Method for correcting motor estimated torque

Abstract

The application discloses a motor estimated torque correction method, characterized in that comprising the following steps: S1: presetting multiple rotating speed sampling points of the motor, acquiring motor parameters of each sampling point, calculating a magnetic chain according to the motor parameters through a magnetic chain calculation formula, and calibrating a dynamic magnetic chain; S2: when the motor is in a preset working condition, different input combinations of AC and DC axis currents are given, and an estimated torque value is obtained; S4: three calibration data tables are output, which are used for correcting the estimated torque in the motor control software, and finally an optimized estimated torque is output. The application adopts a formula method of non-constant inductance difference and an actual estimated torque coefficient compensation mode to improve the torque estimation accuracy, correct the torque estimation, improve the torque estimation accuracy, and better support the vehicle torque control and system decision.

Description

Technical Field

[0001] This invention relates to the field of new energy vehicle technology, and in particular to a method for torque estimation correction in the field of permanent magnet synchronous motor control. Background Technology

[0002] Pure electric vehicles move by driving wheels with electric motors, with permanent magnet synchronous motors (PMSMs) currently being the primary application. The core of PMSM control is to maximize torque output and achieve higher speeds within the constraints of DC voltage and voltage utilization. Therefore, it's necessary to find a reasonable combination of AC and DC axis current commands for control to achieve the desired performance goals. Typically, PMSMs undergo bench calibration before vehicle application. The most important task is to test the relationship between the controller's output current and output torque on the bench while fully utilizing the motor's performance. While responding to the torque request from the vehicle controller, the motor controller must estimate the actual torque, compare it with the required torque, and feed this feedback to the vehicle controller for decision-making and control of vehicle commands. Current methods for estimating PMSM torque primarily involve directly calculating the output torque from monitored motor parameters. However, the motor's operating power cannot be accurately calculated, and internal temperature variations lead to inaccurate calculations. Furthermore, the electromagnetic torque formula for PMSMs is an ideal formula, expressed in an AC / DC axis system as follows: Currently, a preferred method for estimating the torque of permanent magnet synchronous motors is to use a non-uniform flux linkage calculation method to obtain the motor torque through a formula. However, this formula is idealized, and motor parameters change with the motor's usage. Furthermore, the internal temperature of the motor fluctuates over time, leading to inaccurate torque estimates and affecting the assessment of vehicle driving conditions. Chinese patent application CN202110681444.6 discloses a method, system, and vehicle for estimating the torque of an electric vehicle motor. This patent addresses the fact that the ideal formula is affected by motor operating conditions. Motor parameters change with usage, temperature, and other factors, all of which impact the accuracy of the estimated torque. Its key feature is the acquisition and correction of flux linkage changes to estimate the motor torque.

[0003] However, the electric vehicle motor torque estimation method, system, and vehicle mentioned above only consider the influence of magnetic flux on torque. Therefore, the torque estimation accuracy of the technical solution disclosed in Chinese patent application number CN202110681444.6 is not high. Summary of the Invention

[0004] This invention primarily addresses the problem of poor torque estimation accuracy in existing motor control technologies, providing a method for correcting estimated motor torque. Most current torque estimation methods for permanent magnet synchronous motors (PMSMs) reference the electromagnetic torque formula for PMSMs, expressed in a direct-orthogonal axis system as follows: This formula is an ideal formula, requiring precise motor parameters to ensure the estimated torque matches the target torque. Currently, a superior method for estimating torque in permanent magnet synchronous motors uses non-uniform flux linkage calculations; however, this method neglects the impact of variations in the quadrature-direct axis inductance difference in the reluctance torque component. In previous calibration methods, both quadrature-direct axis inductances were constant, which is insufficient for torque estimation under different input conditions. This invention improves torque estimation accuracy by employing a formula based on non-constant inductance differences combined with compensation using actual torque estimation coefficients.

[0005] The above-mentioned technical problem of the present invention is mainly solved by the following technical solution: a method for correcting the estimated torque of a motor, comprising the following steps:

[0006] S1: Multiple speed sampling points are preset in the motor, the motor parameters of each sampling point are obtained, and the flux is calculated according to the flux calculation formula based on the motor parameters to calibrate the dynamic flux.

[0007] S2: When the motor is under preset operating conditions, given different combinations of AC and DC axis current inputs, the estimated torque value is obtained through the formula. The difference between the quadrature and direct axes inductance is calculated separately, and the quadrature and direct axis inductance is calibrated; where Te is the motor torque and Pn is the number of motor pole pairs. For the composite flux linkage, Iq is the Q-axis current, Id is the D-axis current, Lq is the Q-axis inductance, and Ld is the D-axis inductance.

