Dual compensation method for current measurement error of permanent magnet synchronous motor and non-linear error of inverter

Abstract

According to the dual compensation method for the current measurement error of the permanent magnet synchronous motor and the non-linear error of the inverter, the non-linear disturbance observer and the low-pass filter are organically combined, and the current measurement error can be accurately estimated and compensated; meanwhile, on the basis of integral comparison of the voltage actually acting on the motor and the reference voltage, the invention provides an effective compensation means suitable for voltage errors caused by nonlinearity of the inverter. According to the method, motor rotating speed information and A / B / C three-phase current symbol judgment are not needed in implementation, current periodic fluctuation and insufficient torque output caused by two errors can be simply and efficiently eliminated, accurate following and accurate control of the current and the torque are achieved, the method has the advantages of being simple in algorithm, short in execution time and the like, and the method is suitable for popularization and application. And the stability and the reliability of the motor control system are effectively guaranteed, so that the motor control system can be widely applied to engineering practice.

Description

Technical Field

[0001] The present invention belongs to the technical field of permanent magnet synchronous motor control, and in particular relates to motor current compensation and control under the dual influence of current bias error and inverter nonlinear voltage error. Background Art

[0002] At present, the control of permanent magnet synchronous motor (PMSM) in this field mainly uses sensors to collect current signals and outputs reference voltages in combination with algorithms such as PI control and model predictive control. However, current sensors often age and fail with use, which may lead to various measurement errors, such as bias error and scaling error. Among them, the bias error will cause the motor current to fluctuate greatly, thereby affecting the stability of the motor output torque. The existing technology for such error compensation methods often relies on the motor speed and involves the adjustment of multiple parameters, which has the disadvantages of complex control and difficult implementation.

[0003] In addition, in PMSM current control, due to the dead zone effect of the inverter, there will be a deviation between the actual voltage applied to the motor and the reference voltage, which will cause the current and torque to fail to accurately follow the target value. Existing voltage compensation methods, such as compensation based on the direction of the A / B / C three-phase current, will have the problem of incorrect judgment of the three-phase current sign when there is a current bias error and zero current clamping effect, resulting in compensation failure and even affecting the stability of the control system.

[0004] It can be seen that there is an urgent need in this field to establish a dual compensation strategy for permanent magnet synchronous motor current control to solve the problem of current periodic fluctuation caused by current bias error and the problem of insufficient output torque caused by inverter nonlinearity, so as to ensure accurate current control and torque output of the motor. Summary of the invention

[0005] In view of this, in order to solve the technical problems existing in the art, the present invention provides a dual compensation method for permanent magnet synchronous motor current measurement error and inverter nonlinear error, which specifically includes the following steps:

[0006] ① Considering the bias error and the disturbance it brings, the voltage equation of the α / β axis of the permanent magnet synchronous motor is established;

[0007] ② Use a nonlinear disturbance observer to estimate the disturbance, and use a low-pass filter to obtain a smooth estimate of the bias error based on the disturbance estimation result, and form a compensation current on this basis;

[0008] ③ Compare the ideal voltage command with the actual voltage when considering the nonlinear effect of the inverter in a control cycle, and calculate the nonlinear voltage error and the corresponding compensation voltage command;

[0009] ④ Establish a beat-free predictive control model for the traditional permanent magnet synchronous motor, use the compensation current obtained in step ② for d / q axis current estimation, and calculate the voltage command at the next moment based on the compensated d / q axis current estimation result and the compensation voltage command obtained in step ③, so as to realize the dual compensation of bias error and inverter nonlinear voltage error, and finally obtain accurate current and output torque.

