Permanent magnet synchronous motor cogging effect parameter identification module and method

Abstract

The invention discloses a permanent magnet synchronous motor cogging effect parameter identification module and method. The module comprises a shell, an IMU, a battery, a Bluetooth module and a bottom plate. The shell is installed on the module bottom plate, the IMU is installed on the bottom plate, the battery and the Bluetooth module are both installed on the bottom plate, the battery is connected with the IMU and the Bluetooth module through two pairs of cables, the IMU is connected with the Bluetooth module through a cable, and the identification method comprises the steps that firstly, the tooth groove identification module is installed; step 2, the tooth space identification module performs uniformly accelerated motion; 3, receiving linear acceleration data and a mechanical rotation angle in real time; 4, processing the data and obtaining a data pair; 5, performing first fitting on the data pair; step 6, obtaining a mechanical rotation angle and cogging torque data pair; and step 7, carrying out second fitting on the data pair, and obtaining the data pair representing the cogging effect according to the fitting result. According to the method, the cogging effect of the permanent magnet synchronous motor driver or the equipment joint driven by the permanent magnet synchronous motor is identified based on the acceleration information.

Description

Technical Field

[0001] The present invention relates to a cogging effect parameter identification system and method, and in particular to a permanent magnet synchronous motor cogging effect parameter identification module and method, and belongs to the technical field of electric drive. Background Art

[0002] Motor drivers have the advantages of simple structure, convenient control and high power density, and are widely used in industrial production activities, among which a considerable number are permanent magnet synchronous motors. The permanent magnet synchronous motor contains several pairs of permanent magnet poles and coils, which are installed on the rotor and stator of the motor respectively. By applying current in the coil to generate a magnetic field, the poles and coils attract and repel each other, thereby driving the rotor to rotate relative to the stator. However, the permanent magnet poles are not arranged continuously, but are discretely and evenly installed on the rotor. There will inevitably be air gaps between adjacent poles. During the operation of the motor, the position corresponding to each coil will present a state of alternating "pole-air gap", which will cause the actual output torque of the motor to fluctuate. This is the cogging effect of the permanent magnet synchronous motor. The electromechanical characteristics of this permanent magnet synchronous motor will introduce high-frequency signal interference in the current, angular velocity and other aspects during operation, reduce the motion control performance of the motor and generate noise. The impact is more obvious when running at low speed, which will cause significant torque fluctuations and position movement. Therefore, parameter identification and compensation work for the cogging effect of permanent magnet synchronous motors is of great significance.

[0003] Current research indicates that the effect of cogging on motor output torque is only related to the relative position of the rotor and stator, and has nothing to do with the motor running speed and load. Therefore, parameter identification of cogging is actually to identify its quantitative relationship with the relative position of the motor rotor and stator. From the perspective of sensor data acquisition, the identification methods of cogging can be divided into three main types: identification based on torque information, identification based on current information, and identification based on acceleration information. (1) Torque information is generally obtained directly through torque sensors. For a small number of motor drives equipped with torque sensors, the torque effect of cogging can be obtained by directly measuring sensor data, but for most other motor drives without torque sensors, it is necessary to obtain relevant data through an additional sensor test platform. For motors that have been installed in equipment, it is necessary to disassemble the equipment and install torque sensors in series in the motor transmission chain. This is not only extremely inconvenient, but the electromechanical characteristics of individual components may also change during the loading and unloading process. (2) For permanent magnet synchronous motors, in actual control, all kinds of control laws at the upper level will eventually be converted into current signals input to the driver. Therefore, current information is the underlying data reflecting the characteristics of the cogging effect, and the cogging effect characteristics can be identified based on the current change trend. In order to improve the authenticity of the current signal in the cogging torque, the motor needs to achieve uniform rotation in a closed loop at the lowest possible speed. In fact, due to the existence of cogging torque, most permanent magnet brushless currents are difficult to ensure the stability of low-speed operation, which leads to distortion of the collected current signal. In addition, due to the existence of electromagnetic interference, the q-axis current of the motor usually has a more obvious high-frequency interference signal. (3) For permanent magnet synchronous motor drives with known load inertia, the effect of the cogging torque will be directly reflected in the change of angular acceleration. According to Newton's second law, the torque is proportional to the angular acceleration of the corresponding inertia. Therefore, the cogging effect can be indirectly identified based on the change trend of the angular velocity. Most motor-driven robot joints are equipped with encoders, but obtaining angular acceleration information requires secondary differential processing, which will cause obvious high-frequency interference.