[0008] S3: Estimation of torque compensation coefficient calibration;

[0009] S4: Outputs three calibration data tables for correction of the estimated torque in the motor control software, and finally outputs the optimized estimated torque.

[0010] Preferably, step S1 specifically includes the following steps:

[0011] S11: The controller is connected to low voltage, and the motor is in a free rotation state while being driven;

[0012] S12: Divide the motor speed into ten points based on the motor's maximum speed;

[0013] S13: The dynamometer drives the motor to rotate in the positive direction of the motor to the evenly distributed target speed point, and records the line voltage displayed on the power analyzer at this time;

[0014] S14: Record the line voltage at each target rotational speed point according to step S13, and finally obtain a composite flux linkage and a three-dimensional table with temperature and rotational speed as the horizontal and vertical axes. Calculate the corresponding flux linkage value by substituting the relevant temperature and rotational speed information into the flux linkage calculation formula, and fill the calculated flux linkage data into the calibration table.

[0015] Preferably, step S2 specifically includes the following steps:

[0016] S21: When the controller connects to the low voltage and high voltage, the dynamometer drives the motor to rotate in the positive direction (the motor speed is 30% of the peak speed);

[0017] S22: Enter the corresponding quadrature and direct axis current values ​​from the current distribution calibration into the calibration tool software;

[0018] S23: Set the calibration tool software to the quadrature-direct axis current distribution mode and set the target current;

[0019] S24: View and record the motor torque value displayed on the test bench and the estimated motor torque value calculated internally by the motor control software; S25: Calculate the quadrature-direct axis inductance difference using the quadrature-direct axis inductance difference formula and fill it into the quadrature-direct axis inductance difference calibration table.

[0020] Preferably, step S3 specifically includes the following steps:

[0021] S31: Set all data in the estimated torque compensation coefficient table in the calibration tool software to 1;

[0022] S32: The dynamometer drives the motor to rotate in the positive direction of the motor.

[0023] S33: Set the motor operating mode to current input in the calibration tool software;

[0024] S34: Given different target currents, view and record the motor torque value displayed on the test bench and the estimated motor torque value calculated internally by the motor control software;

[0025] S35: Sum the test bench torque value recorded in step S34 with the no-load torque at that speed to obtain the actual output torque of the motor. Calculate the estimated torque compensation coefficient using the estimated torque compensation coefficient calculation formula and fill it into the estimated torque coefficient table.

[0026] Preferably, the formula for calculating the estimated torque compensation coefficient in step S3 is as follows:

[0027] Factor_Compensation=Trq_TestRig / Trq_Cal

[0028] In the formula, Factor_Compensation is the estimated torque compensation coefficient, Trq_TestRig is the actual torque of the motor, and Trq_Cal is the estimated torque of the motor.

[0029] Preferably, the magnetic flux linkage calculation formula in step S1 is:

[0030]

[0031] In the formula, For the composite flux linkage, Kt is the temperature coefficient, Pn is the number of pole pairs of the motor, Uab is the line voltage of the motor, and Nspeed is the motor speed.

[0032] The beneficial effects of the present invention are: According to the original formula for electromagnetic torque of motor, the inductance of both the direct and quadrature axes is constant, which cannot well cope with the torque estimation under different input conditions. The present invention provides a method for correcting the torque estimation of motor, (1) using a formula method with non-constant inductance difference, (2) and a method of compensation for actual estimated torque coefficient, to improve the accuracy of torque estimation. The accuracy of torque estimation will be improved, which can better support the torque control of the whole vehicle and system decision-making. Attached Figure Description

[0033] Figure 1 This is a flowchart of the motor torque estimation method according to an embodiment of the present invention. Detailed Implementation

