A self-learning method for delay compensation coefficient of permanent magnet synchronous drive motor for vehicle

Abstract

The present invention relates to the field of permanent magnet synchronous motor control technology, and in particular to a self-learning method for a time delay compensation coefficient of a vehicle permanent magnet synchronous drive motor, comprising: S1, obtaining the no-load torque of the motor under test when the three-phase line is disconnected, and the no-load torque when the torque ring is open; S2, comparing the no-load torque when the torque ring is open and when the three-phase line is disconnected, if the difference is greater than the preset difference, then go to S3; S3, judging whether the self-learning condition is met, if it is met, then go to S4; S4, performing self-learning at the motor inflection point speed and the maximum speed, respectively, calculating the average electrical angles PwrAg1 and PwrAg2 at the two speeds, and combining the preset delay compensation initial value Delta and the delay compensation gain Gain, calculating the self-learning value; S5, comparing the self-learning value with the current default value, if the difference is less than the preset difference value, then using the self-learning value as the new default value. The method can dynamically adjust and self-learn the time delay compensation coefficient of the permanent magnet synchronous motor to adapt to the change of hardware dispersion.

Description

Technical Field

[0001] The present invention relates to the technical field of permanent magnet synchronous motor control, and in particular to a self-learning method for a time delay compensation coefficient of a permanent magnet synchronous drive motor for a vehicle. Background Art

[0002] With the continuous innovation of microprocessor technology, power electronic devices, and the rapid development of motor manufacturing technology, coupled with the widespread application of high-performance rare earth permanent magnet materials, AC servo systems have ushered in unprecedented development opportunities. In particular, the introduction of modern control theory represented by vector control algorithms has enabled permanent magnet synchronous motor (PMSM) servo systems to surpass traditional DC servo systems in terms of control performance. With its many advantages such as small size, large torque, high efficiency, simple control, and convenient maintenance, the application field of permanent magnet synchronous motors is constantly expanding, especially in the field of automotive drive motors, showing great application potential.

[0003] In the control strategy of permanent magnet synchronous motor, the delay compensation coefficient is a crucial parameter. This coefficient will be set to a default value when the product is calibrated for the first time, which is used to compensate for the friction torque of the motor under test at high speed to achieve the goal of torque control 0, that is, to ensure that the no-load torque of the motor under test when not connected to the three-phase line is equal to the no-load torque when the tube is open.

[0004] However, due to hardware dispersion or other unavoidable factors, the default value of the delay compensation coefficient often cannot maintain the optimal state throughout the product cycle. When the deviation of the delay compensation coefficient is too large, it will directly affect the torque control accuracy and operating efficiency of the motor, and may even lead to a series of problems such as increased energy consumption and weakened power output, thereby affecting the performance and user experience of the entire vehicle. The delay compensation coefficient of the permanent magnet synchronous motor cannot meet the control requirements of the entire product cycle. This is mainly due to the fact that factors such as hardware dispersion cause the delay compensation coefficient to be unable to always maintain the optimal state during the first calibration. However, most of the existing control strategies do not take this into account, and do not perform dynamic adjustment or self-learning, so they cannot effectively deal with the situation where the delay compensation coefficient changes with hardware dispersion. In response to the above problems, the reasons are analyzed. The main reason is that there is a close correlation between the delay compensation coefficient of the permanent magnet synchronous motor and the hardware dispersion, but the existing control strategies have failed to fully consider this.

[0005] Therefore, how to achieve dynamic adjustment and self-learning of the permanent magnet synchronous motor delay compensation coefficient to adapt to the changes in hardware dispersion, thereby optimizing the control accuracy of the permanent magnet synchronous motor and extending its life cycle has become a problem that needs to be solved urgently. Summary of the invention

[0006] In view of the deficiencies of the above-mentioned prior art, the present invention provides a self-learning method for the delay compensation coefficient of a vehicle permanent magnet synchronous drive motor, which can dynamically adjust and self-learn the delay compensation coefficient of the permanent magnet synchronous motor to adapt to changes in hardware dispersion, thereby optimizing the control accuracy of the permanent magnet synchronous motor and extending its life cycle.

