Control method and control apparatus for permanent magnet synchronous motor

Abstract

The present invention relates to a control method and control apparatus for a permanent magnet synchronous motor. The control method comprises: a compensation measurement step, which comprises: by means of an experimental test performed in advance, measuring and calculating a plurality of cogging torques of a rotor at a plurality of rotation positions within a rotation cycle, and generating a lookup table that represents mapping relationships between the plurality of rotation positions and the corresponding plurality of cogging torques; and a compensation application step, which comprises when a permanent magnet synchronous motor is operating, on the basis of the lookup table generated in the compensation measurement step, determining a cogging torque corresponding to the current rotation position of the rotor, and applying a corresponding compensation signal to a control circuit of the permanent magnet synchronous motor. The control method and control apparatus in the present invention can improve the effect of cogging torque compensation.

Description

Control method and control device of permanent magnet synchronous motor TECHNICAL FIELD

[0001] The present application relates to the technical field of electric machines. In particular, the present application relates to a control method and a control device of a permanent magnet synchronous motor. BACKGROUND

[0002] A permanent magnet synchronous motor (PMSM) is a device for converting electrical energy into kinetic energy or vice versa. A permanent magnet synchronous motor mainly comprises a stator and a rotor. The radially inner side of the stator is formed with stator teeth and stator slots alternatingly distributed in the circumferential direction, and the coil winding of the stator is wound on the stator teeth. The rotor is installed on the radially inner side of the stator and comprises a permanent magnet. The coil winding of the stator can generate a rotating magnetic field when a three-phase alternating current is applied, thereby driving the rotor to rotate relative to the stator. Due to the magnetic interaction between the material of the stator core and the permanent magnet of the rotor, there is an interaction force between the stator and the rotor even in the case of non-power supply. Influenced by the physical structure of the motor such as the tooth and slot distribution of the inner surface of the stator and the magnetic tile distribution of the outer surface of the rotor, this force usually has a non-zero tangential component, thereby generating an additional torque acting on the rotor, i.e. a cogging torque. Generally, the cogging torque will present a periodic change in each rotation period of the rotor.

[0003] In order to accurately control the operation of the motor, it is necessary to compensate the cogging torque in the control signal of the motor. There are many different control methods for compensating the cogging torque in the prior art. For example, the control method disclosed in CN 115842491 A compensates the cogging torque by harmonic current injection based on the transfer characteristics of the control system, and the control algorithm of this method is very complex, requiring a large amount of algorithm resource of the controller. For another example, the control method disclosed in CN 112994539 A suppresses the cogging torque fluctuation of the permanent magnet synchronous motor based on theoretical analysis and calculation, and this method has very high requirements for the accuracy of the physical parameters of the motor. For another example, the control method disclosed in CN 115800847 A pre-compensates the cogging torque fluctuation by adjusting the PWM (pulse width modulation) duty cycle, and this method has very high requirements for the sampling accuracy of the sensor. Due to the above-mentioned high requirements of the prior art for the sampling and calculation process, these methods are difficult to be well applied to the engineering practice of the motor.

[0004] SUMMARY

[0005] Therefore, the technical problem to be solved by the present application is to provide a method for compensating the cogging torque of a permanent magnet synchronous motor.

[0006] The above technical problem is solved by a control method of a permanent magnet synchronous motor according to the present application. The control method is used to compensate the cogging torque of the permanent magnet synchronous motor, which comprises a stator and a rotor. The control method comprises: a compensation measurement step, in which a plurality of cogging torques of the rotor at a plurality of rotational positions in a rotational cycle are measured and calculated through a pre-experimental test, and a look-up table representing a mapping relationship between the plurality of rotational positions and the corresponding plurality of cogging torques is generated; and a compensation application step, in which when the permanent magnet synchronous motor is running, the cogging torque corresponding to the current rotational position of the rotor is determined according to the look-up table generated in the compensation measurement step, and a corresponding compensation signal is applied in the control circuit of the permanent magnet synchronous motor. By determining the cogging torque through the look-up table generated in the test in advance, the program processing time in running can be saved, the data processing capacity requirement of the processor is reduced, thereby improving the practicability of the control method and reducing the control cost.

[0007] According to a preferred embodiment of the present application, the compensation measurement step can comprise: a data measurement sub-step, in which a plurality of driving torques are measured when the rotor of the permanent magnet synchronous motor is driven by an external driving torque to rotate at a constant speed; and a component separation sub-step, in which a time delay compensation is applied to the plurality of rotational positions, a plurality of cogging torques as fluctuation components in the plurality of driving torques are obtained through digital filtering processing, and the plurality of rotational positions after time delay compensation and the plurality of cogging torques are used to generate the look-up table based on the mapping relationship. By the digital filtering method, the cogging torque can be conveniently and effectively separated from the total torque measured.

[0008] According to another preferred embodiment of the present application, in the data measurement sub-step, for each of the m permanent magnet synchronous motors of the same specification, the plurality of rotational positions of the corresponding rotor are measured n times respectively, thereby obtaining m×n groups of the plurality of driving torques, wherein m and n are each an integer greater than or equal to 1; and the compensation measurement step further comprises a mean value calculation sub-step executed between the data measurement sub-step and the component separation sub-step, in which for each of the plurality of rotational positions, the average value of the n measurements of the m permanent magnet synchronous motors in the data measurement sub-step is calculated as the final measurement value of the corresponding driving torque, and in the component separation sub-step, the look-up table suitable for each of the m permanent magnet synchronous motors is generated based on the final measurement values of the plurality of driving torques. By the multiple measurements for the plurality of permanent magnet synchronous motors, the accuracy of the measurement results can be improved.

