Axial flux permanent magnet synchronous motor accurate position estimation method

The MCT method with a variable step length scheme compensates for estimation errors in axial flux permanent magnet synchronous motors, improving accuracy and timeliness of rotor position estimation by dynamically adjusting the compensation angle.

CN119093800APending Publication Date: 2024-12-06HARBIN INST OF TECH +2
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411192671.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-28
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

Axial flux permanent magnet synchronous motors face significant position estimation errors due to high fundamental frequencies and varying motor parameters, especially at high speeds, making accurate rotor position estimation difficult and affecting motor performance.

Method used

A method using minimum current tracking (MCT) with a variable step length scheme to compensate for estimation errors, adjusting the compensation angle based on current sum comparisons, ensuring rapid convergence and stability without detailed factor-by-factor calculations.

Benefits of technology

Improves the accuracy and timeliness of rotor position estimation by dynamically adjusting the step size, reducing errors and enhancing motor control efficiency and responsiveness.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119093800A_ABST
    Figure CN119093800A_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of magnetic flux permanent magnet synchronous motors, and discloses an axial magnetic flux permanent magnet synchronous motor accurate position estimation method comprising the following steps: 1, obtaining an axial motor estimated rotor position, and carrying out angle compensation of one step length on the rotor position; step 2, according to a set judgment period, calculating a current sum in a single judgment period, and according to a set judgment frequency, comparing the current sum of the current judgment period with the current sum of the previous period to obtain a comparison result; and step 3, according to the comparison result, adjusting the compensation angle step length of the next judgment period, and finally tending the step length to be 0 to obtain the accurate position of the motor rotor. According to the method, the variable step size scheme is adopted, the position of the motor rotor is accurately obtained, the accuracy is improved, and the convergence speed and the stability are both considered. The implementation mode is simple, the efficiency is higher, and the application range is wide.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the technical field of axial flux permanent magnet synchronous motors, and in particular to a method for accurately estimating the position of an axial flux permanent magnet synchronous motor. Background Art

[0002] Axial flux permanent magnet synchronous motors are widely used in electric and hybrid vehicle transmission systems and wind turbines because their short axial length allows the mechanical structure to be integrated into a compact transmission system, and their specific shape makes them suitable for high pole pairs, making them very suitable for direct drive applications. To ensure the operating performance of axial flux permanent magnet synchronous motors, accurate rotor position estimation during the dynamic operation of the motor can ensure precise control of the motor output torque, and also help the motor controller to adjust the control strategy in a timely manner, thereby enhancing the dynamic response capability of the motor.

[0003] At present, in order to improve the reliability of the system and reduce the cost, the control of the rotor of the axial flux permanent magnet synchronous motor at high speed often adopts a position sensorless control method, including algorithms such as model reference adaptive observer, sliding film observer and state expansion observer method.

[0004] However, axial flux permanent magnet synchronous motors generally have more pole pairs, and at higher speeds, they will have higher fundamental frequencies and lower carrier ratios. Under such conditions, the error caused by the observer will increase with the increase in fundamental frequency, resulting in increased error and affecting position accuracy. On the other hand, as the speed increases, the influence of various motor parameters will also increase, resulting in increased position error, which will further increase the error of position estimation and make it difficult to achieve higher accuracy requirements, affecting motor performance. Summary of the invention

[0005] The present invention aims to provide a method for accurately estimating the position of an axial flux permanent magnet synchronous motor, so as to solve the problem in the prior art that the error in estimating the rotor position of an axial flux permanent magnet synchronous motor is large and it is difficult to achieve higher accuracy requirements.