[0007] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0008] A self-learning method for a time delay compensation coefficient of a permanent magnet synchronous drive motor for a vehicle comprises the following steps:

[0009] S1, obtaining the no-load torque of the motor under test when the three-phase line is disconnected, and the no-load torque when the torque ring is open;

[0010] S2. Compare the no-load torque when the torque ring is open and when the three-phase line is disconnected. If the difference is greater than the preset difference, it is determined that the default value of the current delay compensation coefficient does not meet the requirements, and go to S3; if the difference is less than or equal to the preset value, return to S1;

[0011] S3, determine whether the self-learning conditions are met, if so, go to S4;

[0012] S4, perform self-learning at the motor inflection point speed and the maximum speed respectively, calculate the average electrical angles PwrAg1 and PwrAg2 at the two speeds, and calculate the self-learning delay compensation coefficient value in combination with the preset delay compensation initial value Delta and delay compensation gain Gain, and record it as the self-learning value;

[0013] S5. Compare the self-learning value with the current default value. If the difference is less than the preset difference value, update the default value and use the self-learning value as the new default value. If it is greater than or equal to the preset difference value, return to S4 and perform self-learning again.

[0014] Compared with the prior art, the present invention has the following beneficial effects:

[0015] 1. Using this method, first, obtain the no-load torque of the motor under test in different states through S1, which provides basic data for subsequent comparison and analysis. In S2, compare the no-load torque when the torque ring is open and when the three-phase line is disconnected to determine whether the default value of the current delay compensation coefficient meets the actual needs. If the difference is greater than the preset difference, it indicates that the current delay compensation coefficient may be inaccurate and needs to be adjusted.

[0016] Next, in S3, it is determined whether the current conditions meet the requirements for self-learning. This ensures that the self-learning process can be carried out in an appropriate environment and conditions, avoiding unnecessary errors and interference. If the self-learning conditions are met, enter S4, perform self-learning at the motor inflection point speed and the maximum speed, and calculate the average electrical angles PwrAg1 and PwrAg2 at the two speeds. These data are an important basis for the subsequent calculation of the self-learning delay compensation coefficient value. Afterwards, combined with the preset delay compensation initial value Delta and the delay compensation gain Gain, a new self-learning delay compensation coefficient value can be calculated. This value represents the optimal delay compensation coefficient under the current motor state.

[0017] Finally, in S5, the self-learning value is compared with the current default value. If the difference is less than the preset difference value, it means that the self-learning value is close enough to the default value and more accurately reflects the current state of the motor. Therefore, the default value can be updated and the self-learning value can be used as the new default value. If the difference is greater than or equal to the preset difference value, it means that there is a large difference between the self-learning value and the default value, and further adjustment or re-self-learning may be required. Therefore, return to S4 to re-self-learn to ensure the accuracy of the delay compensation coefficient.

[0018] 2. By performing self-learning at different speeds (inflection point speed and maximum speed) and calculating the average electrical angle, you can have a more comprehensive understanding of the motor's operating status. By performing self-learning at the motor's inflection point speed, you can obtain the motor's performance parameters at this critical speed, including the average electrical angle PwrAg1. This is of great significance for optimizing the motor's control strategy and improving its operating efficiency. By performing self-learning at the motor's maximum speed, you can obtain the motor's performance parameters under extreme conditions, including the average electrical angle PwrAg2. This helps to evaluate the motor's extreme performance and stability, and provides an important basis for optimizing motor design and control strategies. This method helps to avoid instability in the motor at a specific speed and improves the stability of the entire drive system.