[0034] Firstly, the calculation of the non-constant inductance difference involves the influence of input variables such as motor temperature and speed. Given that the stator resistance is affected by temperature, and considering the influence of speed on flux linkage and inductance, theoretically calculating the quadrature and direct-axis inductances under different input conditions (including different currents, temperatures, and speeds) is quite cumbersome. Therefore, a method for calibrating the quadrature and direct-axis inductance difference is proposed. This method is based on actual motor calibration data, offering high accuracy and good dynamic response. The derivation process from the torque calculation formula is as follows: Based on the original motor electromagnetic torque formula: In the formula, Te is the motor torque, and Pn is the number of pole pairs of the motor. To synthesize the magnetic flux linkage, Iq is the Q-axis current, Id is the D-axis current, Lq is the Q-axis inductance, and Ld is the D-axis inductance. The original formula can be transformed into: The difference between the quadrature and direct-axis inductances can be considered as being affected only by the quadrature and direct-axis currents under the current motor state, speed, and temperature conditions. Different actual torque values ​​are obtained based on different combinations of quadrature and direct-axis current inputs. The difference between the quadrature and direct-axis inductances is calculated using the above formula, and a three-dimensional table of the difference between the quadrature and direct-axis inductances and currents is fitted. More optimally, the combined flux linkage φf is affected by speed and temperature. This part includes the influence of temperature on the stator resistance. The line voltage under the current conditions can be measured at different speeds and temperatures, and a three-dimensional table relating flux linkage, speed, and temperature can be fitted using the following formula. In the formula, For the composite flux linkage, Kt is the temperature coefficient, Pn is the number of motor pole pairs, Uab is the motor line voltage, and Nspeed is the motor speed. Another method mentioned above for actual torque estimation coefficient compensation is to recalibrate the estimated torque by correcting the ratio between the estimated torque and the bench-measured torque. The theoretical formula is as follows: Factor_Compensation = Trq_TestRig / Trq_Cal, where Factor_Compensation is the estimated torque compensation coefficient, Trq_TestRig is the actual motor torque, and Trq_Cal is the estimated motor torque. This method can use a fixed current as a reference to obtain the estimated torque coefficient compensation value under different target current conditions. Even better, other input conditions such as speed and temperature can be added to obtain multi-dimensional compensation coefficient values. The three tables obtained from the above two parts will be used to correct the estimated torque in the motor control software, ultimately outputting the optimized estimated torque. The method proposed in this invention for improving the torque estimation accuracy of permanent magnet synchronous motors is achieved through the calibration of the quadrature-direct axis inductance difference Ld-Lq and the actual estimated torque compensation coefficient.

[0035] Specific implementation example: A method for correcting the estimated torque of a motor, such as... Figure 1 As shown, it includes:

[0036] S1: Calibration of dynamic flux linkage; specifically: the controller is connected to low voltage (high voltage disconnected), ensuring the motor remains in free rotation while being driven; based on the motor's maximum speed, the motor speed is divided into ten sampling points; the dynamometer drives the motor in the positive direction, and when the motor speed reaches the ten equally divided target speed points, the line voltage displayed on the power analyzer is recorded sequentially, ultimately obtaining a composite flux linkage and a three-dimensional table with temperature and speed as the horizontal and vertical axes. Using the flux linkage calculation formula, the corresponding flux linkage value is calculated by substituting the relevant temperature and speed information. The calculated flux linkage data is then filled into the calibration table. The flux linkage calculation formula is:

[0037]

[0038] In the formula, To synthesize the flux linkage, Kt is the temperature coefficient, Pn is the number of motor pole pairs, Uab is the motor line voltage, and Nspeed is the motor speed. Finally, the calculated dynamic flux linkage data is filled into the calibration table.

[0039] S2: Calibration of the direct-axis inductance difference; specifically: the controller is connected to both low and high voltage, and the dynamometer drives the motor to rotate in the positive direction (motor speed is 30% of peak speed); the corresponding direct-axis current values ​​Id and Iq from the current distribution calibration are entered into the calibration tool software; the calibration tool software is set to direct-axis current distribution mode, a target current is given, and the motor torque value displayed on the test bench and the estimated motor torque value calculated internally by the motor control software are viewed and recorded; the direct-axis inductance difference is calculated using the direct-axis inductance difference formula and entered into the direct-axis inductance difference calibration table, the formula is as follows:

[0040]

[0041] The formula for the difference in inductance between the direct and quadrature axes is as follows: It is based on the original electromagnetic torque formula of the motor The transformed formula shows that the difference between the quadrature and direct-axis inductances can be considered as being affected only by the quadrature and direct-axis currents under the current motor state, speed, and temperature conditions. Te is the motor torque, and Pn is the number of motor pole pairs. For the composite flux linkage, Iq is the Q-axis current, Id is the D-axis current, Lq is the Q-axis inductance, and Ld is the D-axis inductance.

[0042] S3: Calibration of the estimated torque compensation coefficient. This involves recalibrating the estimated torque by correcting the ratio between the estimated torque and the actual torque measured on the test bench. Specifically: Ensure all data in the estimated torque compensation coefficient table in the calibration tool software is set to 1. Use the dynamometer to drive the motor in the positive direction (motor speed at 30% of peak speed). In the calibration tool software, set the motor operating mode to current input. Given different target currents, observe and record the motor torque values ​​displayed on the test bench and the estimated torque values ​​calculated internally by the motor control software. Sum the recorded test bench torque value with the no-load torque at that speed to obtain the actual output torque of the motor. Calculate the estimated torque compensation coefficient using the formula and enter it into the estimated torque coefficient table. The formula is as follows:

[0043] Factor_Compensation=Trq_TestRig / Trq_Cal

[0044] Among them, Factor_Compensation is the estimated torque compensation coefficient, Trq_TestRig is the actual torque of the motor, and Trq_Cal is the estimated torque of the motor.