[0004] In summary, the traditional cogging effect identification equipment requires the disassembly of equipment components, and the traditional identification method is affected by electromagnetic interference signals, resulting in distortion of the cogging effect identification results. Summary of the invention

[0005] In order to overcome the existing technology, the present application provides a permanent magnet synchronous motor cogging effect parameter identification module and method.

[0006] A permanent magnet synchronous motor cogging parameter identification module comprises a shell, an IMU, a battery, a Bluetooth module and a base plate; the shell is installed on the module base plate to cover the IMU, the battery and the Bluetooth module, an IMU2 is installed on the base plate, the battery and the Bluetooth module are both installed on the base plate, and the installation positions of the IMU2, the battery and the Bluetooth module enable the identification module to maintain dynamic balance during rotation, the battery is connected to the IMU2 and the Bluetooth module respectively through two pairs of cables, and the IMU2 is connected to the Bluetooth module through a cable.

[0007] A method for identifying parameters of cogging effect of a permanent magnet synchronous motor comprises:

[0008] Step 1: Install the tooth slot identification module on the permanent magnet synchronous motor driver or the moving part of the equipment joint driven by the permanent magnet synchronous motor that needs to be identified, and the x-axis of IMU2 is in the rotation tangential direction;

[0009] Step 2: The driver or joint under test drives the tooth slot identification module to perform uniform acceleration motion until the running speed reaches the critical speed;

[0010] Step 3: Make the permanent magnet synchronous motor of the tested drive or joint enter the zero torque mode with the target output torque of zero. At this time, the tested drive or joint drives the tooth slot identification module to perform free deceleration movement. In this process, the Bluetooth module 4 is used to receive the x-axis acceleration data α of the IMU2 in real time. x , and collect the motor's mechanical angle θ in real time. All data are accompanied by the system time when collected;

[0011] Step 4: Wait until the free deceleration motion of the tooth slot identification module driven by the tested drive or joint stops, complete the sensor data collection, and process the collected linear acceleration α x and mechanical rotation angle θ, and obtain the mechanical rotation angle and linear acceleration data pair corresponding to each other in time represents the mechanical rotation angle sequence, represents the linear acceleration sequence;

[0012] Step 5: Use linear function to Perform the first least squares fit and the online acceleration series This part of the linear offset is eliminated to remove the low-frequency acceleration component that is not related to the cogging effect, and obtain the mechanical angle and linear acceleration data that are only related to the cogging effect. represents the cogging effect linear acceleration sequence;

[0013] Step 6: Compare the data obtained in step 5 The linear acceleration data in is divided by the distance of the IMU relative to the rotation axis and multiplied by the equivalent inertia of the joint rotation part to obtain the mechanical angle and cogging torque data pair. represents the cogging torque;

[0014] Step 7: Use the fast Fourier method to perform frequency domain analysis on the data obtained in step 6, and select the first n frequencies with the largest amplitude as the main frequencies of the cogging effect vibration, so as to construct the main frequency function Among them: a i represents the amplitude, b i represents frequency, c i Indicates the phase, and the data obtained in step 6 is Perform the second least squares fitting, and finally get the data pair representing the cogging effect according to the fitting results.

[0015] Further optimization steps are included:

[0016] Step 8: Install the tooth slot identification module on the permanent magnet synchronous motor driver or the moving part of the equipment joint driven by the permanent magnet synchronous motor that needs to be identified, and the x-axis of the IMU is in the tangential direction of rotation;

[0017] Step 9: First, the motor under test is made to execute a torque command that increases in a step-like manner starting from zero. When the torque just makes the motor run at a certain speed, the torque is recorded as the critical torque.