[0019] In summary, this method can dynamically adjust and self-learn the delay compensation coefficient of the permanent magnet synchronous motor to adapt to the change of hardware dispersion. By real-time monitoring and adjusting the delay compensation coefficient, the control accuracy of the permanent magnet synchronous motor can be optimized, the error and loss can be reduced, and its life cycle can be extended. This self-learning method not only improves the operating efficiency of the motor, but also enhances its adaptability and stability.

[0020] Preferably, in S3, it is determined whether the self-learning conditions are met. If the following conditions are met at the same time, it is determined that the self-learning conditions are met: the torque loop is open; the self-learning mode is turned on; the current loop PI is applicable, the difference between the request Id and the feedback Id is less than the preset current difference, and the difference between the request Iq and the feedback Iq is less than the preset current difference; wherein, Id is the d-axis current, and Iq is the q-axis current.

[0021] Such a setting, the torque loop is open, means that the torque control system is in an open or adjustable state, which helps to accurately control and adjust the torque during the self-learning process. The self-learning mode is turned on, which ensures that the system is self-learning with the explicit intention of the user or operator, avoiding unnecessary or misoperation-induced self-learning. For example, when offline, the mode of the open tube request is different from the normal open tube mode. The normal open tube is generally a torque loop, and the self-learning mode has a separate mode. The difference between the request Id and the feedback Id, and the difference between the request Iq and the feedback Iq are less than the preset current difference. These two conditions ensure the accuracy of the current control. When the difference between the request current and the feedback current is less than the preset value, it can be considered that the current control has achieved sufficient accuracy, which provides a reliable basis for the system's self-learning.

[0022] Preferably, the preset current difference is 100A.

[0023] In this way, setting the preset current difference to 100A can ensure that the system performs self-learning under more stringent and accurate conditions, thereby optimizing control parameters, improving system performance and efficiency, and enhancing system stability and adaptability. This setting ensures the strictness and accuracy of current control, and helps the system obtain more reliable and stable data during the self-learning process.

[0024] Preferably, in S4, when self-learning is performed at the motor inflection point speed and the average electrical angle PwrAg1 is calculated, PwrAg1 is the average value of PwrAg within X operating cycles;

[0025] Perform self-learning at the highest motor speed and calculate the average electrical angle PwrAg2, which is the average value of PwrAg within X operating cycles;

[0026] Wherein, PwrAg is a preprocessed electrical angle calculated according to the corresponding d-axis voltage Ud and q-axis voltage Uq at a certain speed.

[0027] With this setup, the electrical angle information of the motor at different speeds can be accurately obtained by calculating the preprocessed electrical angle PwrAg based on the d-axis voltage Ud and the q-axis voltage Uq. This is crucial for accurate motor control, because the electrical angle is an important parameter in motor control that determines the direction and speed of motor rotation. Calculating the average electrical angles PwrAg1 and PwrAg2 over X operating cycles can further smooth and stabilize the electrical angle data and reduce errors caused by instantaneous fluctuations or noise. This helps improve the stability and accuracy of motor control.

[0028] By performing self-learning at different speeds and calculating the average electrical angle, the system can more accurately understand the performance of the motor under different operating conditions. This helps the system adjust the control strategy according to the real-time operating conditions to ensure that the motor can maintain stable operation under various conditions. The calculation of the average electrical angle can reduce errors caused by instantaneous fluctuations or noise, thereby improving the reliability of motor control. In addition, the motor performance parameters obtained through self-learning can be used for fault prediction and diagnosis, further improving the reliability and safety of the motor.

[0029] Preferably, when calculating the preprocessing electrical angle PwrAg, assume

[0030] If Ud≤0, then PwrAg=π-Ang;

[0031] If Ud>0 and Uq≥0, then PwrAg=Ang;

[0032] If Ud>0 and Uq<0, then PwrAg=2π+Ang.

