Vector-controlled asynchronous motor parameter adaptive correction method
Through the adaptive correction method, the asynchronous motor parameters are obtained and fitted using the experimental method to establish a relationship curve between parameters and temperature, which solves the problem of inflexible changes in asynchronous motor parameters, improves the accuracy of vector control and the efficient operation of the motor.
Patent Information
- Application Number
- CN202411077226.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-07
- Publication Date
- 2024-11-08
AI Technical Summary
The existing asynchronous motor vector control methods are not flexible enough for motor parameter changes, resulting in inaccurate parameters, affecting control performance, and may cause the motor to heat up too quickly or even damage.
Adaptive correction method is adopted to obtain motor parameters through DC experiments, single-phase experiments and no-load experiments, and the trigonometric function and quadratic polynomials are fitted using the least squares method to establish a relationship curve between parameters and temperature, and correct motor parameters in real time.
Accurate correction of motor parameters is achieved, the accuracy of vector control is improved and the efficient operation of the motor is reduced, and the risk of calculation errors and motor damage is reduced.
Smart Images

Figure CN118920942A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of asynchronous motor parameter control, and in particular relates to a vector controlled asynchronous motor parameter adaptive correction method. Background Art
[0002] Asynchronous motors have the advantages of simple structure, stable reliability and low price. With the development of vector control technology, the control performance of asynchronous motors is no different from that of permanent magnet synchronous motors, and their applications are becoming more and more extensive. However, asynchronous motor vector control is highly dependent on motor parameters, and inaccurate parameters have a greater impact on motor performance.
[0003] Most of the existing motor parameter identification methods use DC experiments, no-load experiments, and locked-rotor experiments to perform offline identification to obtain various parameters of asynchronous motors, but this method has a large identification error. This is because most of the existing motor parameter identification methods perform static and dynamic motor self-learning, and do not effectively correct the parameter changes during the operation of the motor, and fail to obtain more accurate motor parameters. Inaccurate motor parameters will lead to inaccurate magnetic field orientation in vector control, seriously affecting control performance, failing to bring into play the advantages of vector control, and may also cause the motor to heat up too quickly, and in severe cases may burn out the motor. Summary of the invention
[0004] The technical problem to be solved by the present invention is to make up for the deficiencies of the prior art and provide a method for adaptively correcting parameters of an asynchronous motor under vector control.
[0005] To solve the above technical problems, the technical solution of the present invention is: A method for adaptively correcting parameters of a vector-controlled asynchronous motor comprises the following steps: Step 1: Get the motor stator temperature , execute steps 2 to 6 at the current temperature; Step 2: Determine the motor stator resistance through the DC experimental method ; Step 3: Determine the amplitude of the motor line voltage through the single-phase experimental method , initial phase , angular frequency and the amplitude of the phase current , initial phase , and calculate the first power factor , and then calculate the stator leakage inductance , rotor leakage inductance and rotor resistance ; First power factor The calculation formula is: ; Stator leakage inductance and rotor leakage inductance The calculation formula is: ; Rotor resistance The calculation formula is: ; Step 4: Determine the amplitude of the motor phase voltage through the no-load test method , initial phase , angular frequency and the amplitude of the phase current , initial phase , and calculate the second power factor , and then calculate the mutual inductance ; Second power factor The calculation formula is: ; Mutual Induction The calculation formula is: ; Step 5: Determine the optimal time constant of the motor rotor Define the initial value of the time constant , , ; Defining the time constant The value range is , and traverse and solve in this interval, the traversal step length is: , is the number of traversal points, then traverse The value of is: ; By changing from small to large The time constant is gradually changed by taking the value of The value of ,exist Record the corresponding torque output value. If the torque output value - rated torque ≤ setting error, stop changing. The value of The optimal time constant ; Step 6: Record the current stator temperature Stator resistance under , stator leakage inductance , rotor leakage inductance , rotor resistance and the optimal time constant ; Step 7: Heat up the stator by , and determine the current stator temperature Is it greater than the maximum allowable operating temperature? ,if , then loop through steps 2 to 6; if , then execute step eight; Step 8: Use the least squares quadratic polynomial fitting method to fit the stator temperature With stator resistance , stator temperature Stator leakage inductance , stator temperature Rotor leakage inductance , stator temperature With rotor resistance , stator temperature With the optimal time constant , the stator resistance , stator leakage inductance , rotor leakage inductance , rotor resistance and the optimal time constant curves that vary with temperature and apply these curves to motor vector control.
