injection stage.

[0070] For example, since the third target phase has been determined to be the phase closest to the rotor spatial direction in the previous steps, the current response of this phase is most sensitive to changes in the magnetic pole orientation. Therefore, a reverse third pulse voltage is applied to the third target phase, and the current data collected in the third injection stage is uniformly calibrated with the current data of the same phase in the first and second injection stages, so that the current data obtained from the three excitations form a consistent reference in terms of time axis, sampling interval and amplitude reference.

[0071] Step S502: If the third target phase is the injection phase of two injection stages in the entire injection stage, then compare the absolute difference of the phase currents of the third target phase in the two injection stages to determine the polarity position of the rotor.

[0072] For example, if the injection phase in the first injection stage is phase A, the first target phase is C, and the second target phase is A, then the third target phase is either phase A or phase C. Since phase A is the positive excitation direction in the first injection stage and phase C is the positive excitation direction in the second injection stage, in this case, regardless of whether the third target phase is phase A or phase C, the current response accumulated by the third target phase as the excitation direction in each injection stage can be directly used, and the magnetic pole orientation can be determined by calculating the absolute difference between the phase currents under positive and negative excitation.

[0073] The absolute difference reflects the difference in current response of the same phase under forward and reverse excitation. This difference can be used to determine the degree of proximity of the rotor magnetic poles to the phase in the forward and reverse directions. When the phase current amplitude under forward excitation is greater than that under reverse excitation, it means that the rotor N pole is oriented in the forward direction closer to the phase. When the phase current amplitude under forward excitation is less than that under reverse excitation, it means that the rotor N pole is oriented in the reverse direction closer to the phase. This determines the polarity position of the rotor in the corresponding sector.

[0074] For example, when phase A is the third target phase, the amplitude of the phase current under the excitation of the positive phase of phase A in the first injection stage and the excitation of the negative phase of phase A in the third injection stage are compared. If the phase current under the excitation of the positive phase of phase A is greater than the phase current under the excitation of the negative phase of phase A, it means that the rotor N pole is closer to the positive phase of phase A; conversely, if the phase current under the excitation of the positive phase of phase A is less than the phase current under the excitation of the negative phase of phase A, it means that the rotor N pole is closer to the negative phase of phase A.

[0075] Step S503: If the third target phase is an injection phase in one of the injection phases of all injection phases or is not an injection phase in any injection phase, then a positive third pulse voltage is injected into the third target phase, and the absolute difference between the phase currents of the third target phase in the reverse injection phase and the forward injection phase is compared to determine the polarity position of the rotor.

[0076] For example, if the injection phase in the first injection stage is phase A, the first target phase is C, and the second target phase is B, then the third target phase is either phase B or phase C. If phase B is selected as the third target phase, since phase B was not used as the excitation direction in the preceding injection stage, its amplitude cannot be directly compared with the phase current used as the reverse excitation direction in the third injection stage. In this case, a positive pulse corresponding to the current reverse pulse is applied to the third target phase, thereby determining the magnetic pole orientation by calculating the absolute difference between the phase currents under positive and reverse excitation.

[0077] In this embodiment, in step S501, current characteristics for distinguishing magnetic pole orientation are obtained by injecting a reverse pulse voltage into the third target phase and collecting the current response formed by the phase under different injection stages; in step S502, based on the condition that the third target phase is used as the injection phase in both historical injection stages, the current amplitude of the phase under the two excitation directions is compared to directly determine the proximity of the rotor magnetic poles to the positive and negative directions of the phase; in step S503, based on the lack of comparable bidirectional excitation conditions for the third target phase in the historical injection stages, a positive pulse voltage is applied to the phase and compared with the reverse excitation result to ensure that the polarity judgment is analyzable under any injection phase; based on this, in the entire technical solution, the magnetic pole information of the rotor can be reliably distinguished within the determined spatial sector, making the initial position identification have higher directional clarity and judgment stability.

[0078] In an exemplary embodiment, after injecting a reversed third pulse voltage into the third target phase, the method further includes step S601.

[0079] Step S601: By comparing the phase currents of each non-injected phase in the third injection stage based on the third pulse voltage, the closest non-injected phase of the rotor in the third injection stage is determined, so as to determine whether the rotor is in the left or right position in the sector corresponding to the third target, and the sector is refined from a 60-degree range to a 30-degree range.