[0033] Such a setting, 1. Accurate electrical angle calculation can be performed. By setting Ang = Uq / √(UdUd+UqUq), a preliminary electrical angle value can be calculated based on the ratio of the d-axis and q-axis voltages. This method utilizes the direct relationship between voltage and electrical angle in motor control, thereby achieving accurate calculation of electrical angle. According to the signs of Ud and Uq, the preliminary calculated electrical angle Ang is adjusted to obtain the final pre-processed electrical angle PwrAg. This adjustment takes into account the influence of the motor rotation direction and voltage polarity on the electrical angle, thereby ensuring the accuracy of the electrical angle calculation.

[0034] 2. It can adapt to different voltage combinations. When the d-axis voltage Ud is less than or equal to 0, the pre-processed electrical angle is calculated by PwrAg = π-Ang. This processing takes into account the situation that the motor may be in reverse rotation or stopped state, thereby ensuring the correctness of the electrical angle calculation. When the d-axis voltage Ud is greater than 0 and the q-axis voltage Uq is greater than or equal to 0, the electrical angle Ang obtained by preliminary calculation is directly used as the pre-processed electrical angle PwrAg. This reflects the situation that the motor rotates forward and the q-axis voltage is non-negative. When the d-axis voltage Ud is greater than 0 and the q-axis voltage Uq is less than 0, the pre-processed electrical angle is calculated by PwrAg = 2π+Ang. This processing takes into account the situation that the motor rotates forward but the q-axis voltage is negative, and ensures that the electrical angle is between 0 and 2π by adding 2π.

[0035] 3. It can improve the motor control performance. The accurate pre-processed electrical angle PwrAg can provide an important basis for the optimization of the motor control strategy. By calculating PwrAg in real time, the system can more accurately understand the operating status of the motor, thereby adjusting the control parameters to achieve better control effects. Accurate electrical angle calculation helps to reduce system fluctuations and instability caused by electrical angle errors. By using the above method to calculate PwrAg, the stability and reliability of the motor control system can be further improved.

[0036] Preferably, 80≤X≤150.

[0037] With this setting, the lower limit of the value range of X is 80, which means that at least 80 operating cycles of data are required to calculate the average electrical angle PwrAg. This ensures that there are enough data points to reflect the performance of the motor under different working conditions, thereby improving the accuracy and reliability of the calculation results. The upper limit of the value range of X is 150, which limits the number of data points used for calculation. This helps to avoid data redundancy, reduce the amount of calculation and storage requirements, while maintaining the stability and consistency of the calculation results.

[0038] By setting a reasonable range of X values, we can optimize the calculation efficiency while ensuring the calculation accuracy. Too many data points may increase the calculation complexity and time cost, while too few data points may lead to inaccurate calculation results. Therefore, the range of 80≤X≤150 is a reasonable compromise, which not only ensures the calculation accuracy but also improves the calculation efficiency.

[0039] Preferably, in S4, the calculation formula of the self-learning delay compensation coefficient value is (PwrAg2-PwrAg1+Delta)*Gain.

[0040] This setting can accurately compensate for the delay. PwrAg2 and PwrAg1 represent the average electrical angle of the motor at the maximum speed and the inflection point speed, respectively. Their difference (PwrAg2-PwrAg1) reflects the change in the electrical angle of the motor from the inflection point speed to the maximum speed. This change is directly related to the dynamic performance and delay characteristics of the motor. By adding Delta, the calculated change in electrical angle can be fine-tuned. Delta may represent some unconsidered factors or errors, such as sensor accuracy, circuit delay, etc. This adjustment helps to more accurately compensate for the actual delay. Gain, as a gain factor, can amplify or reduce the entire calculation result. This helps to adjust the amplitude of delay compensation according to actual needs to ensure that the compensation effect meets expectations.

[0041] Preferably, in S5, the preset difference value is 0.5 us.