[0010] Furthermore, the specific process of establishing the α / β axis voltage equation in step ① includes:

[0011] For the conventional α / β axis permanent magnet synchronous motor voltage equation:

[0012]

[0013] In the formula, u α 、u β 、i α 、i β are the α / β axis voltage and current respectively; ω e is the rotor electrical angular velocity; θ r is the electrical angle; R s is the stator resistance; L α , L β is the α / β axis inductance; Ψ f is the rotor flux, t is the time;

[0014] Considering the α / β axis offset error Δi α , Δi β , the measured value of α / β axis current It is expressed as:

[0015]

[0016] The bias error Δi is introduced α , Δi β The disturbance caused by α and f β , we get the following new voltage equation:

[0017]

[0018] Among them, the disturbance

[0019] Furthermore, the process of estimating the disturbance using the nonlinear disturbance observer in step ② specifically includes:

[0020] First, the measured current value is derived from the α / β axis voltage equation including the offset error. The derivative expression of is:

[0021]

[0022] According to the principle of nonlinear disturbance observer, the following estimation equation of disturbance is obtained:

[0023]

[0024] In the formula, is the disturbance f α 、f β The estimated value of K α , K β is the observer gain;

[0025] According to the above formula, define the intermediate variable Z α , Z β as follows:

[0026]

[0027] By using the intermediate variable Z α , Z β Take the derivative and convert the derivative of the measured current into Substituting in:

[0028]

[0029] Then the estimated value of disturbance It can be expressed as:

[0030]

[0031] And the estimated value of the disturbance can be obtained To the estimate of the bias error The transfer function G α (s), G β (s) is:

[0032]

[0033] In order to reduce the impact of noise on the observation results, a low-pass filter is used to estimate the bias error. The following filtering is performed:

[0034]

[0035] Where m is the filter coefficient, and its value range is between (0,1); is the estimated value of the α / β axis bias error at time k; are the estimated values ​​of α / β axis bias errors after filtering at time k and time k-1 respectively;

[0036] Estimated value of the α / β axis bias error after filtering at time k After Park transformation, the estimated value of the d / q axis bias error at time k is obtained.

[0037]

[0038] Furthermore, the specific process of calculating the nonlinear voltage error and the corresponding compensation voltage instruction in step ③ includes:

[0039] For conventional permanent magnet synchronous motor d / q axis voltage equation:

[0040]

[0041] In the formula, u d 、u q 、i d 、i q are d / q axis voltage and current respectively; L d , L q is the d / q axis inductance;

[0042] Considering that the nonlinearity of the inverter mainly affects the steady-state error of the q-axis current, and ignoring its influence on the steady-state error of the d-axis current, the ideal voltage command of the q-axis is for:

[0043]

[0044] Under steady-state conditions, At the same time, the d-axis current reference value is 0, that is, vector control, thereby the ideal voltage command of the q axis Simplified to:

[0045]

[0046] In the formula, is the q-axis current reference value. Under steady-state conditions, the ideal voltage command of the q-axis can be considered is a constant.

[0047] According to the average value equivalent principle of the space vector pulse width modulation algorithm, the voltage acting on the motor in an ideal situation is In a control cycle T s The average value within should satisfy:

[0048]

[0049] Taking into account the nonlinear effect of the inverter, the actual voltage u acting on the motor is q (t), ideal voltage acting on the motor And the error voltage Δu q (t) The following relationship exists among the three:

[0050]

[0051] Introducing the constant δu q As the error voltage Δu q (t) In a control cycle T s The average value within, that is:

[0052]

[0053] Then the above three are in one control cycle T s The average value in satisfies the following expression:

[0054]

[0055] Therefore, in order to eliminate the voltage error caused by the nonlinearity of the inverter, the compensated voltage command Set to:

[0056]

[0057] The compensated voltage command The voltage u actually acting on the motor after passing through the inverter q (t) In a control cycle T s The average value within is:

[0058]

[0059] Furthermore, in step ④, the following traditional permanent magnet synchronous motor deadbeat predictive control model is first established:

[0060]

[0061] In the formula, i d (k), i q (k) and u d (k) and u q (k) are the d / q axis current and voltage at time k respectively; ω e (k) is the rotor electrical angular velocity at time k; is the predicted d / q axis current at time k+1; is the voltage command at time k+1;

[0062] The estimated value of the d / q axis bias error at time k obtained in step ① is Compensation to the actual current value In this paper, we can obtain accurate d / q axis current estimation values.