[0006] Furthermore, in step 2, the DC experiment is as follows: short-circuit the motor phase b and phase c, apply a DC voltage between the phase a and phase b windings, and measure the phase a current after the current stabilizes. and the voltage across phases a and b , then the stator resistance The calculation formula is: (1).
[0007] Furthermore, in step 3, the single-phase experiment is as follows: the control signals of the bridge arms of the motor phases b and c are the same, the motor is blocked, a sinusoidal voltage is applied between the motor phases a and b using a servo driver, and after the current is stable, the line voltages between the N phases a and b are detected in the order of the sampling time. and N a-phase currents The sampling time is not less than one positive voltage cycle, and the sampling time and Arrays, sample times, and Array.
[0008] Furthermore, in step 4, the no-load test is as follows: using a servo driver to control the motor to operate at a rated state in a constant voltage-frequency ratio V / F mode, and after the motor runs stably, detecting N a-phase voltages in the order of sampling time , a-phase current The sampling time is not less than one positive voltage cycle, and the sampling time and Arrays, sample times, and Array.
[0009] Furthermore, the sampling time in step 3 and Arrays, sample times, and The array is processed to obtain the amplitude of the motor line voltage in step 3 , initial phase , angular frequency and the amplitude of the phase current , initial phase ; The sampling time in step 4 and Arrays, sample times, and The array is processed to obtain the amplitude of the motor phase voltage in step 4 , initial phase , angular frequency and the amplitude of the phase current , initial phase .
[0010] Furthermore, the least squares fitting trigonometric function method is: Design the trigonometric functions that need to be fitted: (2) Among them, B, C, and D are the coefficients to be solved. is the angular frequency of the trigonometric function to be solved; Substitute the array into equation (2) and use the sum of squared deviations of trigonometric functions as the objective function: (3) in, Number the data points in the array. , are the horizontal and vertical coordinates of the data points, and the horizontal coordinate is the sampling time; the goal is to find a set of parameters Minimize the objective function value W; Since the trigonometric function in equation (3) is nonlinear, for the convenience of calculation, let , then the trigonometric function terms in the objective function become constants; Determine by formula (4): (4) in, and are the horizontal coordinate values of the data points corresponding to two adjacent zero-crossing values, and are the ordinate values of the two data points, namely , ; In order to minimize the objective function value W, the partial derivatives of the three coefficients B, C, and D of formula (3) are calculated, and the partial derivatives are set equal to 0. The matrix can be obtained by rearranging: (5) By calculating formula (5), we can get The specific values of the three coefficients B, C, and D are , , ; definition The value range of is: , traverse and solve in this interval, the traversal step length is (6) Where M is the number of traversal points, then traverse Value collection ,in, (7) in, ; Will , , , Substituting into formula (3), we get M objective function values W, and by comparison we get the minimum objective function value: , small objective function value Corresponding That is the trigonometric function Angular frequency ; Will Substituting into formula (5), we can get the trigonometric function The coefficient of , , ; According to the characteristics of trigonometric functions, we know that trigonometric functions The amplitude is: (8); Trigonometric functions The initial phase is: (9).
[0011] The beneficial effects that can be achieved by the present invention are: 1. Through the motor parameter and time fitting algorithm, the corresponding parameter and temperature change relationship curve is obtained for vector control, and the real-time parameter correction is performed according to the current motor temperature.