[0045] S4: Outputs three calibration data tables for correction of the estimated torque in the motor control software, and finally outputs the optimized estimated torque.

[0046] The embodiments described above are merely preferred embodiments of the present invention and are not intended to limit the present invention in any way. Other variations and modifications are possible without departing from the technical solutions described in the claims.

Claims

1. A method for correcting the estimated torque of a motor, characterized in that, Includes the following steps: S1: Multiple speed sampling points are preset in the motor, the motor parameters of each sampling point are obtained, and the flux is calculated according to the flux calculation formula based on the motor parameters to calibrate the dynamic flux. S2: When the motor is under preset operating conditions, given different combinations of AC and DC axis current inputs, the estimated torque value is obtained through the formula. The difference between the quadrature and direct axes inductances is calculated separately, and the difference is calibrated; where Te is the motor torque and Pn is the number of motor pole pairs. For the composite flux linkage, Iq is the Q-axis current, Id is the D-axis current, Lq is the Q-axis inductance, and Ld is the D-axis inductance. S3: Calibration of estimated torque compensation coefficient, wherein the estimated torque compensation coefficient is obtained by the ratio of the actual output torque of the motor to the estimated torque of the motor; S4: Outputs three calibration data tables for correction of the estimated torque in the motor control software, and finally outputs the optimized estimated torque.

2. The method for correcting the estimated torque of a motor according to claim 1, characterized in that, Step S1 specifically includes the following steps: S11: The controller is connected to low voltage, and the motor is in a free rotation state while being driven; S12: Divide the motor speed into ten points based on the motor's maximum speed; S13: The dynamometer drives the motor to rotate in the positive direction of the motor to the evenly distributed target speed point, and records the line voltage displayed on the power analyzer at this time; S14: Record the line voltage at each target rotational speed point according to step S13, and finally obtain a composite flux linkage and a three-dimensional table with temperature and rotational speed as the horizontal and vertical axes. Calculate the corresponding flux linkage value by substituting the relevant temperature and rotational speed information into the flux linkage calculation formula, and fill the calculated flux linkage data into the calibration table.

3. The method for correcting the estimated torque of a motor according to claim 1 or 2, characterized in that, Step S2 specifically includes the following steps: S21: When the controller connects the low voltage and high voltage, the dynamometer drives the motor to rotate in the positive direction of the motor. S22: Enter the corresponding quadrature and direct axis current values ​​from the current distribution calibration into the calibration tool software; S23: Set the calibration tool software to the quadrature-direct axis current distribution mode and set the target current; S24: View and record the motor torque value displayed on the test bench and the estimated motor torque value calculated internally by the motor control software; S25: Calculate the difference between the quadrature and direct axes inductance using the quadrature-direct axis inductance difference formula and fill it into the quadrature-direct axis inductance difference calibration table.

4. The method for correcting the estimated torque of a motor according to claim 3, characterized in that, Step S3 specifically includes the following steps: S31: Set all data in the estimated torque compensation coefficient table in the calibration tool software to 1; S32: The dynamometer drives the motor to rotate in the positive direction of the motor. S33: Set the motor operating mode to current input in the calibration tool software; S34: Given different target currents, view and record the motor torque value displayed on the test bench and the estimated motor torque value calculated internally by the motor control software; S35: Sum the test bench torque value recorded in step S34 with the no-load torque at that speed to obtain the actual output torque of the motor. Calculate the estimated torque compensation coefficient using the estimated torque compensation coefficient calculation formula and fill it into the estimated torque coefficient table.

5. The method for correcting the estimated torque of a motor according to claim 4, characterized in that, The formula for calculating the estimated torque compensation coefficient is as follows: Factor_Compensation=Trq_TestRig / Trq_Cal In the formula, Factor_Compensation is the estimated torque compensation coefficient, Trq_TestRig is the actual torque of the motor, and Trq_Cal is the estimated torque of the motor.

6. The method for correcting the estimated torque of a motor according to claim 1, characterized in that, The formula for calculating magnetic flux linkage is: In the formula, For the composite flux linkage, Kt is the temperature coefficient, Pn is the number of pole pairs of the motor, Uab is the line voltage of the motor, and Nspeed is the motor speed.

Images