[0018] Step 10: The drive or joint under test drives the cogging identification module with a constant critical torque command within the allowable operating range, and at the same time adds a feedforward compensation torque based on the mechanical angle to the motor control torque. The specific value is obtained through the cogging effect data obtained in step 7. Calculated by interpolation;

[0019] Step 11: According to the initial value of the cogging effect data Establish iterative optimization data pair Represents the sequence of iterative optimization torque, all initial values ​​are zero, and during the execution of the constant torque instruction, the x-axis acceleration α of IMU2 is collected in real time x and the mechanical rotation angle θ of the motor, and the two data collected within the set time are combined into a data pair;

[0020] Step 12: For each data pair collected in step 11, determine the mechanical angle of the iterative optimization data pair closest to the current position according to the mechanical angle. The corresponding iterative optimization torque is: h(θ)[k]=h(θ)[k-1]+K p α x [k], where k is the number of updates of the iterative optimization torque at this position, K p is the iterative learning coefficient, is p-type iterative learning, and iteratively optimizes the data pair Add to the calculation of feedforward compensation torque;

[0021] Step 13: Repeat steps 10 to 12 until the iterative optimization data pair is Convergence, satisfaction ε represents the critical value of the convergence setting, and the data obtained from this is This is the identified permanent magnet synchronous motor cogging effect.

[0022] Compared with the prior art, the present invention has the following beneficial effects:

[0023] The method for identifying the cogging effect parameters of a permanent magnet synchronous motor using an IMU proposed in the present application can identify the cogging effect of a permanent magnet synchronous motor drive or a joint of a device driven by a permanent magnet synchronous motor based on acceleration information.

[0024] Compared with the traditional identification method based on torque information, the required sensor does not need to be installed in series in the torque transmission chain, so there is no need to disassemble the core transmission components of the equipment. The proposed method only needs to be installed on the surface of the moving parts of the equipment, which has a wider applicability. In addition, the price of IMU is much lower than that of torque sensor, and its use cost is also lower.

[0025] Compared with the traditional identification method based on current information, the ratio between acceleration data and cogging torque is the joint inertia. When the system state is known, the joint inertia can be accurately measured, while the ratio between current and cogging torque is the torque coefficient, which may change with the motor operating state and cannot be accurately measured. Secondly, the motor current usually has a more obvious high-frequency electromagnetic interference signal. Filtering it will cause the identification result of the cogging effect to be distorted, while IMU does not have this problem. Therefore, on the whole, acceleration data can more accurately and directly reflect the true state of the cogging effect than current data.

[0026] The technical solution of the present invention is further described below in conjunction with the accompanying drawings and embodiments: BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is a three-dimensional assembly diagram of the tooth groove identification module of the present application;

[0028] Figure 2 An exploded schematic diagram of the tooth slot identification module of the present application;

[0029] Figure 3 This is a flow chart of the free deceleration initial value identification experimental method in the parameter identification method for the cogging effect using the proposed cogging identification module in this application;

[0030] Figure 4This is a flow chart of a uniform speed iterative optimization experimental method in a parameter identification method for the cogging effect using the proposed cogging identification module in this application;

[0031] Figure 5 A schematic diagram of the installation of using the cogging identification module of the present application to identify the cogging effect of a permanent magnet synchronous motor;

[0032] Figure 6 This is a schematic diagram of the installation of using the tooth slot identification module of the present application to identify the joints of industrial robot arm equipment. DETAILED DESCRIPTION

[0033] The following is a detailed description of the embodiments of the technical solution of the present application in conjunction with the accompanying drawings. Unless otherwise specified, the technical terms or scientific terms used in the present application should have the common meanings understood by those skilled in the art to which the present application belongs.