[0009] According to another preferred embodiment of the present application, in the data measuring substep, the m same-specification permanent magnet synchronous motors can be selected from a plurality of permanent magnet synchronous motors produced in the same batch, and in the component separating substep, a lookup table applicable to each of the plurality of permanent magnet synchronous motors produced in the same batch can be generated, wherein m is less than or equal to the number of the plurality of permanent magnet synchronous motors produced in the same batch. By virtue of the structural consistency of the batch-produced motors, data of only part of the motors can be measured, thereby reducing the workload of measurement.

[0010] According to another preferred embodiment of the present application, in the component separating substep, the direct current component in each of the applied driving torques can be obtained first by digital low-pass filtering processing, and then the corresponding fluctuation component can be obtained by subtracting the corresponding direct current component from each of the applied driving torques. The calculation result of the digital low-pass filtering separation has higher accuracy.

[0011] According to another preferred embodiment of the present application, in the compensation applying step, for the rotational positions of the rotor that are not present in the lookup table, an interpolation method can be used to calculate the corresponding cogging torque based on the data in the lookup table. In this way, the control of the complete rotational period of the rotor can be realized based on limited measurement data.

[0012] According to another preferred embodiment of the present application, in the compensation measuring step, the plurality of rotational positions can be distributed in the same minimum cogging torque period, and each minimum cogging torque period corresponds to 1 / p of the rotational period of the rotor, wherein p is the number of pole pairs of the rotor. In this way, the workload of measurement can be reduced.

[0013] According to another preferred embodiment of the present application, the control method can further include an initial correction step performed before the compensation measuring step, in which the zero-degree mechanical angle position of the rotor is corrected. In this way, the accuracy of the sensor measurement result is ensured.

[0014] According to another preferred embodiment of the present application, the compensation measuring step can be performed before the permanent magnet synchronous motor is shipped. In this way, the permanent magnet synchronous motor can accurately compensate the cogging torque when it is put into use.

[0015] The technical problem is also solved by a control device of a permanent magnet synchronous motor according to the present application. The permanent magnet synchronous motor comprises a stator and a rotor, and the control device comprises: a compensation measurement module configured to measure and calculate a plurality of cogging torques of the rotor at a plurality of rotational positions in a rotational cycle through pre-experimental tests, and generate a lookup table representing a mapping relationship between the plurality of rotational positions and the corresponding plurality of cogging torques; and a compensation application module configured to determine a cogging torque corresponding to a current rotational position of the rotor according to the lookup table generated by the compensation measurement module when the permanent magnet synchronous motor is running, and apply a corresponding compensation signal in a control circuit of the permanent magnet synchronous motor. By determining the cogging torque through the lookup table generated in advance in the test, the program processing time in the running can be saved, the data processing capacity requirement of the processor is reduced, thereby improving the practicability of the control device and reducing the control cost.

[0016] According to a preferred embodiment of the present application, the compensation measurement module can comprise: a data measurement submodule configured to measure a plurality of applied driving torques of the rotor at a plurality of rotational positions when the rotor of the permanent magnet synchronous motor is driven to rotate at a constant speed by an applied driving torque; and a component separation submodule configured to apply time delay compensation to the plurality of rotational positions, obtain a plurality of cogging torques as fluctuation components in the plurality of applied driving torques through digital filtering processing, and generate a lookup table based on the mapping relationship between the plurality of rotational positions after time delay compensation and the plurality of cogging torques. By the digital filtering method, the cogging torque can be conveniently and effectively separated from the measured total torque.

[0017] According to another preferred embodiment of the present application, the data measurement submodule can be further configured to measure the plurality of rotational positions of the corresponding rotor n times for each of the m permanent magnet synchronous motors of the same specification, thereby obtaining m×n groups of the plurality of applied driving torques, wherein m and n are each an integer greater than or equal to 1; and the compensation measurement module can further comprise a mean value calculation submodule configured to calculate, for each of the plurality of rotational positions, an average value of n measurements of the data measurement submodule for the m permanent magnet synchronous motors as a final measurement value of the corresponding applied driving torque, and the component separation submodule is further configured to generate a lookup table applicable to each of the m permanent magnet synchronous motors based on the final measurement values of the plurality of applied driving torques. By multiple measurements for a plurality of permanent magnet synchronous motors, the accuracy of the measurement results can be improved.

[0018] According to another preferred embodiment of the present application, the m permanent magnet synchronous motors of the same specification measured by the data measurement sub-module can be selected from a plurality of permanent magnet synchronous motors produced in the same batch, and the component separation sub-module can be further configured to generate a lookup table applicable to each of the plurality of permanent magnet synchronous motors produced in the same batch, wherein m is less than or equal to the number of the plurality of permanent magnet synchronous motors produced in the same batch. By taking advantage of the structural consistency of the batch-produced motors, data of only part of the motors can be measured, thereby reducing the workload of measurement.

[0019] According to another preferred embodiment of the present application, the component separation sub-module can be further configured to first obtain a direct current component in each of the applied driving torques by digital low-pass filtering processing, and then obtain a corresponding fluctuation component by subtracting the corresponding direct current component from each of the applied driving torques. The calculation result of the digital low-pass filtering separation has high accuracy.

[0020] According to another preferred embodiment of the present application, the compensation application module can be further configured to, for a rotating position of the rotor that does not exist in the lookup table, calculate a corresponding cogging torque based on data in the lookup table using an interpolation method. In this way, control over the entire rotation period of the rotor can be achieved based on limited measurement data.

[0021] According to another preferred embodiment of the present application, the compensation measurement module can be configured to distribute the plurality of rotating positions within the same minimum cogging torque period, and each minimum cogging torque period corresponds to 1 / p of the rotation period of the rotor, wherein p is the number of pole pairs of the rotor. In this way, the workload of measurement can be reduced.

[0022] According to another preferred embodiment of the present application, the control device can further include an initial correction module configured to correct the zero-degree mechanical angle position of the rotor before the compensation measurement module measures and calculates the plurality of cogging torques. In this way, the accuracy of the sensor measurement result is ensured.