[0042] With this setup, the preset difference value is set to 0.5us, which means that the system has a very low tolerance for time differences. This helps achieve more precise control and reduces system deviations caused by time errors. In situations where high-precision time control is required, such as motor control, signal processing, etc., the preset difference value of 0.5us can ensure the stability and optimization of system performance. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to make the purpose, technical solution and advantages of the invention more clear, the present invention will be further described in detail below with reference to the accompanying drawings, in which:

[0044] Figure 1 Flowchart of this method. DETAILED DESCRIPTION

[0045] The following is a further detailed description through specific implementation methods:

[0046] Example:

[0047] like Figure 1 As shown, this embodiment discloses a self-learning method for a time delay compensation coefficient of a permanent magnet synchronous drive motor for a vehicle, comprising the following steps:

[0048] S1. Obtain the no-load torque of the motor under test when the three-phase line is disconnected, and the no-load torque when the torque loop is open.

[0049] S2. Compare the no-load torque when the torque ring is open and when the three-phase line is disconnected. If the difference is greater than the preset difference, it is determined that the default value of the current delay compensation coefficient does not meet the requirements, and go to S3; if the difference is less than or equal to the preset value, return to S1, at this time, keep or store the default value of the current delay compensation coefficient unchanged. In specific implementation, the specific value of the preset difference can be set by those skilled in the art according to the actual use scenario (such as motor model, etc.), which will not be repeated here.

[0050] S3: Determine whether the self-learning conditions are met, and if so, go to S4.

[0051] Among them, it is judged whether the self-learning conditions are met. If the following conditions are met at the same time, it is judged that the self-learning conditions are met: the torque loop is open; the self-learning mode is turned on; the current loop PI is applicable, the difference between the request Id and the feedback Id is less than the preset current difference, and the difference between the request Iq and the feedback Iq is less than the preset current difference; wherein, Id is the d-axis current, and Iq is the q-axis current. In this way, the torque loop is open, which means that the torque control system is in an open or adjustable state, which helps to accurately control and adjust the torque during the self-learning process. The self-learning mode is turned on, which ensures that the system is self-learning with the clear intention of the user or operator, avoiding unnecessary or misoperation-induced self-learning. For example, when offline, the mode of the open tube request is different from the normal open tube mode. The normal open tube is generally a torque loop, and the self-learning mode has a separate mode. The difference between the request Id and the feedback Id, and the difference between the request Iq and the feedback Iq are less than the preset current difference. These two conditions ensure the accuracy of current control. When the difference between the requested current and the feedback current is less than the preset value, it can be considered that the current control has achieved sufficient accuracy, which provides a reliable basis for the self-learning of the system.

[0052] In specific implementation, the preset current difference is 100A. Setting the preset current difference to 100A can ensure that the system performs self-learning under more stringent and accurate conditions, thereby optimizing control parameters, improving system performance and efficiency, and enhancing system stability and adaptability. This setting ensures the strictness and accuracy of current control, and helps the system obtain more reliable and stable data during the self-learning process.

[0053] S4. Perform self-learning at the motor inflection point speed and maximum speed respectively, calculate the average electrical angles PwrAg1 and PwrAg2 at the two speeds, and calculate the self-learning delay compensation coefficient value in combination with the preset delay compensation initial value Delta and delay compensation gain Gain, and record it as the self-learning value.

[0054] Among them, when the motor inflection point speed is self-learned and the average electrical angle PwrAg1 is calculated, PwrAg1 is the average value of PwrAg within X operating cycles;

[0055] Perform self-learning at the highest motor speed and calculate the average electrical angle PwrAg2, which is the average value of PwrAg within X operating cycles;

[0056] Wherein, PwrAg is a preprocessed electrical angle calculated according to the corresponding d-axis voltage Ud and q-axis voltage Uq at a certain speed.