[0012] 2. The torque experiment is based on the vector control torque change trend to obtain a more accurate rotor time constant and correct the parameters obtained above.
[0013] 3. Furthermore, the least squares data fitting is performed on the measured voltage and current sinusoidal data to replace the Fourier transform analysis method, thereby reducing the amount of calculation and reducing the calculation error. BRIEF DESCRIPTION OF THE DRAWINGS
[0014] Figure 1 It is the structure diagram of three-phase asynchronous motor vector control; Figure 2 It is the DC experimental equivalent circuit; Figure 3 It is a single-phase experimental equivalent circuit; Figure 4 It is the equivalent circuit of the no-load experiment; DETAILED DESCRIPTION
[0015] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.
[0016] This embodiment is aimed at the three-phase asynchronous motor vector control technology, using the SVPWM modulation method to generate a control voltage vector to accurately control the motor speed, current, etc., mainly considering the impact of temperature on the asynchronous motor vector control. It includes a motor servo drive and an induction motor. The servo drive can detect the DC bus voltage. , three-phase current , , , position sensor and temperature sensor are installed inside the motor.
[0017] Figure 1 It is a typical connection method between voltage inverter and three-phase asynchronous motor, where It is the DC bus voltage, which is obtained by rectifying the industrial power supply AC380V; pass Figure 1 The three-phase inverter in the circuit can convert it into a voltage with variable frequency and amplitude. This step is called inversion. Figure 1 The inverter in the figure connects the three stator windings of the motor. The inverter has three bridge arms a, b, and c, so it is called a three-phase inverter. Each bridge arm has two switches. For example, there is a and Two switches, which respectively control the conduction and disconnection of the upper and lower bridge arms of phase a, and the switch states of the upper and lower bridge arms of the same phase are mutually exclusive, that is, the upper and lower bridge arms will not be turned on at the same time. This embodiment is mainly aimed at servo drive controlling induction motors, Figure 1 As a part of the servo drive hardware, the inverter provides the motor with a voltage with variable frequency and amplitude by controlling the switching state of each phase bridge arm.
[0018] A method for adaptively correcting parameters of a vector-controlled asynchronous motor comprises the following steps: Step 1: Get the motor stator temperature , execute steps 2 to 6 at the current temperature.
[0019] Step 2: Determine the motor stator resistance through the DC experimental method .
[0020] The DC experiment is: Figure 1 Middle a phase upper bridge arm The lower bridge arm is turned on because the switch states of the lower bridge arm and the upper bridge arm are mutually exclusive. Here, the lower bridge arm of phase a is That is, it is in the off state, making the lower bridge arm of phase b and c , The motor is conducting, that is, the motor phase b and phase c are short-circuited. A DC voltage is applied between the phase a and phase b windings. The equivalent circuit is as follows Figure 2 As shown, after the current stabilizes, measure the current of phase a and the voltage across phases a and b ( With bus voltage equal, is the DC voltage, which is the measured value), according to and The stator resistance is calculated as: (1) Least squares fitting trigonometric function method: Before conducting subsequent experiments, a least squares fitting trigonometric function method for data processing of the sampling array is designed as follows: Design the trigonometric functions that need to be fitted: (2) Among them, B, C, and D are the coefficients to be solved. is the angular frequency of the trigonometric function to be solved; Substitute the array into equation (2) and use the sum of squared deviations of trigonometric functions as the objective function: (3) in, Number the data points in the array. , are the horizontal and vertical coordinates of the data points, and the horizontal coordinate is the sampling time; the goal is to find a set of parameters Minimize the objective function value W; Since the trigonometric function in equation (3) is nonlinear, for the convenience of calculation, let , then the trigonometric function terms in the objective function become constants; Determine by formula (4): (4) in, and are the horizontal coordinate values of the data points corresponding to two adjacent zero-crossing values, and are the ordinate values of the two data points, namely , ; In order to minimize the objective function value W, the partial derivatives of the three coefficients B, C, and D of formula (3) are calculated, and the partial derivatives are set equal to 0. The matrix can be obtained by rearranging: (5) By calculating formula (5), we can get The specific values of the three coefficients B, C, and D are , , ; definition The value range of is: , traverse and solve in this interval, the traversal step length is (6) Where M is the number of traversal points, then traverse Value collection ,in, (7) in, ; Will , , , Substituting into formula (3), we get M objective function values W, and by comparison we get the minimum objective function value: , small objective function value Corresponding That is the trigonometric function Angular frequency ; Will Substituting into formula (5), we can get the trigonometric function The coefficient of , , ; According to the characteristics of trigonometric functions, we know that trigonometric functions The amplitude is: (8); Trigonometric functions The initial phase is: (9).