[0023] According to another preferred embodiment of the present application, the compensation measurement module is further configured to measure and calculate the plurality of cogging torques and generate the lookup table before the permanent magnet synchronous motor is shipped. The permanent magnet synchronous motor can thus accurately compensate for the cogging torque when put into use. BRIEF DESCRIPTION OF DRAWINGS

[0024] The present application will be further described below in conjunction with the accompanying drawings. The same reference numerals are used to represent the same functional elements in the drawings. Among them:

[0025] FIG. 1 shows a schematic diagram of a permanent magnet synchronous motor to which a control method according to an exemplary embodiment of the present application is applied;

[0026] FIG. 2 shows a flowchart of the control method according to an exemplary embodiment of the present application;

[0027] Fig. 3 shows a flow chart of the compensation measurement step of the control method according to an exemplary embodiment of the present application;

[0028] Fig. 4 shows a control logic diagram of the component separation step of the control method according to an exemplary embodiment of the present application;

[0029] Figs. 5a and 5b show a schematic diagram of the time delay calculation of the control method according to an exemplary embodiment of the present application; and

[0030] Fig. 6 shows a control logic diagram of the compensation application step of the control method according to an exemplary embodiment of the present application. DETAILED DESCRIPTION

[0031] The following will describe specific embodiments of a control method and a control device for a permanent magnet synchronous motor according to the present application with reference to the accompanying drawings. The following detailed description and drawings are provided as an example to illustrate the principles of the present application, the present application is not limited to the described preferred embodiments, the scope of protection of the present application is defined by the claims.

[0032] According to an embodiment of the present application, a control method for a permanent magnet synchronous motor and a corresponding control device are provided. Such control method and control device aim at compensating the cogging torque inherent to the permanent magnet synchronous motor.

[0033] First, the theoretical background of cogging torque will be introduced. As already introduced in the background art, in the physics of electrical machines, cogging torque is an inherent physical property of permanent magnet synchronous motors. In order to describe cogging torque from the level of electromagnetic field theory, the Maxwell stress tensor and tension method are introduced to quantitatively estimate the air gap region, as shown in equation (1):

[0034] where L denotes the effective axial rotor length; l g denotes the length of the air gap; μ0denotes the vacuum permeability; B n and B t denote the radial and tangential components of the air gap flux density; S g denotes the surface area of the air gap. The general description of cogging torque is given by equation (2):

[0035] Here the Fourier expansion method is used to derive the sum of the harmonics. Where N cog denotes the fundamental frequency of the cogging torque; n denotes the order of any harmonic; T n denotes the amplitude of the harmonic of order n; θ m denotes the mechanical rotor angle; The phase angle represents the harmonic of order n. Equation (2) is a strict theoretical description of the cogging torque, including an infinite number of harmonic orders. To meet the requirements of practical engineering applications, the harmonic components should be analyzed for order, which can reveal which harmonic components contribute to the main amplitude. Generally, the torque fluctuation caused by the cogging effect is mainly contributed by the fundamental component, which depends on the number of cogging torque cycles per mechanical angle period of the rotor and the motor speed. As the motor speed changes from low to high within the operating speed range, the frequency of the cogging torque fundamental component also changes.

[0036] FIG. 1 shows a schematic diagram of a permanent magnet synchronous motor to which a control method according to an exemplary embodiment of the present application is applied. As shown in FIG. 1, such a permanent magnet synchronous motor has a typical structure of a conventional permanent magnet synchronous motor. Specifically, the permanent magnet synchronous motor mainly includes a stator 10 and a rotor 20. The stator 10 has a substantially ring-shaped structure, and the rotor 20 is coaxially arranged on the radially inner side of the stator 10 and rotates relative to the stator 10 about a common central axis. The stator 10 is formed with a plurality of stator teeth 11 and stator slots 12 alternately distributed in the circumferential direction on the radially inner side. Coil windings (not shown) of the stator 10 are wound around each of the stator teeth 11. When three-phase alternating current is applied to the coil windings of the stator 10, a rotating magnetic field can be generated in the stator 10. The rotor 20 includes one or more permanent magnets 21. When there are a plurality of permanent magnets 21, the permanent magnets 21 are distributed in the circumferential direction, and the north pole N and the south pole S of each permanent magnet 21 are arranged in the same direction in the circumferential direction. In the embodiment shown in FIG. 1, four permanent magnets 21 are shown schematically, but the rotor 20 can also include other numbers of permanent magnets 21, such as two or six, etc.

[0037] The main steps of the control method of the present application will be described below in conjunction with FIGS. 2 and 3. As shown in the flowchart of FIG. 2, the control method mainly includes a compensation measurement step S2 and a compensation application step S3. In addition, the control method can additionally include a preliminary initial correction step S1. In the control device according to the present application, an initial correction module can be correspondingly provided to perform the initial correction step S1. As shown in the flowchart of FIG. 2, when the initial correction step S1 exists, it needs to be performed as the first step first. In the initial correction step S1, the zero-degree mechanical angle position of the rotor 20 of the permanent magnet synchronous motor is corrected.

[0038] In a permanent magnet synchronous motor, the geometric angle of the rotor 20 rotating in the physical space is referred to as the mechanical angle, while the corresponding position of the rotor 20 in a complete magnetic field variation cycle is referred to as the electrical angle. The angle sensor (e.g. a rotary transformer) in the permanent magnet synchronous motor can directly detect the mechanical angle of the rotor 20 and convert the mechanical angle into the electrical angle through a certain correspondence. In other words, the rotating position of the rotor 20 can be directly obtained through the angle sensor. The zero-degree mechanical angle position of the rotor 20 is a specific position depending on the structure of the stator 10. However, due to the assembly process, the installation position of the angle sensor is usually random, thereby often causing a deviation between the predetermined zero-degree mechanical angle position of the angle sensor and the actual zero-degree mechanical angle position. In practice, this deviation needs to be considered and the reading of the angle sensor needs to be corrected in order to ensure that the measured value of the rotating position of the rotor 20 is accurate, thereby ensuring that accurate currents are generated for FOC (field oriented control).