[0063]

[0064] Using the above d / q axis current estimates Accurately predict the d / q axis current at time k+1:

[0065]

[0066] On this basis, referring to the established deadbeat current prediction control model, the voltage command at time k+1 is obtained.

[0067]

[0068] Through the same voltage command compensation method in step ②, the voltage command at time k+1 after compensation is obtained. for:

[0069]

[0070] The dual compensation method for the current measurement error of the permanent magnet synchronous motor and the nonlinear error of the inverter provided by the present invention can accurately estimate and compensate the current measurement error by organically combining the nonlinear disturbance observer with the low-pass filter; at the same time, based on the integral comparison of the voltage actually acting on the motor and the reference voltage, the present invention proposes an effective compensation method for the voltage error caused by the nonlinearity of the inverter. In the implementation of the present invention, there is no need for motor speed information and A / B / C three-phase current sign judgment, and the current periodic fluctuation and torque output shortage caused by the two errors can be simply and efficiently eliminated, and the current and torque can be accurately followed and precisely controlled. It has the advantages of simple algorithm and short execution time, which helps to provide effective guarantee for the stability and reliability of the motor control system, so that it can be widely used in engineering practice. BRIEF DESCRIPTION OF THE DRAWINGS

[0071] Figure 1 The block diagram of the current bias error estimation including the nonlinear disturbance observer and the low-pass filter;

[0072] Figure 2 Schematic diagram of before and after compensation of voltage error caused by inverter nonlinearity;

[0073] Figure 3 The control block diagram of the dual compensation method for current bias error and inverter nonlinear voltage error;

[0074] Figure 4 The current and torque results before and after the inverter nonlinear voltage error compensation when there is no current bias error;

[0075] Figure 5 Figure 2 shows the current and torque results before and after current offset error compensation when voltage error compensation is used. DETAILED DESCRIPTION

[0076] The technical solution of the present invention will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0077] The dual compensation method for permanent magnet synchronous motor current measurement error and inverter nonlinear error provided by the present invention specifically comprises the following steps:

[0078] ① Considering the bias error and the disturbance it brings, the voltage equation of the α / β axis of the permanent magnet synchronous motor is established;

[0079] ② Use a nonlinear disturbance observer to estimate the disturbance, and use a low-pass filter to obtain a smooth estimate of the bias error based on the disturbance estimation result, and form a compensation current on this basis;

[0080] ③ Compare the ideal voltage command with the actual voltage when considering the nonlinear effect of the inverter in a control cycle, and calculate the nonlinear voltage error and the corresponding compensation voltage command;

[0081] ④ Establish a beat-free predictive control model for the traditional permanent magnet synchronous motor, use the compensation current obtained in step ② for d / q axis current estimation, and calculate the voltage command at the next moment based on the compensated d / q axis current estimation result and the compensation voltage command obtained in step ③, so as to realize the dual compensation of bias error and inverter nonlinear voltage error, and finally obtain accurate current and output torque.

[0082] Furthermore, the specific process of establishing the α / β axis voltage equation in step ① includes:

[0083] For the conventional α / β axis permanent magnet synchronous motor voltage equation:

[0084]

[0085] In the formula, u α 、u β 、i α 、i β are the α / β axis voltage and current respectively; ω e is the rotor electrical angular velocity; θ r is the electrical angle; R s is the stator resistance; L α , L β is the α / β axis inductance; Ψ f is the rotor flux, t is the time;

[0086] Considering the α / β axis offset error Δi α , Δi β , the measured value of α / β axis current It is expressed as:

[0087]

[0088] The bias error Δi is introduced α , Δi β The disturbance caused by α and f β , we get the following new voltage equation:

[0089]

[0090] Among them, the disturbance

[0091] Furthermore, the process of estimating the disturbance using the nonlinear disturbance observer in step ② specifically includes:

[0092] First, the measured current value is derived from the α / β axis voltage equation including the offset error. The derivative expression of is:

[0093]

[0094] According to the principle of nonlinear disturbance observer, the following estimation equation of disturbance is obtained:

[0095]

[0096] In the formula, is the disturbance f α 、f β The estimated value of K α , K β is the observer gain;

[0097] According to the above formula, define the intermediate variable Z α , Z β as follows:

[0098]

[0099] By using the intermediate variable Z α , Z β Take the derivative and convert the derivative of the measured current into Substituting in:

[0100]

[0101] Then the estimated value of disturbance It can be expressed as:

[0102]

[0103] And the estimated value of the disturbance can be obtained To the estimate of the bias error The transfer function G α (s), G β (s) is:

[0104]

[0105] In order to reduce the impact of noise on the observation results, a low-pass filter is used to estimate the bias error. The following filtering is performed:

[0106]

[0107] Where m is the filter coefficient, and its value range is between (0,1); is the estimated value of the α / β axis bias error at time k; are the estimated values ​​of α / β axis bias errors after filtering at time k and time k-1 respectively;

[0108] Estimated value of the α / β axis bias error after filtering at time k After Park transformation, the estimated value of the d / q axis bias error at time k is obtained.

[0109]

[0110] The control block diagram of the estimated current bias error through the nonlinear disturbance observer and low-pass filter in the above step ① is shown in the attached figure. Figure 1 As shown, it is easy to understand the data flow and processing steps.

[0111] Furthermore, the specific process of calculating the nonlinear voltage error and the corresponding compensation voltage instruction in step ③ includes:

[0112] For conventional permanent magnet synchronous motor d / q axis voltage equation:

[0113]

[0114] In the formula, u d 、u q 、i d 、i q are d / q axis voltage and current respectively; L d , L q is the d / q axis inductance;

[0115] Considering that the nonlinearity of the inverter mainly affects the steady-state error of the q-axis current, and ignoring its influence on the steady-state error of the d-axis current, the ideal voltage command of the q-axis is for:

[0116]

[0117] Under steady-state conditions, At the same time, the d-axis current reference value is 0, that is, vector control, thereby the ideal voltage command of the q axis Simplified to:

[0118]

[0119] In the formula, is the q-axis current reference value. Under steady-state conditions, the ideal voltage command of the q-axis can be considered is a constant.

[0120] According to the average value equivalent principle of the space vector pulse width modulation algorithm, the voltage acting on the motor in an ideal situation is In a control cycle T s The average value within should satisfy:

[0121]

[0122] Taking into account the nonlinear effect of the inverter, the actual voltage u acting on the motor is q (t), ideal voltage acting on the motor And the error voltage Δu q (t) The following relationship exists among the three:

[0123]

[0124] Introducing the constant δu q As the error voltage Δu q (t) In a control cycle T s The average value within, that is:

[0125]

[0126] Then the above three are in one control cycle T s The average value in satisfies the following expression:

[0127]

[0128] Therefore, in order to eliminate the voltage error caused by the nonlinearity of the inverter, the compensated voltage command Set to:

[0129]

[0130] The compensated voltage command The voltage u actually acting on the motor after passing through the inverter q (t) In a control cycle T s The average value within is:

[0131]

[0132] Therefore, the compensation voltage command can effectively correct the influence of the inverter nonlinearity, thereby improving the control accuracy of the system and eliminating the following error.

[0133] The voltage error caused by the nonlinearity of the inverter in step ② above is as follows before and after compensation: Figure 2 shown. Figure 2 (a) shows the voltage integral drop caused by the nonlinearity of the inverter without compensation; Figure 2 In (b), thanks to the voltage command compensation, the integral increase caused by the compensation voltage is offset by the voltage integral decrease caused by the inverter nonlinearity, so that the actual voltage integral acting on the motor ∫u q (t)dt and ideal voltage command integral The overlap clearly shows the voltage command compensation effect.