[0080] For example, after the controller completes the reverse pulse injection of the third target phase, in order to further distinguish whether the rotor is in the left or right position of the sector within a given sector, it needs to compare the current response of each non-injected phase to the third pulse voltage during the third injection phase. Specifically, the current amplitude of each non-injected phase is compared, and by identifying which non-injected phase exhibits a higher response amplitude under the current excitation, the subdivision pointing relationship of the rotor position within the corresponding sector of the third target phase can be determined.

[0081] In other words, once the third target phase has locked onto the 60-degree sector where the rotor is located, the left and right sides of this sector correspond to the magnetic flux coupling directions of two different non-injected phases. If the rotor faces the left side of the sector, the coupling between the non-injected phase and the magnetic pole corresponding to that side position will be stronger under the current excitation, resulting in a higher current amplitude. If the rotor faces the right side of the sector, the other non-injected phase will generate a stronger response under the current excitation, resulting in a higher current amplitude.

[0082] Therefore, after the comparison is completed, the side position corresponding to the non-injected phase with higher current amplitude is taken as the subdivided position of the rotor in the corresponding sector of the third target. Based on this, by comparing the difference in phase current amplitude between the two non-injected phases, the basic sector can be further refined from a 60-degree range to a 30-degree range, so that the rotor position in the corresponding sector of the third target can obtain more accurate left and right interval information.

[0083] In this embodiment, in step S601, after applying a pulse voltage to the third target phase in the third injection stage, the phase currents formed by each non-injected phase are compared. This allows for the acquisition of further spatial orientation information within the sector corresponding to the third target, inferring whether the rotor is on the left or right side within that sector. This enables the traditional 60-degree sector to be refined to a 30-degree range. Based on this, fine-grained positioning within the sector is achieved in the entire technical solution, giving the overall solution higher angle resolution and a shorter position convergence path.

[0084] In an exemplary embodiment, a direct-axis pulse voltage is injected at each angle, and the angular position of the rotor is determined by comparing the direct-axis currents that respond to the direct-axis pulse voltage at each angle, including steps S701 to S703.

[0085] Step S701: Inject direct-axis pulse voltage at each angle. By performing inverse Park transformation on the voltage at each angle based on the direct-axis pulse voltage in a preset virtual rotating coordinate system, the voltage at each angle in a preset two-phase stationary coordinate system is obtained.

[0086] For example, the sector corresponding to the third target is subdivided into discrete angles, and a corresponding direct-axis pulse voltage is set for each discrete angle in a preset virtual rotating coordinate system, so that the direct-axis pulse voltage can accurately describe the direct-axis voltage component consistent with the corresponding angle in the virtual rotating coordinate system. Then, the direct-axis pulse voltage set for each angle is transformed to a preset two-phase stationary coordinate system through inverse Park transformation, so that the direct-axis voltage component originally in the virtual rotating coordinate system obtains a clear α-axis voltage component and β-axis voltage component in the two-phase stationary coordinate system.

[0087] At this time, each angle forms a set of α-axis voltage components and β-axis voltage components corresponding to that angle. Through these components, a spatial voltage direction consistent with that angle can be constructed during the voltage modulation process of the inverter, thereby realizing the direct-axis pulse voltage as a physically applicable voltage command in the inverter.

[0088] Step S702: Perform space vector pulse modulation on the voltage at each angle in the two-phase stationary coordinate system to obtain duty cycle data for driving the preset three-phase inverter.

[0089] For example, firstly, each set of α-axis voltage components and β-axis voltage components is mapped onto a spatial voltage vector plane, and its vector representation in the spatial voltage vector plane is determined according to its sector and amplitude position. Subsequently, according to the spatial vector pulse modulation algorithm, the duration of the corresponding voltage vector within the corresponding sector, which is composed of two basic voltage vectors and a zero voltage vector, is calculated, so that this combination can equivalently generate the required voltage direction and amplitude within one modulation cycle.

[0090] In this process, the algorithm decomposes the required voltage into the switching action time of each arm of the three-phase inverter, and then further converts it into the three-phase duty cycle, enabling the three-phase inverter to output the voltage corresponding to that angle according to these duty cycles. By performing the same modulation calculation for each angle, a series of duty cycle data sets covering all discrete angles can be obtained, ensuring that the direct-axis pulse voltage at different angles is consistently applied to the motor windings.