[0006] To achieve the above-mentioned purpose, the present invention adopts the following technical scheme, an axial flux permanent magnet synchronous motor accurate position estimation method, after estimating the position information of the axial flux permanent magnet synchronous motor, this scheme uses the minimum current tracking scheme (MCT) to compensate for its estimation error, and adopts a variable step length scheme to achieve rapid convergence of the MCT scheme and avoid the jitter of the fixed step length scheme after the system is stabilized, thereby improving the accuracy and timeliness of rotor positioning. The estimation method is to estimate and compensate the motor rotor position at a higher speed. Considering that the axial flux permanent magnet synchronous motor has a large number of pole pairs, a general speed ≥5000r / min is considered to be a high speed, and specifically includes the following steps:

[0007] Step 1, obtaining the axial motor rotor position, and performing a step angle compensation on the rotor position;

[0008] Step 2, according to the set judgment cycle, calculate the current sum in a single judgment cycle, and compare the current sum of the current judgment cycle with the current sum of the previous cycle according to the set judgment frequency to obtain a comparison result;

[0009] Step 3: According to the comparison result, the compensation angle step size of the next judgment cycle is adjusted, and the final step size tends to 0 to obtain the motor rotor position.

[0010] The principles and advantages of this solution are:

[0011] In the existing sensorless control method, the position estimation error comes from two main sources. First, it comes from the observer and the high-pass filter (HPF) and low-pass filter (LPF) in the system. Such errors can be quantitatively analyzed, and the resulting position error can be expressed as:

[0012]

[0013] Where Vθ f is the position estimation error, ω e is the motor angular velocity, ω H is the cutoff frequency of the HPF, ω L is the cutoff frequency of the LPF.

[0014] From the above formula, it can be seen that the position error caused by the filter can be calculated and increases with the increase of the fundamental frequency.

[0015] Secondly, there are many factors involved in the operation of the motor, such as the deviation of motor parameters. As the speed increases, the cross-coupling effect becomes increasingly serious, and the inductance parameters change greatly, resulting in certain errors in position estimation. In addition, position errors caused by factors such as current and voltage sampling errors and inverter nonlinearity, especially when the speed increases, the fundamental frequency = speed * pole pair number / 60, that is, generally expressed as n = 60f / p, where n is the speed, f is the fundamental frequency, and p is the pole pair number, resulting in an increase in the fundamental frequency, and the carrier ratio = fundamental frequency / carrier frequency. When the number of axial motor pole pairs is large, the carrier is large. In this case, the PWM delay and the impact on position estimation are very large, and the above factors are difficult to quantitatively analyze and compensate, resulting in this part of the error being difficult to determine and calculate. In the case of low speed, this part of the error is small and can generally be ignored, but when the speed is too high, the error caused by this part of the factors will increase significantly, affecting the accuracy of the final position determination. However, it would be difficult to perform estimated calculations for these factors one by one. Even if it could be achieved, the amount of individual data involved in each factor is small, and the amount of calculation would be very large, resulting in a sharp increase in workload, which would affect the motor control efficiency and reduce the timeliness and responsiveness of the control.

[0016] Therefore, after considering the error influence caused by two aspects, this scheme does not only analyze and calculate a certain error factor, but after estimating the position information of the axial flux permanent magnet synchronous motor, first makes a small adjustment, and then combines the minimum current tracking method to check whether the motor current decreases after the adjustment, and uses the variable step size method according to the comparison result to dynamically adjust the compensation, so as to achieve full compensation for the position estimation error of the axial flux permanent magnet synchronous motor, and obtain accurate axial motor rotor position, reduce position error, take into account convergence speed and stability, and at the same time do not need to calculate the error influencing factors one by one, so its implementation method is simpler and faster, with lower calculation amount, higher efficiency, and wider range of application.

[0017] Furthermore, in step 1, the step length of compensation is expressed as:

[0018] dθ c (k)=μ*|i sum (k)-i sum (k-1)|;

[0019] In the formula, k represents the current calculation judgment cycle, k-1 represents the previous calculation current judgment cycle, μ represents the step coefficient, and i sum It indicates that the current sum within the calculation judgment period is calculated, and the initial state is k=1.

[0020] Furthermore, the judgment period is thirty carrier periods; and the judgment frequency is 30 divided by the carrier frequency.