[0039] In motor physics, there are many different methods that can be used to measure and calibrate this deviation. In the initial correction step S1, the reading of the angle sensor can be corrected in the zero-degree mechanical angle position using any method known in the prior art. The present application does not limit the specific method used to correct the zero-degree mechanical angle position. The initial correction step S1 needs to be performed separately for each permanent magnet synchronous motor. Since the initial correction is a process that all motors in the prior art usually need to go through before being shipped or used, this step can not be considered as the core feature of the present application. The compensation measurement step S2 and the compensation application step S3 performed after the initial correction step S1 are the core features of this control method.

[0040] After the initial correction step S1 is performed, the compensation measurement step S2 is performed. In the control device according to the present application, the compensation measurement module can be correspondingly arranged to perform the above-mentioned compensation measurement step S2. In the compensation measurement step S2, a plurality of cogging torques of the rotor 20 at a plurality of rotating positions in a rotating cycle are measured and calculated through a preliminary experimental test, and a lookup table representing the mapping relationship between these rotating positions and the corresponding plurality of cogging torques is generated. In other words, before the permanent magnet synchronous motor is put into normal operation, the cogging torques of the rotor 20 at different positions in a rotating cycle are obtained through an experimental test method, and each cogging torque is stored in a lookup table in a mutually corresponding manner (mapping relationship) with the corresponding rotating position. In this way, when the motor is in operation, the corresponding cogging torque can be found in the pre-stored lookup table according to the rotating position of the rotor 20, and the corresponding compensation torque is provided. Since the measurement process is performed in advance, during the operation of the motor, only the corresponding cogging torque needs to be found in the lookup table, without the need to calculate the cogging torque, so the processing during the operation of the motor can be significantly simplified and the requirement for data processing capacity can be reduced.

[0041] In the compensation measurement step S2, the process of measuring the cogging torque by experimental test can be implemented in various ways. The compensation measurement step S2 will be described in detail below in connection with FIG. 3. In a preferred embodiment, as shown in the flowchart of FIG. 3, the compensation measurement step S2 can specifically include a data measurement sub-step S21 and a component separation sub-step S23.

[0042] The data measurement sub-step S21 aims to collect the original torque data of the motor, and can be performed by a data measurement sub-module in the compensation measurement module. Specifically, in the data measurement sub-step S21, when the rotor 20 of the permanent magnet synchronous motor is driven by an external driving torque to rotate at a predetermined speed, a plurality of external driving torques at a plurality of predetermined rotational positions of the rotor 20 are measured. The torque results measured at each rotational position of the rotor 20 are recorded in the memory for further processing. During the measurement process of step S21, the measured permanent magnet synchronous motor is driven by an external driving torque to passively rotate rather than actively rotate in the measurement system, in which case the torque balance equation under uniform rotation conditions can be expressed as formula (3):

[0043] T dr -T fri -T damp -T cog (θ m )=0 (3)

[0044] wherein T dr represents the external driving torque (e.g. from an external driving motor) applied to the target permanent magnet synchronous motor, T fri represents the friction torque of the target permanent magnet synchronous motor, T damp represents the damping torque of the target permanent magnet synchronous motor, θ m represents the mechanical angular position of the rotor 20, T cog (θ m ) represents the cogging torque of the target permanent magnet synchronous motor, which varies periodically with the rotational position θ m of the rotor 20.

[0045] In actual measurement, the torque balance equation can be further expressed as formula (4):

[0046] T′ dr -T err =T fri +T damp +T cog (θ m ) (4)

[0047] wherein T′ dr represents T drmeasured by a torque sensor installed at the connection between the motor shaft of the target permanent magnet synchronous motor and the transmission shaft for inputting the applied driving torque (for example, the motor shaft of an external driving motor); T err represents the inherent measurement error of the torque sensor; T' dr -T err corresponds to the actual value of the applied driving torque T dr . The sum of the three torque parameters T fri , T damp , and T cog (θ m ) corresponds to the actual value of the applied driving torque T dr , which can be referred to as the total fluctuation torque. In the total fluctuation torque, the cogging torque T cog (θ m ) is referred to as the fluctuation component or alternating component due to periodic changes, and the sum of T fri and T damp is referred to as the direct current component. Considering that there are three physical quantities T' dr , T err , and θ m that can be directly measured in formula (4), and there are also three physical quantities T fri , T damp , and T cog (θ m ) that cannot be directly measured, the control method of the present application solves the fluctuation component of the cogging torque based on the three directly measured physical quantities.

[0048] In motor physics, there are two physical quantities related to the rotor position, i.e., the mechanical angle (mechanical angular position) θ m of the rotor 20 and the electrical angle θ e of the rotor. The mechanical angle refers to the angular position of the rotor 20 measured by the angle sensor after offset compensation in one rotation period of the rotor 20 in the physical space. The electrical angle of the rotor 20 refers to dividing the mechanical angle period of the rotor into several sub-periods according to the variation period of the magnetic flux generated in the stator coil by the rotating magnetic field of the rotor permanent magnet 21. Therefore, for a permanent magnet synchronous motor with p pairs of rotor magnetic poles, the relationship between the electrical angle and the mechanical angle can be represented by formula (5):

[0049] θ e = mod(p·θ m , 2π) (5)

[0050] where θ m represents the mechanical angle of the rotor 20; θ erepresents the electrical angle of the rotor 20; p represents the number of pole pairs of the rotor 20; mod(x, y) represents a function of taking the remainder of x with respect to y.