[0091] Step S703: Drive the three-phase inverter to output three-phase current to the preset motor winding according to the duty cycle data, perform Clarke transformation and Park transformation on the three-phase current to obtain the direct-axis current of each angle in the preset real rotating coordinate system, and take the angle corresponding to the maximum direct-axis current as the angular position of the rotor.

[0092] For example, firstly, the inverter controls the switching time of each arm based on duty cycle data, enabling the three-phase inverter to synthesize a voltage direction and amplitude consistent with a specified angle within one modulation cycle, thereby applying a direct-axis excitation voltage to the motor windings. Then, the motor windings generate three-phase currents under the excitation voltage, which exhibit response characteristics in a three-phase stationary coordinate system that are related to the direct-axis direction of that angle.

[0093] Next, the three-phase currents are transformed to a two-phase stationary coordinate system using the Clarke transformation, allowing the current distribution to be presented in an ordered form with α-axis and β-axis current components. Subsequently, the Park transformation is used to transform these current components from the two-phase stationary coordinate system to a preset real rotating coordinate system, resulting in directly identifiable direct-axis currents in the real d-axis direction. By performing the same output control and transformation calculations for each angle, a set of direct-axis currents covering all angles can be obtained.

[0094] Finally, the direct-axis currents at all angles are compared. When the amplitude of the direct-axis current at a certain angle reaches the maximum among all angles, it indicates that the direction of that angle is closest to the direction of the rotor's direct axis, and its current response best reflects the true flux linkage alignment characteristics. Therefore, this angle is taken as the final angular position of the rotor.

[0095] Furthermore, when the permanent magnet synchronous motor is stationary, its rotational speed is zero. Therefore, its voltage equation expression in the preset rotating coordinate system is as follows: (2) In equation (2), Represents the direct-axis voltage component. Represents the direct-axis current component. Indicates direct-axis inductance. Indicates the rate of change of the direct-axis current component; Represents the quadrature-axis voltage component. Represents the quadrature-axis current component. Indicates quadrature axis inductance. This represents the rate of change of the quadrature-axis current component; This indicates the stator resistance.

[0096] If the pulse injection period is discrete time The current expression for each pulse injection cycle in the rotating coordinate system is as follows: (3) in, and Satisfy the following formula: (4) In equation (3), This represents the direct-axis current component at the nth sampling point. Represents the direct-axis voltage component. This represents the discretized response strength of the direct-axis current to the voltage pulse at the nth sampling point; This represents the quadrature-axis current component at the nth sampling point. Represents the quadrature-axis voltage component. This represents the discretized response strength of the quadrature-axis current to a voltage pulse. In equation (4), Indicates stator resistance. Indicates direct-axis inductance. Indicates quadrature axis inductance. Represents discrete time.

[0097] The above expression is based on the current components in the rotor's actual rotating coordinate system. However, the rotor's exact position is unknown. Therefore, when a direct-axis pulse voltage is injected at a selected angle in the virtual rotating coordinate system, the rotor's current expression in the virtual rotating coordinate system is as follows: (5) In equation (5), Represents the virtual direct-axis current component. Represents the virtual quadrature-axis current component. This represents the discretized response strength of the direct-axis current to the voltage pulse at the nth sampling point. This represents the discrete response strength of the quadrature-axis current to a voltage pulse. This represents the virtual angle at which a direct-axis pulse voltage is injected. Represents the actual electrical angle. This indicates the amplitude of the injected direct-axis pulse voltage.

[0098] Based on this, the above sets of expressions jointly characterize the mathematical relationship between the direct-axis current response and quadrature-axis current response induced by the direct-axis pulse voltage under different injection angles in the virtual rotating coordinate system, so that the offset between the virtual angle and the real electrical angle can be explicitly represented by the law of current change with angle. Then, by scanning multiple virtual angles and comparing the corresponding current responses, the initial position of the rotor can be deduced.