[0021] Furthermore, in step 3, the step size of the compensation angle of the next judgment cycle is adjusted in the following two ways:

[0022] Mode 1: When the calculated current sum of the current judgment cycle is less than the current sum of the previous cycle, the compensation angle of the next judgment cycle is adjusted to increase by one step in the direction of the previous step;

[0023] Mode 2: When the calculated current sum of the current judgment cycle is greater than the current sum of the previous cycle, the compensation angle of the next judgment cycle is adjusted to increase by one step in the opposite direction of the previous step.

[0024] Furthermore, the compensation angle is calculated according to the compensation step length, which is expressed as:

[0025]

[0026] Furthermore, it also includes calculating the estimation error compensation according to the calculated compensation angle, which is expressed as:

[0027]

[0028] In the formula Estimate the rotor position for k periods.

[0029] Furthermore, the current sum of the current cycle is expressed as:

[0030]

[0031] Wherein, α represents the α axis in the coordinate axis obtained by the CLARKE transformation of the three-phase current, and β represents the β axis.

[0032] Furthermore, in method 1, the adjusted step size can be expressed as:

[0033] -1*dθ c (k+1).

[0034] Furthermore, in method 1, the adjusted step size can be expressed as:

[0035] -1*dθ c (k+1).

[0036] Furthermore, the estimation method uses a FOC algorithm to perform operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 It is a schematic flow chart of the precise position estimation method of the axial flux permanent magnet synchronous motor of the present invention.

[0038] Figure 2It is a schematic diagram of the MCT algorithm flow of the axial flux permanent magnet synchronous motor accurate position estimation method of the present invention. DETAILED DESCRIPTION

[0039] The following is further described in detail through specific implementation methods:

[0040] Example

[0041] As attached Figure 1 As shown in the figure, this scheme uses the minimum current tracking scheme (MCT) to compensate for its estimation error and realize the control of the maximum torque per ampere (MTPA). It is no longer just for a certain error factor to fully compensate for the position estimation error of the axial flux permanent magnet synchronous motor. Instead, a variable step size scheme is adopted to accurately obtain the motor rotor position, improve accuracy, and take into account the convergence speed and stability. The implementation method is simple, more efficient, and has a wide range of practical applications.

[0042] The estimation method is to estimate and compensate the motor rotor position at a higher speed (above 5000r / min). This method is mainly based on the FOC algorithm to perform operations. This solution is aimed at obtaining the estimated rotor position through the position estimation algorithm without using a position sensor. The general position estimation method usually has errors, so the estimated position is compensated on this basis. The FOC algorithm and the position estimation algorithm will be calculated once in each carrier cycle, and the new voltage output to the motor is calculated according to the current state of the motor. Specifically, the following steps are included:

[0043] S1, obtains the estimated rotor position of the axial motor and performs a step angle compensation on the rotor position.

[0044] In this embodiment, as shown in the attached Figure 2 As shown, the axial flux permanent magnet synchronous motor position information is first estimated through the state expansion observer, and the estimated rotor position can be expressed as Then the rotor position is compensated for a step angle. In this embodiment, the compensation step is expressed as:

[0045] dθ c (k)=μ*|i sum (k)-i sum (k-1)|(1);

[0046] In the formula, k represents the current calculation judgment cycle, k-1 represents the previous calculation current judgment cycle, μ represents the step coefficient, and i sum It indicates that the current sum within the calculation judgment period is calculated, and the initial state is k=1.

[0047] In the initial stage, the first compensation step length is a constant, that is, when k=1, k-1=0, there is no previous cycle, and there is no current difference. The first compensation step length can be randomly assigned according to the formula.

[0048] Among them, the coefficient μ needs to be determined according to the actual coefficient. When the coefficient error is large, the step size is larger and the convergence is faster; when the system error is small, the step size is smaller and the system is more stable. The step size is dynamically adjusted by using a variable step size. When the system converges, the step size approaches 0, which effectively avoids the jitter of the fixed step size and effectively improves the accuracy.

[0049] Among them, according to the estimated rotor position obtained, the compensation angle is calculated and expressed as:

[0050]

[0051] According to the compensation angle, the estimation error compensation can be obtained, which is expressed as:

[0052]

[0053] In the formula, Estimate the rotor position for k periods.