[0051] For example, in the exemplary permanent magnet synchronous motor shown in Fig. 1, the number of pole pairs p of the rotor 20 is 4, and the number of slots s of the stator 10 is 12. As can be clearly seen from Fig. 1, the relative positional relationship of the physical structures of the stator 10 and the rotor 20 repeats once every time the rotor 20 rotates through 1 / 4 of a circle. This means that for every 1 / 4 of a circle of the rotation process of the rotor 20, the interaction state between the teeth and slots of the stator 10 and the permanent magnets 21 of the rotor 20 repeats, and thus the cogging torque generated also repeats. Therefore, it is possible to focus only on the cogging torque in 1 / p of a circle of the rotation process of the rotor. At the same time, according to electromagnetism, the number of periods of the electrical angle contained in each complete mechanical angle period (rotation period of the rotor) of the rotor is equal to the number of pole pairs p of the rotor. It can be concluded that one complete minimum cogging torque period corresponds to exactly one electrical angle period, and the minimum cogging torque period is 1 / p of a mechanical angle period of the rotor 20. Based on this theory, in the preferred embodiment, in the data measurement sub-step S21 of the compensation measurement step S2, the selected plurality of rotation positions can be distributed within the same minimum cogging torque period.

[0052] As mentioned before, in the measurement process of the data measurement sub-step S21, the rotor 20 is driven to rotate at a predetermined speed by an externally applied driving torque. The predetermined speed is not specifically limited here. However, the lower the speed of the permanent magnet synchronous motor during this process, the more accurate the measured cogging torque will be. In electromagnetism, there is a well-known fundamental model equation for representing the voltage, as shown in equation (6):

[0053] where Ldrepresents the inductance on the d-axis (direct axis); Lqrepresents the inductance on the q-axis (quadrature axis); R represents the resistance of each phase; Idrepresents the current on the d-axis; Iqrepresents the current on the q-axis; ω represents the mechanical speed of the rotor; ψrepresents the magnetic flux generated by the rotor 20. The magnetic flux generated by the rotor 20 is constant after the initial assembly of the motor, and depends on the permanent magnets 21 distributed on the outer surface of the rotor 20. It can be concluded that: d q d q r According to equation (6), when the permanent magnet synchronous motor rotates at a high speed without being powered, since there is a fixed physical quantity, the magnetic flux ψ r ​​​​​, strong counter electromotive force will be generated in the stator coil, which makes the magnetic field around the stator teeth 11 and the stator slots 12 more complex and severely distorted. This will result in superimposing more additional complex fluctuation torque on the cogging torque. Therefore, it is preferred to choose a lower rotational speed to drive the rotor 20 to measure the physical quantities T' dr , T err and θ m . The specific rotational speed value can be determined according to practical experience.

[0054] In the preferred embodiment, for each predetermined rotation position to be measured, the measurement can be performed only once or repeated multiple times; at the same time, the measurement can be performed only for a single permanent magnet synchronous motor or a plurality of permanent magnet synchronous motors of the same specification. If each rotation position to be measured is measured only once for a single permanent magnet synchronous motor, the plurality of measurement results respectively corresponding to the plurality of rotation positions can be directly provided to the component separation step S23 for subsequent processing. If the measurement is repeated multiple times for each rotation position to be measured and / or is performed for a plurality of permanent magnet synchronous motors of the same specification, the average value for each rotation position needs to be calculated first, and then the average value is provided as the measurement result to the subsequent component separation step S23.

[0055] The above process can be summarized as follows: in the data measurement sub-step S21, for each of the m permanent magnet synchronous motors of the same specification, n measurements are performed for the plurality of rotation positions of the corresponding rotor 20, thereby obtaining m x n groups of applied driving torques, each group containing a plurality of applied driving torques corresponding to the aforementioned plurality of rotation positions respectively. Here, m and n are each an integer greater than or equal to 1. At the same time, the compensation measurement step S2 further includes a mean value calculation sub-step S22 performed between the data measurement sub-step S21 and the component separation sub-step S23, which can be performed by a mean value calculation sub-module in the compensation measurement module. In the mean value calculation sub-step S22, for each rotation position, the average value of the n measurements of the aforementioned m permanent magnet synchronous motors is calculated as the final measurement value of the corresponding applied driving torque. This final measurement value will be applied to the subsequent component separation sub-step S23, in which a lookup table is generated based on the final measurement values of the applied driving torques. The lookup table is a common lookup table applicable to each of the aforementioned m permanent magnet synchronous motors.

[0056] The specific averaging process is as follows. First, assume that m permanent magnet synchronous motors of the same specification are selected, and each permanent magnet synchronous motor is repeatedly measured n times. Thus, for each directly measured physical quantity, there will be n x m data records of measurement results. The measurement results can be expressed as: T'dr[i][j][k] represents the kth sample value of the total fluctuation torque measured in the jth measurement process of the ith motor; θm[i][j][k] represents the kth sample value of the rotor rotation position measured in the jth measurement process of the ith motor; T err[k] represents the kth sample value of the fixed measurement error of the torque sensor. In the mean calculation sub-step S22, for the total fluctuation torque T'dr[i][j][k] measured by the torque sensor, the average value of each measurement point is calculated as shown in equation (7):

[0057] According to equation (7), equation (4) can be converted to equation (8):

[0058] wherein, is referred to as the total fluctuation torque compensated for measurement error.

[0059] In addition, according to equation (5), equation (8) can be converted to equation (9):

[0060] Thus, by separating the direct current component from the total fluctuation torque compensated for measurement error, the fluctuation component of the cogging torque can be easily obtained from the total fluctuation torque compensated for measurement error, as shown in equation (10):

[0061] In equation (10), it can be found that there are three directly measured physical quantities, the original torque signal contains the total fluctuation torque on the motor shaft, the torque measurement error or offset of the measurement system, and the angular position of the rotor 20. Therefore, in order to facilitate engineering application and achieve high compensation performance, a solution is proposed to construct a compensation lookup table from the rotor electrical angle θ e and the total fluctuation torque compensated for measurement error compensation .