[0099] In this embodiment, in step S701, the excitation voltages at different angles are accurately constructed under a unified reference system by injecting direct-axis pulse voltages at each angle and obtaining the voltages in the two-phase stationary coordinate system through inverse Park transformation. In step S702, space vector pulse modulation is performed on the voltages in the two-phase stationary coordinate system to obtain duty cycle data, so that the excitation voltages at each angle can act on the three-phase inverter in a consistent manner. In step S703, the direct-axis current is obtained by driving the three-phase inverter with the duty cycle data and obtaining the direct-axis current through two coordinate transformations, so that the flux linkage response at each angle can be expressed in amplitude form, thereby determining the angular position of the rotor in the real rotating coordinate system. Based on this, in the entire technical solution, by constructing an angle scanning link with unified coordinates, unified excitation, and unified response, the precise identification of the rotor angular position is achieved.

[0100] In one exemplary embodiment, Figure 2A schematic diagram of the sector positions of a permanent magnet synchronous motor is shown, where the voltage vector in the positive phase direction of A is... The voltage vector in the positive phase direction of B is 100. The voltage vector in the positive phase direction of C is 010. It is 001.

[0101] First, in the first injection phase, a first pulse voltage is injected into phase A. By comparing phase B and phase C based on the first pulse voltage If the phase current amplitude of phase C is greater than that of phase B, then the closest first target phase among the non-injection phases of the rotor in the first injection stage is determined to be phase C; otherwise, it is phase B.

[0102] Furthermore, in the second injection stage, if the first target phase is phase C, then a second pulse voltage is injected into phase C. By comparing phase A and phase B based on the second pulse voltage If the phase current amplitude of phase A is greater than that of phase B, then the closest second target phase among the non-injection phases of the rotor in the second injection stage is determined to be phase A; otherwise, it is phase B. Further, the closest third target phase of the rotor is determined among the first and second target phases.

[0103] Furthermore, for example Figure 2 As shown, the permanent magnet synchronous motor involves 6 basic sectors, including sector I corresponding to A positive phase (A+), sector II corresponding to C reverse phase (C-), sector III corresponding to B positive phase (B+), sector IV corresponding to A reverse phase (A-), sector V corresponding to C positive phase (C+), and sector VI corresponding to B reverse phase (B-).

[0104] If the third target phase is phase C, then the sector corresponding to phase C includes sector II corresponding to the reverse phase of C and sector V corresponding to the forward phase of C; based on this, in the third injection stage, a reverse third pulse voltage is injected into phase C. By comparing the phase currents of phase C under forward and reverse pulses, if the amplitude of the phase current of phase C under the forward pulse is greater than the amplitude of the phase current of phase C under the reverse pulse, then the positive phase of C is close to the rotor N pole, that is, the rotor N pole falls in sector V and the rotor S pole falls in sector II.

[0105] Furthermore, in the third injection phase, phase A and phase B are compared based on the reverse third pulse voltage. If the phase current amplitude of phase A is greater than that of phase B, it indicates that the rotor is biased towards the side position corresponding to A within the corresponding sector. That is, the rotor's N pole falls in sector V-a, and the rotor's S pole falls in sector II-a. Based on this, the traditional 6 basic sectors are refined into 12 basic sectors to improve the resolution of angle positioning.

[0106] In one exemplary embodiment, Figure 3 A structural block diagram of a direct-axis pulse voltage injection process is shown: First, the sector containing the third target phase is divided into several discrete angles according to a preset angle step size, and each discrete angle is used as a virtual angle for injecting direct-axis pulse voltage in the dq virtual rotating coordinate system. The direct-axis component of the direct-axis pulse voltage Cross-axis components .

[0107] From various virtual perspectives After injecting a direct-axis pulse voltage, the direct-axis voltage component and the quadrature-axis voltage component of its response in the dq virtual rotating coordinate system are subjected to an inverse Park transformation to obtain each virtual angle. The α-axis voltage components and β-axis voltage components in the α-β two-phase stationary coordinate system, respectively.

[0108] Next, for each virtual angle The α-axis voltage component and the β-axis voltage component are subjected to space vector pulse modulation (SVPWM) to obtain the duty cycle data for preset three-phase inverter.

[0109] Furthermore, based on various virtual perspectives The corresponding duty cycle data drive the three-phase inverter to output three-phase current to the motor windings. , , Then, the Clarke transform is performed on the three-phase currents to obtain the α-axis current components in the α-β two-phase stationary coordinate system. With β-axis current component Furthermore, the virtual angle is placed in the real rotating coordinate system of dq. As the angle to be measured θ, the current component along the α-axis With β-axis current component Perform Park transformation to obtain various virtual angles. Direct-axis current components in the real rotating coordinate system dq With cross-axis current component .