[0134] Furthermore, in step ④, the following traditional permanent magnet synchronous motor deadbeat predictive control model is first established:

[0135]

[0136] In the formula, i d (k), i q (k) and u d (k) and u q (k) are the d / q axis current and voltage at time k respectively; ω e (k) is the rotor electrical angular velocity at time k; is the predicted d / q axis current at time k+1; is the voltage command at time k+1;

[0137] The estimated value of the d / q axis bias error at time k obtained in step ① is Compensation to the actual current value In this paper, we can obtain accurate d / q axis current estimation values.

[0138]

[0139] Using the above d / q axis current estimates Accurately predict the d / q axis current at time k+1:

[0140]

[0141] On this basis, referring to the established deadbeat current prediction control model, the voltage command at time k+1 is obtained.

[0142]

[0143] Through the same voltage command compensation method in step ②, the voltage command at time k+1 after compensation is obtained. for:

[0144]

[0145] The control block diagram of the dual compensation method of current bias error and inverter nonlinear voltage error in the above step ③ is as follows: Figure 3 shown. Figure 3 The functions and data flows of each module, including deadbeat predictive current control, space vector pulse width modulation, nonlinear disturbance observer, and voltage command / current error compensation, are clearly identified, demonstrating the combination and control implementation of the dual compensation method.

[0146] In a specific example of the present invention, the motor parameters are as follows: DC bus voltage V dc is 135V, stator resistance R s 0.365Ω, α / β axis inductance (L α , L β ) and d / q axis inductance (L d , L q ) are both 0.001225H, rotor flux Ψ f is 0.1667Wb, control period T s 5×10 -5 s.

[0147] When there is no current bias error, the current and torque results before and after the inverter nonlinear voltage error compensation are as follows: Figure 4 shown. Figure 4 (a) shows the output before compensation. It can be seen that the q-axis current i q With its reference value There is always a deviation between the motor torque T e It is also impossible to follow the reference torque Figure 4 (b) shows the output after compensation. It can be seen that the current and torque following deviations are eliminated. The results clearly demonstrate the effect of the voltage command compensation method in improving control accuracy.

[0148] Figure 5 The current and torque results before and after current offset error compensation are shown when voltage error compensation is used. Figure 5 The current bias errors of the α / β axes are Δiα =0.500A, Δi β =0.866A, due to the lack of current bias error compensation, Figure 5 (a) The d / q axis current and torque both show large fluctuations; Figure 5 In (b), after compensation, the output current fluctuation is small and the torque pulsation is small. This example intuitively shows the improvement of the ripple of the two by the compensation measures, reflecting the practical application value of the present invention.

[0149] It should be understood that the size of the serial numbers of the steps in the embodiment of the present invention does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.

[0150] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A dual compensation method for permanent magnet synchronous motor current measurement error and inverter nonlinear error, characterized in that: The specific steps include: ① Considering the bias error and the disturbance it brings, the voltage equation of the α / β axis of the permanent magnet synchronous motor is established; ② Use a nonlinear disturbance observer to estimate the disturbance, and use a low-pass filter to obtain a smooth estimate of the bias error based on the disturbance estimation result, and form a compensation current on this basis; ③ Compare the ideal voltage command with the actual voltage when considering the nonlinear effect of the inverter in a control cycle, and calculate the nonlinear voltage error and the corresponding compensation voltage command; ④ Establish a beat-free predictive control model for the traditional permanent magnet synchronous motor, use the compensation current obtained in step ② for d / q axis current estimation, and calculate the voltage command at the next moment based on the compensated d / q axis current estimation result and the compensation voltage command obtained in step ③, so as to realize the dual compensation of bias error and inverter nonlinear voltage error, and finally obtain accurate current and output torque.