[0062] It should be noted that the mean calculation sub-step S22 actually exists only when at least one of m and n is greater than 1; when m and n are both 1, the average value does not need to be actually calculated, and therefore the mean calculation sub-step S22 does not need to be actually executed.

[0063] Since the motor production and assembly process is usually batch production, the physical characteristics of the motors produced in the same batch are usually consistent. Therefore, in the data measurement sub-step S21, m permanent magnet synchronous motors of the same specification can be preferably selected from the permanent magnet synchronous motors produced in the same batch. The lookup table generated in the component separation sub-step S23 can thus be applicable to each of the permanent magnet synchronous motors in the batch. Among them, m is less than or equal to the number of permanent magnet synchronous motors in the batch. It is particularly preferred that m can be less than the number of permanent magnet synchronous motors in the batch, that is, since the physical characteristics of the permanent magnet synchronous motors produced in the same batch are usually substantially the same, only a part of the motors can be measured without having to measure all the motors.

[0064] In the component separation sub-step S23, time delay compensation is applied to the aforementioned plurality of rotation positions, and a plurality of cogging torques as fluctuation components in the applied driving torques are obtained by digitally filtering the total fluctuation torque obtained in the step S22, and the plurality of rotation positions after time delay compensation and the plurality of cogging torques are used to generate the required lookup table based on the mapping relationship. The component separation sub-step S23 can be performed by a component separation sub-module in the compensation measurement module.

[0065] In the process of implementing the FOC (Field Oriented Control) function, there is a lag phenomenon in the rotation position of the rotor 20, as shown in FIG. 5a, where k-1 represents the previous planning period sequence, k represents the current planning period sequence, k+1 represents the next planning period sequence, T s represents the duration of a single planning period, θ represents the rotation position of the rotor, U s represents the target voltage vector. According to FIG. 5a, when the planning period sequence k starts, the motor torque generated at this time is actually the target voltage vector U s (k-1) in the previous planning period sequence k-1. The reason for this is that after the target voltage vector U s (k-1) is calculated, it will be transmitted to the hardware-related basic software to drive the circuit to generate the actual three-phase voltage according to the target voltage vector U s (k-1). This process takes time, resulting in control delay. Therefore, the current target voltage vector U s (k) is generated by the current rotation position θ(k), but when the current target voltage vector U s (k) is applied and takes effect in the torque value, the actual rotation position of the rotor 20 has reached the next planning period θ'(k+1). Obviously, there is a delay in the control of the rotor rotation position. Therefore, the rotor rotation position lag can be represented as formula (11):

[0066] Δθ = θ'(k + 1) - θ(k) (11)

[0067] In motor physics, when the motor is running, due to the delay Δθ of the rotor rotation position, the d-q axis reference frame based on the rotor position when the control effect is generated in the torque level is also delayed, and is different from the reference frame based on the original sampled rotor rotation position. This deviation will cause the magnetic field direction to be inaccurate, and thus the expected actual d-q axis current will also deviate from the physical true current. Therefore, for motor cogging torque compensation, at the moment when the rotor rotation position is just at the directly sampled position, it is inaccurate to directly incorporate the assumed torque compensation into the calculation of the final output torque. If this is ignored, the compensation effect will be poor.

[0068] There are various methods for estimating the time delay compensation. An exemplary method is given below. In motor physics, when the target d-q axis current is i q = 0, i d = 0, according to formula (6), the target voltage vector U s at this time is equal to U q in the d-q axis reference frame. However, once there is a time delay in the rotor rotation angle Δθ, as shown in Figure 5b, the components of the actual voltage vector and in the d-q axis reference frame are no longer zero, because and the time delay rotor angle rotation Δθ can be estimated by formula (12):

[0069] The above estimation method is only exemplary, and various known estimation methods in the prior art can be used to estimate the time delay Δθ. The estimated time delay of the rotation position Δθ should be incorporated into the compensation calculation of the current rotor position in real time. This can ensure that the final rotor position used to calculate the target voltage is highly close to the rotor position when the calculated target voltage takes effect in the torque value. After time delay compensation of the rotor position, the problem of delay in the effectiveness of cogging torque compensation is solved, and then the current compensation for the cogging torque fluctuation component can be calculated based on the earlier rotor position.

[0070] In addition, as shown in Figure 4, in order to separate the fluctuation component of the cogging torque from the original fluctuating torque collected on the motor output shaft, a digital bandpass filter is used to digitally filter the previously obtained total fluctuating torque. Preferably, the direct current component in each applied driving torque can be first obtained by digital low-pass filtering, and then the corresponding fluctuation component can be obtained by subtracting the corresponding direct current component from each applied driving torque. Specifically, a typical first-order RC circuit low-pass filter is discretized, and then a discretized digital low-pass filter is further designed, as shown in formula (13):

[0071] where y n represents the filter output of the current period; y n-1 represents the filter output of the previous period; x n represents the filter input of the current period; T s represents the sampling period; τ represents the filter time constant; f cog represents the fluctuation frequency of the cogging torque. The filter time constant τ can be estimated by the target cut-off frequency through equation (14):

[0072] For a permanent magnet synchronous motor with a representative operating speed of n r , a number of magnetic pole pairs of p, and a number of stator slots of s, the fluctuation frequency of the cogging torque can be calculated according to equation (15) at a typical operating speed:

[0073] where lcm(x, y) represents the calculation of the least common multiple of x and y.