[0057] In this way, by calculating the preprocessed electrical angle PwrAg based on the d-axis voltage Ud and the q-axis voltage Uq, the electrical angle information of the motor at different speeds can be accurately obtained. This is crucial for the precise control of the motor, because the electrical angle is an important parameter in motor control, which determines the rotation direction and speed of the motor. Calculating the average electrical angles PwrAg1 and PwrAg2 within X operating cycles can further smooth and stabilize the electrical angle data and reduce errors caused by instantaneous fluctuations or noise. This helps to improve the stability and accuracy of motor control. By performing self-learning at different speeds and calculating the average electrical angle, the system can more accurately understand the performance of the motor under different operating conditions. This helps the system adjust the control strategy according to the real-time operating conditions to ensure that the motor can maintain stable operation under various conditions. The calculation of the average electrical angle can reduce errors caused by instantaneous fluctuations or noise, thereby improving the reliability of motor control. In addition, the motor performance parameters obtained through self-learning can be used for fault prediction and diagnosis, further improving the reliability and safety of the motor.

[0058] The calculation formula of the self-learning delay compensation coefficient value is (PwrAg2-PwrAg1+Delta)*Gain.

[0059] In this way, the delay can be accurately compensated. PwrAg2 and PwrAg1 represent the average electrical angle of the motor at the maximum speed and the inflection point speed, respectively. Their difference (PwrAg2-PwrAg1) reflects the change in the electrical angle of the motor from the inflection point speed to the maximum speed. This change is directly related to the dynamic performance and delay characteristics of the motor. By adding Delta, the calculated change in electrical angle can be fine-tuned. Delta may represent some unconsidered factors or errors, such as sensor accuracy, circuit delay, etc. This adjustment helps to compensate for the actual delay more accurately. Gain, as a gain factor, can amplify or reduce the entire calculation result. This helps to adjust the amplitude of delay compensation according to actual needs to ensure that the compensation effect meets expectations.

[0060] In specific implementation, 80≤X≤150.

[0061] In this way, the lower limit of the value range of X is 80, which means that at least 80 operating cycles of data are required when calculating the average electrical angle PwrAg. This ensures that there are enough data points to reflect the performance of the motor under different working conditions, thereby improving the accuracy and reliability of the calculation results. The upper limit of the value range of X is 150, which limits the number of data points used for calculation. This helps to avoid data redundancy, reduce the amount of calculation and storage requirements, while maintaining the stability and consistency of the calculation results. By setting a reasonable value range of X, the calculation efficiency can be optimized while ensuring the calculation accuracy. Too many data points may increase the calculation complexity and time cost, while too few data points may lead to inaccurate calculation results. Therefore, the value range of 80≤X≤150 is a reasonable compromise, which not only ensures the calculation accuracy but also improves the calculation efficiency.

[0062] When calculating the preprocessing electrical angle PwrAg, assume

[0063] If Ud≤0, then PwrAg=π-Ang;

[0064] If Ud>0 and Uq≥0, then PwrAg=Ang;

[0065] If Ud>0 and Uq<0, then PwrAg=2π+Ang.

[0066] Such a setting can accurately calculate the electrical angle. By setting Ang = Uq / √(UdUd+UqUq), a preliminary electrical angle value can be calculated based on the ratio of the d-axis and q-axis voltages. This method utilizes the direct relationship between voltage and electrical angle in motor control to achieve accurate calculation of the electrical angle. According to the signs of Ud and Uq, the preliminary calculated electrical angle Ang is adjusted to obtain the final pre-processed electrical angle PwrAg. This adjustment takes into account the influence of the motor rotation direction and voltage polarity on the electrical angle, thereby ensuring the accuracy of the electrical angle calculation. It can also adapt to different voltage combinations. When the d-axis voltage Ud is less than or equal to 0, the pre-processed electrical angle is calculated by PwrAg = π-Ang. This processing takes into account the situation where the motor may be in reverse rotation or stopped state, thereby ensuring the correctness of the electrical angle calculation. When the d-axis voltage Ud is greater than 0 and the q-axis voltage Uq is greater than or equal to 0, the preliminary calculated electrical angle Ang is directly used as the pre-processed electrical angle PwrAg. This reflects the situation where the motor rotates forward and the q-axis voltage is non-negative. When the d-axis voltage Ud is greater than 0 and the q-axis voltage Uq is less than 0, the preprocessed electrical angle is calculated by PwrAg = 2π + Ang. This processing takes into account the situation where the motor rotates forward but the q-axis voltage is negative, and ensures that the electrical angle is between 0 and 2π by adding 2π.