[0034] Combination Figure 1-Figure 2 It is explained that a permanent magnet synchronous motor slot effect parameter identification module based on IMU in this embodiment includes a housing 1, an IMU 2, a battery 3, a Bluetooth module 4 and a base plate 5;

[0035] IMU2 is fixed to the module base plate 5 by bolts, and the battery 3 and the Bluetooth module 4 are both installed on the module base plate 5 by grooves, and the IMU2 is installed on one side of the module base plate 5, and the battery 3 and the Bluetooth module 5 are installed on the other side of the module base plate 5, so as to ensure the dynamic balance of the overall structure during rotation. The module housing 1 is fixed to the module base plate 5 by bolts, and the internal core components and cables are wrapped inside;

[0036] In this solution, the module housing 1 and the module base plate 5 can be manufactured by 3D printing, and the material can be resin.

[0037] Preferably, the IMU 2, the battery 3 and the Bluetooth module 4 are existing products, and the battery 3 is respectively connected to the IMU 2 and the Bluetooth module 4 through two pairs of cables for energy supply. The IMU 2 and the Bluetooth module 4 are connected through a cable for serial communication data transmission.

[0038] Inertial sensors represented by IMU can directly measure acceleration data. When the IMU is fixed relative to the motor shaft, the angular acceleration information can be directly calculated without differentiation. In the identification of the cogging effect of actual equipment, unlike torque sensors, IMUs can be installed on the surface of the moving parts of the equipment without being installed in series in the transmission chain. Angular velocity data is easier to obtain than torque data, and there is no need to disassemble the core transmission components of the equipment when collecting data. Therefore, the method of using IMU to collect acceleration signals has application prospects in the identification of permanent magnet synchronous motor cogging effect parameters.

[0039] Embodiment 1, combination Figure 2-Figure 5 It is explained that the cogging effect of the permanent magnet synchronous motor driver II is identified by using the cogging identification module I proposed in the present application. The method comprises two parts: a free deceleration initial value identification experiment and a uniform speed operation iterative optimization experiment.

[0040] The free deceleration initial value identification experiment is carried out for the permanent magnet synchronous motor drive, which includes:

[0041] Step 1: Install the tooth slot identification module I on the rotor output part of the permanent magnet synchronous motor driver II, and the x-axis of IMU2 is in the rotation tangential direction of the driver II;

[0042] Step 2: Control driver II to drive tooth slot identification module I to perform uniform acceleration motion until the running speed reaches the critical speed v s , the critical speed needs to ensure that the rotor part can rotate at least one circle during the subsequent free rotation process;

[0043] Step 3: Control driver II to enter zero torque mode. During free rotation, use Bluetooth module 4 to receive the x-axis acceleration data of IMU2 in real time, and collect the mechanical angle of the motor in real time. All data are accompanied by the system time when collected.

[0044] Step 4: After the free deceleration motion of driver II stops, the sensor data collection is completed, and the collected linear acceleration data α is processed using the time-based interpolation method. x and mechanical rotation angle data θ, and obtain a pair of mechanical rotation angle and linear acceleration data corresponding to each other in time.

[0045] Step 5: Use linear function to Perform the first least squares fit and the online acceleration data This part of the linear offset is eliminated to remove the low-frequency acceleration component that is not related to the cogging effect, and obtain the mechanical angle and linear acceleration data that are only related to the cogging effect. represents the cogging effect linear acceleration;

[0046] Step 6: Compare the data obtained in step 5 The linear acceleration data in is divided by the distance L between IMU2 and the rotor axis of rotation, and multiplied by the equivalent inertia J of the joint rotation part to obtain the mechanical angle and cogging torque data pair.

[0047]

[0048] Step 7: Use the fast Fourier transform method (FFT) to perform frequency domain analysis on the data obtained in step 6, and select the first 10 frequencies with the largest amplitude as the main frequencies of the cogging effect vibration, so as to construct the main frequency function And for the data obtained in step 6 The second least squares fitting is performed, and the data pair representing the initial value of the cogging effect is finally obtained according to the fitting results.

[0049] The uniform speed operation iterative optimization experiment is carried out for the permanent magnet synchronous motor drive, which includes:

[0050] Step 1: Install the tooth slot identification module I on the rotor output part of the permanent magnet synchronous motor driver II, and the x-axis of IMU2 is in the rotation tangential direction of the driver II;

[0051] Step 2: Control driver II to execute a torque command that increases stepwise from zero. When the torque just makes driver II run at a certain speed, record the torque as the critical torque.