2. The method according to claim 1, characterized in that: The specific process of establishing the α / β axis voltage equation in step ① includes: For the conventional α / β axis permanent magnet synchronous motor voltage equation: In the formula, u α 、u β 、i α 、i β are the α / β axis voltage and current respectively; ω e is the rotor electrical angular velocity; θ r is the electrical angle; R s is the stator resistance; L α , L β is the α / β axis inductance; Ψ f is the rotor flux, t is the time; Considering the α / β axis offset error Δi α , Δi β , the measured value of α / β axis current It is expressed as: The bias error Δi is introduced α , Δi β The disturbance caused by α and f β , we get the following new voltage equation: Among them, the disturbance 3. The method according to claim 2, characterized in that: The process of estimating the disturbance using the nonlinear disturbance observer in step ② specifically includes: First, the measured current value is derived from the α / β axis voltage equation including the offset error. The derivative expression of is: According to the principle of nonlinear disturbance observer, the following estimation equation of disturbance is obtained: In the formula, is the disturbance f α 、f β The estimated value of K α , K β is the observer gain; According to the above formula, define the intermediate variable Z α , Z β as follows: By using the intermediate variable Z α , Z β Take the derivative and convert the derivative of the measured current into Substituting in: Then the estimated value of disturbance It can be expressed as: And the estimated value of the disturbance can be obtained To the estimate of the bias error The transfer function G α (s), G β (s) is: In order to reduce the impact of noise on the observation results, a low-pass filter is used to estimate the bias error. The following filtering is performed: Where m is the filter coefficient, and its value range is between (0,1); is the estimated value of the α / β axis bias error at time k; are the estimated values ​​of α / β axis bias errors after filtering at time k and time k-1 respectively; Estimated value of the α / β axis bias error after filtering at time k After Park transformation, the estimated value of the d / q axis bias error at time k is obtained.

4. The method according to claim 3, characterized in that: The specific process of calculating the nonlinear voltage error and the corresponding compensation voltage instruction in step ③ includes: For conventional permanent magnet synchronous motor d / q axis voltage equation: In the formula, u d 、u q 、i d 、i q are d / q axis voltage and current respectively; L d , L q is the d / q axis inductance; Considering that the nonlinearity of the inverter mainly affects the steady-state error of the q-axis current, and ignoring its influence on the steady-state error of the d-axis current, the ideal voltage command of the q-axis is for: Under steady-state conditions, At the same time, the d-axis current reference value is 0, that is, vector control, thereby the ideal voltage command of the q axis Simplified to: In the formula, is the q-axis current reference value; According to the average value equivalent principle of the space vector pulse width modulation algorithm, the voltage acting on the motor in an ideal situation is In a control cycle T s The average value within should satisfy: Taking into account the nonlinear effect of the inverter, the actual voltage u acting on the motor is q (t), ideal voltage acting on the motor And the error voltage Δu q (t) The following relationship exists among the three: Introducing the constant δu q As the error voltage Δu q (t) In a control cycle T s The average value within, that is: Then the above three are in one control cycle T s The average value in satisfies the following expression: Therefore, in order to eliminate the voltage error caused by the nonlinearity of the inverter, the compensated voltage command Set to: The compensated voltage command The voltage u actually acting on the motor after passing through the inverter q (t) In a control cycle T s The average value within is:

5. The method according to claim 4, characterized in that: In step ④, the following traditional permanent magnet synchronous motor deadbeat predictive control model is first established: In the formula, i d (k), i q (k) and u d (k) and u q (k) are the d / q axis current and voltage at time k respectively; ω e (k) is the rotor electrical angular velocity at time k; is the predicted d / q axis current at time k+1; is the voltage command at time k+1; The estimated value of the d / q axis bias error at time k obtained in step ① is Compensation to the actual current value In this paper, we can obtain accurate d / q axis current estimation values. Using the above d / q axis current estimates Accurately predict the d / q axis current at time k+1: On this basis, referring to the established deadbeat current prediction control model, the voltage command at time k+1 is obtained. Through the same voltage command compensation method in step ②, the voltage command at time k+1 after compensation is obtained. for:

Images