[0067] In addition, the motor control performance can be improved. Accurate pre-processed electrical angle PwrAg can provide an important basis for optimizing the motor control strategy. By calculating PwrAg in real time, the system can more accurately understand the operating status of the motor, thereby adjusting the control parameters to achieve better control effects. Accurate electrical angle calculation helps reduce system fluctuations and instability caused by electrical angle errors. By using the above method to calculate PwrAg, the stability and reliability of the motor control system can be further improved.

[0068] S5. Compare the self-learning value with the current default value. If the difference is less than the preset difference value, update the default value and use the self-learning value as the new default value. If it is greater than or equal to the preset difference value, return to S4 and perform self-learning again.

[0069] In specific implementation, the preset difference value is 0.5us. The preset difference value is set to 0.5us, which means that the system has a very low tolerance for time differences. This helps to achieve more precise control and reduce system deviations caused by time errors. In situations where high-precision time control is required, such as motor control, signal processing, etc., the preset difference value of 0.5us can ensure the stability and optimization of system performance.

[0070] Using this method, first, the no-load torque of the motor under test in different states is obtained through S1, which provides basic data for subsequent comparison and analysis. In S2, the no-load torque when the torque ring is open and when the three-phase line is disconnected can be compared to determine whether the default value of the current delay compensation coefficient meets the actual needs. If the difference is greater than the preset difference, it indicates that the current delay compensation coefficient may be inaccurate and needs to be adjusted. Then, in S3, it is determined whether the current conditions meet the requirements for self-learning. This ensures that the self-learning process can be carried out in an appropriate environment and conditions, avoiding unnecessary errors and interference. If the self-learning conditions are met, enter S4, perform self-learning at the motor inflection point speed and the maximum speed, and calculate the average electrical angles PwrAg1 and PwrAg2 at the two speeds. These data are important bases for the subsequent calculation of the self-learning delay compensation coefficient value. Afterwards, combined with the preset delay compensation initial value Delta and the delay compensation gain Gain, a new self-learning delay compensation coefficient value can be calculated. This value represents the optimal delay compensation coefficient under the current motor state. Finally, in S5, the self-learning value is compared with the current default value. If the difference is less than the preset difference value, it means that the self-learning value is close enough to the default value and more accurately reflects the current state of the motor. Therefore, the default value can be updated and the self-learning value can be used as the new default value. If the difference is greater than or equal to the preset difference value, it means that there is a large difference between the self-learning value and the default value, and further adjustment or re-self-learning may be required. Therefore, return to S4 to re-self-learn to ensure the accuracy of the delay compensation coefficient.

[0071] By performing self-learning at different speeds (inflection point speed and maximum speed) and calculating the average electrical angle, a more comprehensive understanding of the motor's operating status can be obtained. By performing self-learning at the motor's inflection point speed, the motor's performance parameters at this critical speed can be obtained, including the average electrical angle PwrAg1. This is of great significance for optimizing the motor's control strategy and improving its operating efficiency. By performing self-learning at the motor's maximum speed, the motor's performance parameters under extreme conditions can be obtained, including the average electrical angle PwrAg2. This helps to evaluate the motor's extreme performance and stability, and provides an important basis for optimizing motor design and control strategies. This method helps to avoid instability of the motor at a specific speed and improves the stability of the entire drive system.