[0052] Step 3: Control driver II to drive cogging identification module I to operate at a constant critical torque. At the same time, a feedforward compensation torque based on the mechanical angle is added to the control torque. The specific value is obtained by the data of the initial value of the cogging effect. Calculated by interpolation;

[0053] Step 4: According to the initial value of the cogging effect data Establish iterative optimization data pair All initial values ​​are zero. During the execution of the constant torque command, the x-axis acceleration data α of IMU2 is collected in real time. x and the mechanical angle data θ of the motor, and the two data collected within 1ms are combined into a data pair;

[0054] Step 5: For each data pair collected in step 4, determine the mechanical angle of the iterative optimization data pair closest to the current position according to the mechanical angle, and the corresponding iterative optimization torque h(θ) is: h(θ)[k]=h(θ)[k-1]+K p α x [k], and iteratively optimize the data pair It is also added to the calculation of feedforward compensation torque;

[0055] Step 6: Repeat steps 3 to 5 until the iterative optimization data pair is tends to converge, and the data obtained is That is the cogging effect of permanent magnet synchronous motor driver II obtained by identification.

[0056] Embodiment 2, combination Figure 2-Figure 4 as well as Figure 6 It is explained that the cogging effect of each joint of a six-degree-of-freedom industrial robot arm III is identified using the cogging identification module I proposed in this application. For each joint, the method includes two parts: free deceleration initial value identification experiment and uniform speed operation iterative optimization experiment.

[0057] Preferably, the six-degree-of-freedom industrial robot arm III used is an existing product, wherein the operating rotation angles of the 1st to 5th joints are limited values, and the operating rotation angle of the 6th joint is not limited.

[0058] In the specific implementation of the scheme, the cogging identification module I is installed on the flange of the end actuator of the robot arm by bolts. The cogging effect identification of the sixth joint can be operated according to Example 1. For the identification operation scheme of the remaining joints, taking the fifth joint as an example, its free deceleration initial value identification experiment includes:

[0059] Step 1: Adjust the positions of the joints of the robot arm III so that the x-axis direction of the IMU2 is located in the rotation tangential direction of the fifth joint, and control the fifth joint to drive the tooth slot identification module I to perform uniform acceleration movement within the allowable operating range until the operating speed reaches the critical speed;

[0060] Step 2: Control the fifth joint to enter the zero torque mode. During the free rotation process, use the Bluetooth module 4 to receive the x-axis acceleration data of the IMU 2 in real time, and collect the mechanical angle of the motor in real time. All data are accompanied by the system time when collected.

[0061] Step 3: Since the allowed running range is limited, the running starting position and running direction can be changed, and step 2 can be repeated until the free rotation angle covers the entire allowed running range;

[0062] Step 4: Combine all collected data to obtain sensor data within the entire allowable operating range, and use the time-based interpolation method to process the collected linear acceleration data α x and mechanical rotation angle data θ, and obtain a pair of mechanical rotation angle and linear acceleration data corresponding to each other in time.

[0063] Step 5: Use linear function to Perform the first least squares fit and the online acceleration data This part of the linear offset is eliminated to remove the low-frequency acceleration component that is not related to the cogging effect, and obtain the mechanical angle and linear acceleration data that are only related to the cogging effect.

[0064] Step 6: Compare the data obtained in step 5 The linear acceleration data in is divided by the distance L between IMU2 and the rotation axis of the fifth joint, and multiplied by the equivalent inertia J of the rotating part of the joint to obtain the mechanical angle and cogging torque data pair.

[0065]

[0066] Step 7: Use the FFT method to perform frequency domain analysis on the data obtained in step 6, and select the first 10 frequencies with the largest amplitude as the main frequencies of the cogging effect vibration to construct the function And for the data obtained in step 6 The second least squares fitting is performed, and the data pair representing the initial value of the cogging effect is finally obtained according to the fitting results.