[0110] Finally, the various virtual angles The corresponding direct-axis current components By comparing these values, the rotor position is calculated, thereby identifying the maximum direct-axis current component. Corresponding virtual angle As the actual electrical angle of the rotor .

[0111] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages of other steps.

[0112] Based on the same inventive concept, this application also provides an intelligent improved permanent magnet synchronous motor position prediction system for implementing the aforementioned intelligent improved permanent magnet synchronous motor position prediction method. The solution provided by this system is similar to the implementation scheme described in the above method. Therefore, the specific limitations of one or more intelligent improved permanent magnet synchronous motor position prediction system embodiments provided below can be found in the limitations of the intelligent improved permanent magnet synchronous motor position prediction method described above, and will not be repeated here.

[0113] In one exemplary embodiment, such as Figure 4 As shown, an intelligent improved permanent magnet synchronous motor position prediction system is provided, including: a first positioning module 101, a second positioning module 102, a polarity prediction module 103, and an angle prediction module 104, wherein: The first positioning module 101 is used to inject a first pulse voltage into a preset injection phase, and by comparing the phase currents of each phase based on the first pulse voltage, determine the first target phase that the rotor is closest to among the non-injection phases in the first injection stage. The second positioning module 102 is used to inject a second pulse voltage into the first target phase, and by comparing the phase currents of each phase based on the second pulse voltage, determine the closest second target phase among the non-injected phases of the rotor in the second injection stage, and determine the closest third target phase of the rotor between the first target phase and the second target phase. The polarity prediction module 103 is used to inject a reverse third pulse voltage into the third target phase and determine the polarity position of the rotor by comparing the phase currents that the third target phase responds to at each injection stage. The angle prediction module 104 is used to divide the sector corresponding to the third target into multiple angles according to a preset angle step size, inject direct-axis pulse voltage into each angle, and determine the angular position of the rotor by comparing the direct-axis currents that respond to the direct-axis pulse voltages at each angle.

[0114] Each module in the aforementioned intelligent improved permanent magnet synchronous motor position prediction system can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of a computer device in hardware form or independent of it, or stored in the memory of a computer device in software form, so that the processor can call and execute the corresponding operations of each module.

[0115] In one exemplary embodiment, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of any of the above embodiments.

[0116] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the above embodiments.

[0117] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above methods.

[0118] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0119] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A smart improved permanent magnet synchronous motor position prediction method, characterized in that, The method comprises: injecting a first pulse voltage in a preset injection phase, determining a first target phase closest to the rotor among non-injection phases in a first injection stage by comparing phase currents of each phase respectively based on responses of the first pulse voltage; injecting a second pulse voltage in the first target phase, determining a second target phase closest to the rotor among non-injection phases in a second injection stage by comparing phase currents of each phase respectively based on responses of the second pulse voltage, and determining a third target phase closest to the rotor among the first target phase and the second target phase; injecting a third pulse voltage in reverse in the third target phase, and determining a polarity position of the rotor by comparing phase currents of the third target phase respectively in each injection stage; dividing a sector corresponding to the third target phase into multiple angles according to a preset angle step, injecting a direct-axis pulse voltage in each angle, and determining an angle position of the rotor by comparing direct-axis currents of each angle respectively based on responses of the direct-axis pulse voltage.

2. The method of claim 1, wherein, The method comprises: injecting a first pulse voltage in a preset injection phase, comparing phase currents of each non-injection phase respectively based on responses of the first pulse voltage in a first injection stage; if an absolute difference of the phase currents between each non-injection phase is greater than a preset threshold, taking a non-injection phase corresponding to a maximum phase current as a first target phase; if the absolute difference of the phase currents between each non-injection phase is less than or equal to the preset threshold, determining a new injection phase among each non-injection phase and injecting a preset pulse voltage, and determining a first target phase according to phase currents of each phase respectively based on responses of the pulse voltage in a current injection stage.