[0021] In this embodiment, a trigonometric function fitting algorithm is used instead of Fourier transform analysis, which can reduce the amount of calculation and reduce errors.
[0022] Step 3: Solve the leakage inductance and rotor resistance through the single-phase experimental method.
[0023] The single-phase experiment is as follows: make the control signals of the motor b and c phase bridge arms the same, the motor is blocked, and a servo driver is used to apply a sinusoidal voltage between the motor a and b phases. After the current stabilizes, the line voltages between N a and b phases are detected in the sampling time sequence. and N a-phase currents The sampling time is not less than one positive voltage cycle, and the sampling time and Arrays, sample times, and Array. (Here, the amplitude and frequency of the sinusoidal AC voltage are controllable and are output by the servo driver, for example, 380V50Hz. The sampling device is the servo driver, and the driver can collect the line voltage Vab2 and the phase current signal Ia2. For example, the sampling interval is 1ms, the sampling time is 300ms, the signal frequency is 50Hz, and the array length is 300).
[0024] The above sampling time and Arrays, sample times, and The array is processed to obtain the amplitude of the motor line voltage in step S3 , initial phase , angular frequency and the amplitude of the phase current , initial phase .
[0025] Calculate the first power factor : The equivalent circuit of the single-phase asynchronous motor experiment is as follows Figure 3 As shown in the figure (which is the effective circuit of the three-phase asynchronous motor under this control state), the stator leakage inductance is calculated , rotor leakage inductance and rotor resistance ; Stator leakage inductance and rotor leakage inductance The calculation formula is: (10) Rotor resistance The calculation formula is: (11) Step 4: Solve the mutual inductance through the no-load test method.
[0026] The no-load test is as follows: the servo driver is used to control the motor to run at rated state in a constant voltage-frequency ratio V / F mode. After the motor runs stably, the N a-phase voltages are detected in the sampling time sequence. , a-phase current The sampling time is not less than one positive voltage cycle, and the sampling time and Arrays, sample times, and Array.
[0027] The above sampling time and Arrays, sample times, and The array is processed to obtain the amplitude of the motor phase voltage in step S4 , initial phase , angular frequency and the amplitude of the phase current , initial phase .
[0028] Calculate the second power factor : according to Figure 4 The equivalent circuit shown in Figure 1 can be used to calculate the mutual inductance. : (12) Step 5: Determine the optimal time constant of the motor rotor .
[0029] The accurate motor rotor time constant can more accurately calculate the direction of the motor magnetic field in space during vector control, realize independent and efficient control of the motor excitation component and torque component, and improve the motor control performance. Implementation method: Use the servo drive to control the output of the rated torque command value Teref, and use the torque sensor to measure the actual output torque of the motor (at this time, the measured torque deviates from the rated torque).
[0030] Define the initial value of the time constant : (13) in, .
[0031] Defining the time constant The value range is , and traverse and solve in this interval, the traversal step length is: (14) in, is the number of traversal points, then traverse The value of is: (15) in, .
[0032] By changing from small to large The time constant is gradually changed by taking the value of The value of ,exist Record the corresponding torque output value. If the torque output value - rated torque ≤ the setting error (the setting error is set by the technicians according to the actual situation), stop changing. The value of The optimal time constant .