[0074] For different motor speeds in a wide range, the time constant τ corresponding to the target cut-off frequency should be set according to different f cog . Then the time constant τ can be used as a key parameter in the digital low-pass filter for a wide range of motor speeds. The range of motor speeds applied in this scheme should be corrected according to the actual implementation effect of the scheme.

[0075] According to practical experience, the result of separating the fluctuation component and the direct current component by the digital low-pass filter is more accurate, so the digital low-pass filter processing is the preferred filtering method. However, a digital high-pass filter can also be used to directly filter the fluctuation component from the total fluctuation torque.

[0076] In the embodiments of the present application, both the initial correction step S1 and the compensation measurement step S2 can be performed before the permanent magnet synchronous motor is shipped. In this way, the above lookup table can be obtained before the permanent magnet synchronous motor is put into operation and stored in the controller.

[0077] The compensation applying step S3 is performed when the permanent magnet synchronous motor is running, and can be performed by a compensation applying module in the control device. In the compensation applying step S3, when the permanent magnet synchronous motor is running, the cogging torque corresponding to the current rotational position of the rotor 20 is determined according to the look-up table generated in the compensation measuring step S2, and a corresponding compensation signal is applied in the control circuit of the permanent magnet synchronous motor. This process is performed in real time during the running of the motor. The compensation signal is usually applied as a compensation current corresponding to the cogging torque in the current control loop. Preferably, in the compensation applying step S3, for the rotational position of the rotor 20 that does not exist in the look-up table, an interpolation method can be used to calculate the corresponding cogging torque based on the data in the look-up table.

[0078] Specifically, after obtaining the compensation d-q axis current, the compensation d-q axis current is introduced into the current control loop. After introducing the motor cogging torque compensation signal, the FOC flowchart of the permanent magnet synchronous motor is shown in FIG. 6. Wherein, I q_Ref is the q-axis current reference value of the control loop; I q_fb is the q-axis current feedback value of the control loop; I d_Ref is the d-axis current reference value of the control loop; I d_fb is the d-axis current feedback value of the control loop; I q * is the q-axis current contribution part caused by the cogging torque fluctuation component; I d * is the d-axis current contribution part caused by the cogging torque fluctuation component. From this step, the cogging torque fluctuation in the current level can be overcome by the compensation signal. Since the cogging torque fluctuation component is solved in real time, the contribution part of the cogging torque fluctuation component on the d-q axis current level can be calculated from the look-up table. Subsequently, when the d-q axis contribution part generated by the cogging torque fluctuation component is introduced into the FOC current control loop, the d-q axis reference current should be adjusted to increase or decrease its value, so as to overcome the fluctuation of the final output torque.

[0079] The control method of the permanent magnet synchronous motor according to the present application obtains the compensation torque for the cogging torque based on the pre-tested look-up table, which can conveniently and quickly obtain the compensation torque when the motor is running, so as to reduce the occupation of the operation resources of the processor on the basis of ensuring good compensation effect. The look-up table used in this control method can be universally applied to the motors with physical consistency, especially all the motors produced in the same batch, thereby reducing the measurement workload. This control method considers the time delay effect when calculating the compensation torque, thereby avoiding the inaccurate compensation result caused by the time delay effect of the rotor position.

[0080] Although the possible embodiments have been described in the above description by illustration, it is not the intention of the applicant to restrict or in any way limit the scope of the appended claims to such specifically described embodiments, but rather, any changes and modifications to the specifically described embodiments should be considered as within the scope of the claims. The same can also be said of the various implementations and embodiments of the inventions. As such, the applicants want to protect fully their right to claim equally broad and equivalent constructions be they generally known in the art and / or or newly invented on the effective date of this application but not specifically shown in the above description. Also, the goals of all patents are to protect new and inventive subject matter, and since the phraseology or structure of a claim does not create, nor it destroy the novelty and nonobviousness of the protected invention, the applicants desire that there be no structural limitations of either the claim or the invention itself merely based on the recitation of a particular phrase. Person skilled in the art should understand that there is a distinction between simply "illustrative" or "exemplary" embodiments, and "preferred" embodiments. As such, no language in the specification should be construed as indicating any non-claimed element as essential to the practice of the inventions.

[0081] Legend of the Figures

[0082] 10 stator

[0083] 11 stator tooth

[0084] 12 stator slot

[0085] 20 rotor

[0086] 21 permanent magnet

[0087] N north pole

[0088] S south pole

Claims

1. A control method of a permanent magnet synchronous motor for compensating for cogging torque of the permanent magnet synchronous motor, the permanent magnet synchronous motor including a stator (10) and a rotor (20), characterized by, The control method comprises: a compensation measurement step (S2), in which a plurality of cogging torques of the rotor (20) at a plurality of rotational positions in a rotational cycle are measured and calculated through a prior experimental test, and a lookup table representing a mapping relationship between the plurality of rotational positions and the corresponding plurality of cogging torques is generated; and a compensation application step (S3), in which, when the permanent magnet synchronous motor is running, a cogging torque corresponding to a current rotational position of the rotor (20) is determined according to the lookup table generated in the compensation measurement step (S2), and a corresponding compensation signal is applied in a control circuit of the permanent magnet synchronous motor.

2. The control method according to claim 1, characterized by, The compensation measurement step (S2) comprises: a data measurement sub-step (S21), in which, when the rotor (20) of the permanent magnet synchronous motor is driven to rotate at a constant speed by an applied driving torque, a plurality of applied driving torques of the rotor (20) at the plurality of rotational positions are measured; and a component separation sub-step (S23), in which, time delay compensation is applied to the plurality of rotational positions, the plurality of cogging torques as fluctuation components in the plurality of applied driving torques are obtained through digital filtering processing, and the plurality of rotational positions after time delay compensation and the plurality of cogging torques generate the lookup table based on a mapping relationship.