[0054] S2, according to the set judgment cycle, calculate the current sum in a single judgment cycle, and compare the current sum of the current judgment cycle with the current sum of the previous cycle according to the set judgment frequency to obtain a comparison result.

[0055] In this embodiment, the judgment period is set to thirty carrier cycles, that is, thirty single-chip microcomputer calculation cycles; the judgment frequency is 30 divided by the carrier frequency, the carrier frequency = 1 / carrier cycle, and the judgment frequency = carrier frequency / 30. In order to ensure the stability and rapidity of the MCT scheme, it is necessary to match the step size of the judgment period and the compensation angle. In theory, in order to ensure rapidity, a larger step size and a shorter judgment period are required, so that the system can converge to the precise rotor position as soon as possible. However, if the step size is too long, the position information changes too much, resulting in a large current change, which will affect the stability of the system. And the MCT scheme will not eventually converge to a fixed compensation angle, but will jump up and down at a certain compensation angle, and the step size determines its jumping amplitude. If the judgment period is too short, the current response will not be stable yet, which will affect the polarity judgment of the step size and the stability of the system. Therefore, after comprehensive consideration, setting the judgment frequency to 30 divided by the carrier frequency can ensure the stability of the current response, and in order to ensure the polarity judgment of the step size, the judgment current used is the sum of the current responses of these 30 carrier cycles.

[0056] As attached Figure 2As shown, in this embodiment, the current sum in a single judgment cycle is first calculated, that is, the current sum of thirty carrier cycles, which can be expressed as:

[0057]

[0058] Wherein, α represents the α axis in the coordinate axis obtained by the CLARKE transformation of the three-phase current, and β represents the β axis.

[0059] When the next judgment cycle (i.e., the kth cycle) is generated, the current sum of the next judgment cycle, i.e., the current sum of the current in the current judgment cycle is calculated, and the current sum calculated in the previous judgment cycle (i.e., the k-1 cycle) is compared, which can be expressed as a judgment relationship:

[0060] i sum |(k)<|i sum |(k-1), and get a yes or no comparison result.

[0061] S3, according to the comparison result, adjust the compensation angle step of the next judgment cycle (i.e., k+1 cycle), and finally the step size tends to 0, so as to obtain the precise motor rotor position, so that the motor rotor position is more in line with the actual state, so as to improve the accuracy and timeliness of the real-time estimation of the rotor position.

[0062] In this embodiment, the step size of the compensation angle of the next judgment cycle is adjusted in the following two ways:

[0063] Mode 1: When the calculated current sum of the current judgment cycle is less than the current sum of the previous judgment cycle, the compensation angle of the next judgment cycle is adjusted to increase by one step in the direction of the previous step.

[0064] That is when |i sum |(k)<|i sum |(k-1), when the comparison result is yes, the adjustment step can be expressed as: -1*dθ c (k+1).

[0065] Mode 2: When the calculated current sum of the current judgment cycle is greater than the current sum of the previous judgment cycle, the compensation angle of the next judgment cycle is adjusted to increase by one step in the opposite direction of the previous step.

[0066] That is when |i sum |(k)<|i sum |(k-1), when the comparison result is negative, the adjustment step can be expressed as: 1*dθ c (k+1).

[0067] According to the obtained real-time adjustment step value, the initial angle compensation step input in S1 is adjusted in real time according to formula (1), so as to further accurately compensate the rotor position to improve the accuracy of the rotor position. The final step will approach zero, avoiding the final system jump and improving the overall accuracy and timeliness.

[0068] In this embodiment, a small adjustment is first made based on the estimated rotor position to check whether the motor current decreases after the adjustment. If the current decreases, continue to adjust in this direction; if it increases, adjust in the opposite direction. The adjustment step adopts a variable step length scheme. When the error is large, the convergence speed is faster, and when the error is small, the adjustment step size is smaller. The MCT scheme can achieve rapid convergence and avoid the jitter of the fixed step length scheme after the system is stabilized. Finally, the motor current will converge to the minimum. At this time, it is considered that there is no error in the estimated rotor position, and the accurate motor rotor position is obtained as the real motor rotor position, thereby improving the accuracy and timeliness of estimating the motor rotor position.