[0033] Step 6: Record the current stator temperature Stator resistance under , stator leakage inductance , rotor leakage inductance , rotor resistance and the optimal time constant .
[0034] Step 7: Rapidly heat up the stator by , and determine the current stator temperature Is it greater than the maximum allowable operating temperature? ( It can be known from the motor technical parameters) if , then loop through steps 2 to 6; if , then execute step eight.
[0035] According to the motor thermal power formula It can be seen that the larger the motor current, the faster the heat is generated. The servo driver is used to control the motor to run for a period of time to increase the motor temperature by △T (for example, △T=10℃).
[0036] Step 8: Use the least squares quadratic polynomial fitting method to fit the stator temperature With stator resistance , stator temperature Stator leakage inductance , stator temperature Rotor leakage inductance , stator temperature With rotor resistance , stator temperature With the optimal time constant , the stator resistance , stator leakage inductance , rotor leakage inductance , rotor resistance and the optimal time constant curves (quadratic polynomials) that vary with temperature and apply these curves (quadratic polynomials) to motor vector control.
[0037] Assume that the polynomial to be fitted is: (16) in , , is the coefficient to be solved, and the sum of squares of deviations from the sample point to the curve in equation (16) is as follows: (17) In order to minimize the deviation, the partial derivatives of the coefficients of formula (17) are calculated, and the partial derivatives are equal to zero. The following matrix form can be obtained by rearranging: (18) Solving the matrix (18) yields the coefficient matrix, and we get the coefficients , , . Take rotor resistance as an example: , T is the current temperature of the motor, Rr is the rotor resistance that changes with temperature, and similar expressions can be obtained for other motor parameters.
[0038] At this point, the motor parameter identification is completed, and the relationship between each parameter and temperature is obtained. The temperature-based motor parameter expressions are substituted into the servo drive to perform motor vector control. This can achieve high-precision control and high-efficiency operation of the motor at different temperatures and realize adaptive parameter adjustment.
Claims
1. A method for adaptively correcting parameters of an asynchronous motor under vector control, characterized in that: The steps include: Step 1: Get the motor stator temperature , execute steps 2 to 6 at the current temperature; Step 2: Determine the motor stator resistance through the DC experimental method ; Step 3: Determine the amplitude of the motor line voltage through the single-phase experimental method , initial phase , angular frequency and the amplitude of the phase current , initial phase , and calculate the first power factor , and then calculate the stator leakage inductance , rotor leakage inductance and rotor resistance ; First power factor The calculation formula is: ; Stator leakage inductance and rotor leakage inductance The calculation formula is: ; Rotor resistance The calculation formula is: ; Step 4: Determine the amplitude of the motor phase voltage through the no-load test method , initial phase , angular frequency and the amplitude of the phase current , initial phase , and calculate the second power factor , and then calculate the mutual inductance ; Second power factor The calculation formula is: ; Mutual Induction The calculation formula is: ; Step 5: Determine the optimal time constant of the motor rotor Define the initial value of the time constant , , ; Defining the time constant The value range is , and traverse and solve in this interval, the traversal step length is: , is the number of traversal points, then traverse The value of is: ; By changing from small to large The time constant is gradually changed by taking the value of The value of ,exist Record the corresponding torque output value. If the torque output value - rated torque ≤ setting error, stop changing. The value of The optimal time constant ; Step 6: Record the current stator temperature Stator resistance under 、Stator leakage inductance , rotor leakage inductance , rotor resistance and the optimal time constant ; Step 7: Heat up the stator by , and judge the current stator temperature Is it greater than the maximum allowable operating temperature? ,if , then loop through steps 2 to 6; if , then execute step eight; Step 8: Use the least squares quadratic polynomial fitting method to fit the stator temperature With stator resistance , stator temperature Stator leakage inductance , stator temperature Rotor leakage inductance , stator temperature With rotor resistance , stator temperature With the optimal time constant , the stator resistance 、Stator leakage inductance , rotor leakage inductance , rotor resistance and the optimal time constant curves that vary with temperature and apply these curves to motor vector control.