3. The control method according to claim 2, characterized by, In the data measurement sub-step (S21), for each of m permanent magnet synchronous motors of the same specification, the plurality of rotational positions of the corresponding rotor (20) are measured n times respectively, thereby obtaining m×n groups of the plurality of applied driving torques, wherein m and n are each an integer greater than or equal to 1; and The compensation measurement step (S2) further comprises a mean value calculation sub-step (S22) executed between the data measurement sub-step (S21) and the component separation sub-step (S23), in which, for each of the plurality of rotational positions, the The average value of n measurements of the m permanent magnet synchronous motors in the data measurement sub-step (S21) is taken as the final measurement value of the corresponding applied driving torque, and in the component separation sub-step (S23), the lookup table suitable for each of the m permanent magnet synchronous motors is generated based on the final measurement value of the plurality of applied driving torques.

4. The control method according to claim 3, characterized by, In the data measurement sub-step (S21), the m permanent magnet synchronous motors of the same specification are selected from a plurality of permanent magnet synchronous motors produced in the same batch, and in the component separation sub-step (S23), the lookup table suitable for each of the plurality of permanent magnet synchronous motors produced in the same batch is generated, wherein m is less than or equal to the number of the plurality of permanent magnet synchronous motors produced in the same batch.

5. The control method according to claim 2, characterized by, In the component separation sub-step (S23), the direct current component in each applied driving torque is first obtained through digital low-pass filtering processing, and then the corresponding fluctuation component is obtained by subtracting the corresponding direct current component from each applied driving torque.

6. The control method according to claim 1, characterized by In the compensation applying step (S3), for the rotational positions of the rotor (20) not existing in the lookup table, an interpolation method is used to calculate the corresponding cogging torque based on the data in the lookup table.

7. The control method according to claim 1, characterized by, In the compensation measuring step (S2), the plurality of rotational positions are distributed in the same minimum cogging torque period, and each minimum cogging torque period corresponds to 1 / p of a rotational period of the rotor (20), where p is the number of pole pairs of the rotor (20).

8. The control method according to claim 1, characterized by, The control method further comprises an initial correction step (S1) performed before the compensation measuring step (S2), in which the zero-degree mechanical angle position of the rotor (20) is corrected.

9. The control method according to any one of claims 1 to 8, characterized by, The compensation measuring step (S2) is performed before the permanent magnet synchronous motor is shipped.

10. A control device of a permanent magnet synchronous motor including a stator (10) and a rotor (20), characterized by, The control device comprises: a compensation measuring module configured to measure and calculate a plurality of cogging torques of the rotor (20) at a plurality of rotational positions in a rotational period through pre-experimental tests, and generate a lookup table representing a mapping relationship between the plurality of rotational positions and the corresponding plurality of cogging torques; and a compensation applying module configured to determine a cogging torque corresponding to a current rotational position of the rotor (20) according to the lookup table generated by the compensation measuring module when the permanent magnet synchronous motor is running, and apply a corresponding compensation signal in a control circuit of the permanent magnet synchronous motor. The compensation measuring module comprises:

11. The control device of claim 10, wherein a data measuring sub-module configured to measure a plurality of applied driving torques of the rotor (20) at the plurality of rotational positions when the rotor (20) of the permanent magnet synchronous motor is driven to rotate at a constant speed by an applied driving torque; and a component separation sub-module configured to apply time delay compensation to the plurality of rotational positions, obtain the plurality of cogging torques as fluctuation components in the plurality of applied driving torques through digital filtering processing, and generate the lookup table based on a mapping relationship between the plurality of rotational positions after time delay compensation and the plurality of cogging torques. In the data measuring sub-module, for each of m permanent magnet synchronous motors of the same specification, n measurements are respectively performed on the plurality of rotational positions of the corresponding rotor (20), thereby obtaining m×n groups of the plurality of applied driving torques, where m and n are each an integer greater than or equal to 1; and 12. The control device of claim 11, wherein, The compensation measuring module further comprises a mean value calculation sub-module configured to, for each of the plurality of rotational positions, calculate an average value of n measurements of the m permanent magnet synchronous motors by the data measuring sub-module as a final measurement value of the corresponding applied driving torque, and the component separation sub-module is further configured to generate the lookup table applicable to each of the m permanent magnet synchronous motors based on the final measurement values of the plurality of applied driving torques. ​ 13. The control device of claim 12, wherein, The m same-specification permanent magnet synchronous motors measured by the data measurement submodule are selected from a plurality of permanent magnet synchronous motors produced in the same batch, and the component separation submodule is further configured to generate the lookup table applicable to each of the plurality of permanent magnet synchronous motors produced in the same batch, wherein m is less than or equal to the number of the plurality of permanent magnet synchronous motors produced in the same batch.

14. The control device of claim 11, wherein, The component separation submodule is further configured to first obtain a direct current component in each of the external driving torques through digital low-pass filtering processing, and then obtain a corresponding fluctuation component by subtracting the corresponding direct current component from each of the external driving torques.

15. The control device of claim 10, wherein The compensation application module is further configured to, for a rotation position of the rotor (20) that does not exist in the lookup table, calculate a corresponding cogging torque based on data in the lookup table using an interpolation method.

16. The control device of claim 10, wherein The compensation measurement module is configured to distribute the plurality of rotation positions within the same minimum cogging torque period, and each minimum cogging torque period corresponds to 1 / p of a rotation period of the rotor (20), wherein p is the number of pole pairs of the rotor (20).

17. The control device of claim 10, wherein The control device further includes an initial correction module configured to correct a zero-degree mechanical angle position of the rotor (20) before the compensation measurement module measures and calculates the plurality of cogging torques.

18. The control device according to any one of claims 10 to 17, characterized by The compensation measurement module is further configured to measure and calculate the plurality of cogging torques and generate the lookup table before the permanent magnet synchronous motor is shipped.

Images