[0067]

[0068] The uniform speed iterative optimization experiment includes:

[0069] Step 1: Adjust the positions of the joints of the robot arm III so that the x-axis direction of IMU2 is located in the rotation tangential direction of the fifth joint. Control the fifth joint to execute a step-by-step torque command starting from zero, and add the dynamic feedforward torque of the fifth joint to the torque command to offset the influence of gravity. When the torque just makes the fifth joint run at a constant speed, record the torque as the critical torque.

[0070] Step 2: Control the fifth joint to drive the tooth slot identification module I to operate at a constant critical torque. At the same time, add the dynamic feedforward torque based on the mechanical angle and the tooth slot feedforward compensation torque to the control torque. The specific value of the tooth slot feedforward compensation torque is obtained by the data of the initial value of the tooth slot effect. Calculated by interpolation;

[0071] Step 3: Since the allowed running range is limited, when the fifth joint runs to the range boundary, the direction of the critical moment is changed, and the running direction of the fifth joint is changed;

[0072] Step 4: According to the initial value of the cogging effect data Establish iterative optimization data pair All initial values ​​are zero. During the execution of the constant torque command, the x-axis acceleration data of IMU2 during the uniform speed operation of the 5th joint is collected in real time. and the mechanical angle data θ of the motor, and the two data collected within 1ms are combined into a data pair;

[0073] Step 5: For each data pair collected in step 4, the mechanical angle of the iterative optimization data pair closest to the current position is determined according to the mechanical angle, and the corresponding iterative optimization torque value is updated as: h(θ)[k]=h(θ)[k-1]+K p α x [k], and iteratively optimize the data pair Add to the calculation of feedforward compensation torque;

[0074] Step 6: Repeat steps 2 to 5 until the iterative optimization data pair is Convergence, satisfaction ε represents the critical value of the convergence setting, and the data obtained from this is This is the identified cogging effect of the fifth joint of the industrial robot arm III.

[0075] The above method can also be applied to joints 1-4 and will not be repeated here.

[0076] The present application has been disclosed as above with preferred implementation cases, but it is not used to limit the present invention. Any technician familiar with this profession can make slight changes or modifications to equivalent implementation cases with equivalent changes by using the above-disclosed structures and technical contents without departing from the scope of the technical solution of the present application, which still fall within the scope of the technical solution of the present application.

Claims

1. A permanent magnet synchronous motor cogging effect parameter identification module, characterized in that: It includes a housing 1, an IMU 2, a battery 3, a Bluetooth module 4 and a base plate 5; The housing 1 is installed on the module base plate 5 to cover the IMU2, the battery 3 and the Bluetooth module 4. The IMU2 is installed on the base plate 5. The battery 3 and the Bluetooth module 4 are both installed on the base plate 5. The installation positions of the IMU2, the battery 3 and the Bluetooth module 5 enable the identification module to maintain dynamic balance during rotation. The battery 3 is respectively connected to the IMU2 and the Bluetooth module 4 through two pairs of cables, and the IMU2 is connected to the Bluetooth module 4 through a cable.

2. According to claim 1, a permanent magnet synchronous motor cogging parameter identification module is characterized in that: The IMU 2 is installed on one side of the module base plate 5 , and the battery 3 and the Bluetooth module 4 are installed on the other side of the module base plate 5 , wherein the one side is opposite to the other side.

3. According to claim 1, a permanent magnet synchronous motor cogging parameter identification module is characterized in that: The housing 1 and the base plate 5 are respectively manufactured by 3D printing.

4. According to claim 1, a permanent magnet synchronous motor cogging parameter identification module is characterized in that: The housing 1 and the bottom plate 5 are both made of resin.