2. The method for adaptively correcting parameters of an asynchronous motor controlled by vector control according to claim 1 is characterized in that: In step 2, the DC test is as follows: short-circuit the motor phases b and c, apply a DC voltage between the phase a and phase b windings, and measure the current of phase a after the current stabilizes. and the voltage across phases a and b , then the stator resistance The calculation formula is: (1).
3. The method for adaptively correcting parameters of an asynchronous motor controlled by vector control according to claim 1 is characterized in that: In step 3, the single-phase experiment is as follows: the control signals of the motor b and c phase bridge arms are the same, the motor is blocked, and a servo driver is used to apply a sinusoidal voltage between the motor a and b phases. After the current is stable, the line voltages between N a and b phases are detected in the sampling time sequence. and N a-phase currents The sampling time is not less than one positive voltage cycle, and the sampling time and Arrays, sample times, and Array.
4. The method for adaptively correcting parameters of an asynchronous motor controlled by vector control according to claim 3 is characterized in that: In step 4, the no-load test is as follows: the servo driver is used to control the motor to operate at rated state in a constant voltage-frequency ratio V / F mode, and after the motor runs stably, the N a-phase voltages are detected in the sampling time sequence. , a-phase current The sampling time is not less than one positive voltage cycle, and the sampling time and Arrays, sample times, and Array.
5. The method for adaptively correcting parameters of an asynchronous motor controlled by vector control according to claim 4 is characterized in that: The sampling time in step 3 is compared with the trigonometric function using the least squares method. Arrays, sample times, and The array is processed to obtain the amplitude of the motor line voltage in step 3 , initial phase , angular frequency and the amplitude of the phase current , initial phase ; The sampling time in step 4 is Arrays, sample times, and The array is processed to obtain the amplitude of the motor phase voltage in step 4 , initial phase , angular frequency and the amplitude of the phase current , initial phase .
6. The method for adaptively correcting parameters of an asynchronous motor controlled by vector control according to claim 4 is characterized in that: The least squares fitting trigonometric function method is: Design the trigonometric functions that need to be fitted: (2) Among them, B, C, and D are the coefficients to be solved. is the angular frequency of the trigonometric function to be solved; Substitute the array into equation (2) and use the sum of squared deviations of trigonometric functions as the objective function: (3) in, Number the data points in the array. , are the horizontal and vertical coordinates of the data points, and the horizontal coordinate is the sampling time; the goal is to find a set of parameters Minimize the objective function value W; Since the trigonometric function in equation (3) is nonlinear, for the convenience of calculation, let , then the trigonometric function terms in the objective function become constants; Determine by formula (4): (4) in, and are the horizontal coordinate values of the data points corresponding to two adjacent zero-crossing values, and are the ordinate values of the two data points, namely , ; In order to minimize the objective function value W, the partial derivatives of the three coefficients B, C, and D of formula (3) are calculated, and the partial derivatives are set equal to 0. The matrix can be obtained by rearranging: (5) By calculating formula (5), we can get The specific values of the three coefficients B, C, and D are , , ; definition The value range of is: , traverse and solve in this interval, the traversal step length is (6) Where M is the number of traversal points, then traverse Value collection ,in, (7) in, ; Will , , , Substituting into formula (3), we get M objective function values W, and by comparison we get the minimum objective function value: , small objective function value Corresponding That is the trigonometric function Angular frequency ; Will Substituting into formula (5), we can get the trigonometric function The coefficient of , , ; According to the characteristics of trigonometric functions, we know that trigonometric functions The amplitude is: (8); Trigonometric functions The initial phase is: (9)。
Citation Information
Cited By
Parameter calibration method, device and equipment of asynchronous motor and storage medium
CN119535207A
Parameter calibration method, device, equipment and storage medium for asynchronous motor
CN119535207B