[0072] This method can dynamically adjust and self-learn the delay compensation coefficient of the permanent magnet synchronous motor to adapt to the change of hardware dispersion. By real-time monitoring and adjusting the delay compensation coefficient, the control accuracy of the permanent magnet synchronous motor can be optimized, the error and loss can be reduced, and its life cycle can be extended. This self-learning method not only improves the operating efficiency of the motor, but also enhances its adaptability and stability.

[0073] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit the technical solution. Those skilled in the art should understand that those modifications or equivalent substitutions of the technical solution of the present invention that do not depart from the purpose and scope of the technical solution should be included in the scope of the claims of the present invention.

Claims

1. A self-learning method for delay compensation coefficient of a permanent magnet synchronous drive motor for a vehicle, characterized in that: The following steps are involved: S1, obtaining the no-load torque of the motor under test when the three-phase line is disconnected, and the no-load torque when the torque ring is open; S2, comparing the no-load torque when the torque ring is open and when the three-phase line is disconnected. If the difference is greater than the preset difference, it is determined that the default value of the current delay compensation coefficient does not meet the demand, and the process goes to S3; If the difference is less than or equal to the preset value, return to S1; S3, determine whether the self-learning conditions are met, if so, go to S4; S4, perform self-learning at the motor inflection point speed and the maximum speed respectively, calculate the average electrical angles PwrAg1 and PwrAg2 at the two speeds, and calculate the self-learning delay compensation coefficient value in combination with the preset delay compensation initial value Delta and delay compensation gain Gain, and record it as the self-learning value; S5. Compare the self-learning value with the current default value. If the difference is less than the preset difference value, update the default value and use the self-learning value as the new default value. If it is greater than or equal to the preset difference value, return to S4 and perform self-learning again.

2. The self-learning method for delay compensation coefficient of a permanent magnet synchronous drive motor for a vehicle according to claim 1, characterized in that: In S3, it is determined whether the self-learning conditions are met. If the following conditions are met at the same time, it is determined that the self-learning conditions are met: the torque loop is open; the self-learning mode is turned on; The current loop PI is applicable, the difference between the request Id and the feedback Id is less than the preset current difference, and the difference between the request Iq and the feedback Iq is less than the preset current difference; wherein, Id is the d-axis current, and Iq is the q-axis current.

3. The method for self-learning the delay compensation coefficient of a permanent magnet synchronous drive motor for a vehicle as claimed in claim 2, characterized in that: The preset current difference is 100A.

4. The method for self-learning the delay compensation coefficient of a permanent magnet synchronous drive motor for a vehicle as claimed in claim 1, characterized in that: In S4, when self-learning is performed at the motor inflection point speed and the average electrical angle PwrAg1 is calculated, PwrAg1 is the average value of PwrAg within X operating cycles; Perform self-learning at the highest motor speed and calculate the average electrical angle PwrAg2, which is the average value of PwrAg within X operating cycles; Wherein, PwrAg is a preprocessed electrical angle calculated according to the corresponding d-axis voltage Ud and q-axis voltage Uq at a certain speed.

5. The method for self-learning the time delay compensation coefficient of a permanent magnet synchronous drive motor for a vehicle as claimed in claim 4, characterized in that: When calculating the preprocessing electrical angle PwrAg, assume If Ud≤0, then PwrAg=π-Ang; If Ud>0 and Uq≥0, then PwrAg=Ang; If Ud>0 and Uq<0, then PwrAg=2π+Ang.

6. The method for self-learning the delay compensation coefficient of a permanent magnet synchronous drive motor for a vehicle as claimed in claim 4, characterized in that: 80≤X≤150。 7. The self-learning method for delay compensation coefficient of a permanent magnet synchronous drive motor for a vehicle as claimed in claim 4, characterized in that: In S4, the calculation formula of the self-learning delay compensation coefficient value is (PwrAg2-PwrAg1+Delta)*Gain.

8. The method for self-learning the delay compensation coefficient of a permanent magnet synchronous drive motor for a vehicle as claimed in claim 1, characterized in that: In S5, the preset difference value is 0.5us.

Images