[0069] Based on the traditional position estimation scheme, this scheme no longer only targets a certain error factor to fully compensate for the position estimation error of the axial flux permanent magnet synchronous motor. Instead, a variable step length scheme is adopted to obtain accurate axial motor rotor position and realize the control of the maximum torque per ampere (MTPA). Under the same load conditions, the motor current is smaller and the efficiency is higher, taking into account the convergence speed and stability. Its implementation is simple and has a wide range of practical applications.

[0070] The above is only an embodiment of the present invention, and the common knowledge such as the known specific technical solutions and / or characteristics in the solution is not described in detail here. It should be pointed out that for those skilled in the art, without departing from the technical solution of the present invention, several modifications and improvements can be made, which should also be regarded as the protection scope of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A method for accurately estimating the position of an axial flux permanent magnet synchronous motor, characterized in that: The motor rotor position is estimated and compensated at a higher speed, including the following steps: Step 1, obtaining the estimated rotor position of the axial motor and performing an angle compensation of one step on the rotor position; Step 2, according to the set judgment cycle, calculate the current sum in a single judgment cycle, and compare the current sum of the current judgment cycle with the current sum of the previous cycle according to the set judgment frequency to obtain a comparison result; Step 3: According to the comparison result, the compensation angle step size of the next judgment cycle is adjusted, and the final step size tends to 0 to obtain the precise motor rotor position.

2. The method for accurate position estimation of an axial flux permanent magnet synchronous motor according to claim 1 is characterized in that: In step 1, the compensation angle step is expressed as: dθ c (k)=μ*|i sum (k)-i sum (k-1)|; In the formula, k represents the current calculation judgment cycle, k-1 represents the previous calculation current judgment cycle, μ represents the step coefficient, and i sum It indicates that the current sum within the calculation judgment period is calculated, and the initial state is k=1.

3. The method for accurate position estimation of an axial flux permanent magnet synchronous motor according to claim 1 is characterized in that: In step 2, the judgment period is thirty carrier periods; the judgment frequency is 30 divided by the carrier frequency.

4. The method for accurately estimating the position of an axial flux permanent magnet synchronous motor according to claim 1, characterized in that: In step 3, the step size of the compensation angle of the next judgment cycle is adjusted in the following two ways: Mode 1: when the calculated current sum of the current judgment cycle is less than the current sum of the previous judgment cycle, the compensation angle of the next judgment cycle is adjusted to increase by one step in the direction of the previous step; Mode 2: When the calculated current sum of the current judgment cycle is greater than the current sum of the previous judgment cycle, the compensation angle of the next judgment cycle is adjusted to increase by one step in the opposite direction of the previous step.

5. The method for accurately estimating the position of an axial flux permanent magnet synchronous motor according to claim 2, characterized in that: It also includes calculating the compensation angle according to the compensation step length, which is expressed as:

6. The method for accurate position estimation of an axial flux permanent magnet synchronous motor according to claim 5, characterized in that: It also includes calculating the estimated error compensation based on the calculated compensation angle, which is expressed as: In the formula, Estimate the rotor position for k periods.

7. The method for accurate position estimation of an axial flux permanent magnet synchronous motor according to claim 2, characterized in that: The current sum of the current judgment cycle is expressed as: Wherein, α represents the α axis in the coordinate axis obtained by the CLARKE transformation of the three-phase current, and β represents the β axis.

8. The method for accurate position estimation of an axial flux permanent magnet synchronous motor according to claim 4, characterized in that: In method 1, the adjusted step size can be expressed as: -1*dθ c (k+1)。 9. The method for accurately estimating the position of an axial flux permanent magnet synchronous motor according to claim 4, characterized in that: In mode 2, the adjusted step size can be expressed as: 1*dθ c (k+1)。 10. The method for accurate position estimation of an axial flux permanent magnet synchronous motor according to claim 1, characterized in that: The estimation method uses the FOC algorithm to perform the operation.