5. A method for identifying parameters of cogging effect of a permanent magnet synchronous motor, characterized in that: Based on the identification module according to any one of claims 1 to 4, the identification method comprises: Step 1: Install the tooth slot identification module on the permanent magnet synchronous motor driver or the moving part of the equipment joint driven by the permanent magnet synchronous motor that needs to be identified, and the x-axis of IMU2 is in the rotation tangential direction; Step 2: The driver or joint under test drives the tooth slot identification module to perform uniform acceleration motion until the running speed reaches the critical speed; Step 3: Make the permanent magnet synchronous motor of the tested drive or joint enter the zero torque mode with the target output torque of zero. At this time, the tested drive or joint drives the tooth slot identification module to perform free deceleration movement. In this process, the Bluetooth module 4 is used to receive the x-axis acceleration data α of the IMU2 in real time. x , and collect the motor's mechanical angle θ in real time, all data are accompanied by the system time when collected; Step 4: Wait until the free deceleration motion of the tooth slot identification module driven by the tested drive or joint stops, complete the sensor data collection, and process the collected linear acceleration α x and mechanical rotation angle θ, and obtain the mechanical rotation angle and linear acceleration data pair corresponding to each other in time represents the mechanical rotation angle sequence, represents the linear acceleration sequence; Step 5: Use linear function to Perform the first least squares fit and the online acceleration series This part of the linear offset is eliminated, and the mechanical angle and linear acceleration data related only to the cogging effect are obtained. represents the cogging effect linear acceleration sequence; Step 6: Compare the data obtained in step 5 The linear acceleration data in is divided by the distance of the IMU relative to the rotation axis and multiplied by the equivalent inertia of the joint rotation part to obtain the mechanical angle and cogging torque data pair. represents the cogging torque; Step 7: Use the fast Fourier method to perform frequency domain analysis on the data obtained in step 6, and select the first n frequencies with the largest amplitude as the main frequencies of the cogging effect vibration, so as to construct the main frequency function Among them: a i represents the amplitude, b i represents frequency, c i Indicates the phase, and the data obtained in step 6 is Perform the second least squares fitting, and finally get the data pair representing the cogging effect according to the fitting results.

6. The method for identifying cogging parameters of a permanent magnet synchronous motor according to claim 5, characterized in that: The following optimization steps are also included: Step 8: Install the tooth slot identification module on the permanent magnet synchronous motor driver or the moving part of the equipment joint driven by the permanent magnet synchronous motor that needs to be identified, and the x-axis of the IMU is in the tangential direction of rotation; Step 9: First, the motor under test is made to execute a torque command that increases in a step-like manner starting from zero. When the torque just makes the motor run at a certain speed, the torque is recorded as the critical torque. Step 10: The drive or joint under test drives the cogging identification module with a constant critical torque command within the allowable operating range, and at the same time adds a feedforward compensation torque based on the mechanical angle to the motor control torque. The specific value is obtained through the cogging effect data obtained in step 7. Calculated by interpolation; Step 11: According to the initial value of the cogging effect data Establish iterative optimization data pair Represents the sequence of iterative optimization torque, all initial values ​​are zero, and during the execution of the constant torque instruction, the x-axis acceleration α of IMU2 is collected in real time x and the mechanical rotation angle θ of the motor, and the two data collected within the set time are combined into a data pair; Step 12: For each data pair collected in step 11, determine the mechanical angle of the iterative optimization data pair closest to the current position according to the mechanical angle. The corresponding iterative optimization torque is: h(θ)[k]=h(θ)[k-1]+K p α x [k], where k is the number of updates of the iterative optimization torque at this position, K p is the iterative learning coefficient, is p-type iterative learning, and iteratively optimizes the data pair Add to the calculation of feedforward compensation torque; Step 13: Repeat steps 10 to 12 until the iterative optimization data pair is Convergence, satisfaction ε represents the critical value of the convergence setting, and the data obtained from this is This is the identified permanent magnet synchronous motor cogging effect.

7. The method for identifying cogging parameters of a permanent magnet synchronous motor according to claim 5, characterized in that: In step 4, the collected linear acceleration and mechanical rotation angle are processed using a time-based interpolation method.

8. The method for identifying cogging parameters of a permanent magnet synchronous motor according to claim 5, characterized in that: In step 7, the first 10 frequencies with the largest amplitudes are selected as the main frequencies of the cogging effect vibration.

9. The method for identifying cogging parameters of a permanent magnet synchronous motor according to claim 6, characterized in that: In step 11, two data collected within 1 ms are combined into a data pair.

Images