3. The method of claim 1, wherein, The method comprises: injecting a second pulse voltage in the first target phase, comparing phase currents of each non-injection phase respectively based on responses of the second pulse voltage in a second injection stage; if an absolute difference of the phase currents between each non-injection phase is greater than a preset threshold, taking a non-injection phase corresponding to a maximum phase current as a second target phase; if the absolute difference of the phase currents between each non-injection phase is less than or equal to the preset threshold, taking the first target phase as the second target phase.

4. The method of claim 1, wherein, The method comprises: aligning and comparing the phase current corresponding to the first target phase in the first injection stage and the phase current corresponding to the second target phase in the second injection stage according to a phase relationship of voltage vectors corresponding to each injection stage. if the amplitude of the phase current corresponding to the second target phase in the second injection stage is greater than the amplitude of the phase current corresponding to the first target phase in the first injection stage, the second target phase is taken as the third target phase; if the amplitude of the phase current corresponding to the first target phase in the first injection stage is greater than or equal to the amplitude of the phase current corresponding to the second target phase in the second injection stage, the first target phase is taken as the third target phase.

5. The method of claim 1, wherein, the injecting a third pulse voltage in the third target phase, and determining the polarity position of the rotor by comparing the phase currents of the third target phase in response to each injection stage, comprises: injecting a third pulse voltage in the third target phase, and comparing the phase currents of the third target phase in response to each injection stage; if the third target phase is the injection phase of two of the injection stages, comparing the absolute difference of the phase currents of the third target phase in response to the two injection stages to determine the polarity position of the rotor; if the third target phase is the injection phase of one of the injection stages or is not the injection phase of any injection stage, injecting a third pulse voltage in the third target phase, and comparing the absolute difference of the phase currents of the third target phase in response to the reverse injection stage and the forward injection stage to determine the polarity position of the rotor.

6. The method of claim 1, wherein, after the injecting a third pulse voltage in the third target phase, the method further comprises: determining the closest non-injection phase of the rotor in the third injection stage based on the phase currents of each non-injection phase in response to the third pulse voltage, to determine that the rotor is in a left position or a right position in the sector corresponding to the third target phase, and to refine the sector from a 60-degree range to a 30-degree range.

7. The method of claim 1, wherein, the injecting a direct-axis pulse voltage in each angle, and determining the angle position of the rotor by comparing the direct-axis currents of each angle in response to the direct-axis pulse voltage, comprises: injecting a direct-axis pulse voltage in each angle, and performing inverse Park transformation on the voltage of each angle in a preset virtual rotating coordinate system based on the direct-axis pulse voltage to obtain the voltage of each angle in a preset two-phase stationary coordinate system; performing space vector pulse modulation on the voltage of each angle in the two-phase stationary coordinate system to obtain duty cycle data for driving a preset three-phase inverter; driving the three-phase inverter to output a three-phase current to a preset motor winding according to the duty cycle data, and performing Clarke transformation and Park transformation on the three-phase current to obtain the direct-axis current of each angle in a preset real rotating coordinate system, and taking the angle corresponding to the maximum direct-axis current as the angle position of the rotor.

8. An intelligent improved permanent magnet synchronous motor position prediction system, characterized in that, the system comprises: The first positioning module is configured to inject a first pulse voltage in a preset injection phase, and determine a first target phase closest to the rotor among non-injection phases in a first injection stage by comparing phase currents respectively responded by each phase based on the first pulse voltage. The second positioning module is configured to inject a second pulse voltage in the first target phase, and determine a second target phase closest to the rotor among non-injection phases in a second injection stage by comparing phase currents respectively responded by each phase based on the second pulse voltage, and determine a third target phase closest to the rotor among the first target phase and the second target phase. The polarity prediction module is configured to inject a third pulse voltage in reverse in the third target phase, and determine a polarity position of the rotor by comparing phase currents respectively responded by the third target phase in each injection stage. The angle prediction module is configured to divide a sector corresponding to the third target phase into multiple angles according to a preset angle step, inject a direct-axis pulse voltage in each angle, and determine an angle position of the rotor by comparing direct-axis currents respectively responded by each angle based on the direct-axis pulse voltage. 9.A computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the computer device is configured to perform the method according to any one of claims 1-8 when the computer program is executed by the processor. The processor executes the computer program to implement the steps of the method in any one of claims 1 to 7.

10. A computer-readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the method in any one of claims 1 to 7.

Images