Sampling current compensation method based on extended state observer

Abstract

The invention discloses a sampling current compensation method based on an extended state observer, solves the problems that an existing current sampling compensation method based on a current loop is small in application range and is limited by the number and position of current sensors, and belongs to the field of motor driving. The method comprises the steps that sampling currents idm, iqm, iz1m and iz2m and sampling voltages ud, uq, uz1 and uz2 are acquired; inputting ud, uq, uz1, uz2, idm, iqm, iz1m, iz2m and error currents id and iqe obtained by current observation into an extended state observer to obtain estimated values # imgabs0 # # imgabs1 #, wherein the extended state observer comprises a d-q plane extended state observer and a z1-z2 plane extended state observer; according to the obtained estimated value # imgabs2, error currents id, iqe, iz1e and iz2e are obtained; and subtracting the obtained error current from the sampling current under the rotating coordinate system to realize error compensation of the sampling current. The compensation method is irrelevant to the number and position of the current sensors, and the application range is wide.

Description

Technical Field

[0001] The invention relates to a sampling current compensation method based on an extended state observer, and belongs to the field of motor driving. Background Art

[0002] The dual three-phase permanent magnet synchronous motor drive control system has the advantages of low voltage, high power, and low torque pulsation, and has been widely used in new energy vehicles, ships, aerospace and other fields. In order to suppress zero-sequence current, the two sets of windings of the dual three-phase motor generally adopt the neutral point isolation operation mode. Under this operation mode, the dual three-phase electric drive system can be arranged with six-phase current sensors or set to 2 groups of 4 to reduce costs. The two sets of current sensors correspond to two sets of windings, and the two current sensors are placed in any two phases of a set of windings. Current sampling must go through the processes of current sensor, operation circuit, AD conversion, etc., among which sensor error, zero drift of op amp, AD conversion error, calibration error and other problems will cause measurement errors. The measurement error is generally manifested as gain error and offset error, which will cause 1st and 2nd torque pulsation respectively. The winding setting of the dual three-phase motor can theoretically suppress the prominent 6th torque pulsation problem in the three-phase motor, which makes the low-frequency torque pulsation caused by the sampling current error more significant. Therefore, it is necessary to adopt a suitable compensation method to compensate for the current sampling.

[0003] One of the traditional current sampling compensation methods is to regard current sampling as interference in the forward channel of the speed loop and suppress and compensate it. This method is simple in principle and has strong robustness. It starts with the idea of ​​torque pulsation suppression and suppresses the i related to torque pulsation. q Compensation is performed to achieve the effect of suppressing the torque pulsation caused by current sampling. However, the above method does not eliminate the phase current bias and asymmetry caused by current sampling. The phase current still contains abundant harmonics. The stator resistance and leakage inductance corresponding to the resistance inductance of the z1-z2 plane of the dual three-phase motor are small, the harmonic current is large, the heating of the winding increases, and the efficiency of the electric drive control system is reduced. Another method is based on the current loop, subtracting the estimated compensation current from the feedback current or correcting the gain and offset of the current sampling online. The core of this type of method is to extract the relevant current pulsation and estimate the compensation current, which can effectively improve the phase current waveform, suppress torque pulsation, and speed up the response speed of compensation. However, under low-speed conditions, the 1- and 2-fold current frequencies are very close, the cascaded resonator or trap has many setting parameters, and the extraction effect of the pulsating current is poor; the estimation of the compensation current is related to the motor model, the number and position of the current sensors, and the algorithm has a small scope of application. Summary of the invention

[0004] Aiming at the problem that the existing current sampling compensation method based on current loop has a small application scope and is limited by the number and position of current sensors, the present invention provides a sampling current compensation method based on extended state observer.

[0005] The present invention provides a sampling current compensation method based on an extended state observer, comprising:

[0006] S1, collect the current and voltage signals of the motor to obtain the current i dm , i qm , i z1m , i z2m and voltage u d 、u q 、u z1 、u z 2;

[0007] i dm , i qm are the d-axis current and q-axis current of the sampled current in the rotating coordinate system, i z1m , i z2m are the z1 axis current and z2 axis current of the sampled current in the rotating coordinate system, u d and u q are the d-axis voltage and q-axis voltage of the sampled voltage in the rotating coordinate, u z1 、u z2 are the z1-axis voltage and z2-axis voltage of the sampled voltage in the z1-z2 plane respectively;

[0008] S2. d 、u q 、u z1 、u z2 , i dm , i qm , i z1m , i z2m and the error current i de , i qe Enter into the expansion state observer to obtain the estimated value i de , i qe They are the d-axis error current and q-axis error current of the sampling current in the rotating coordinate system respectively;

[0009] The extended state observer is a four-dimensional current loop extended state observer, including a dq plane extended state observer and a z1-z2 plane extended state observer. The dq plane extended state observer is:

[0010]

[0011] where e d 、e q For i dm , i qm estimation error, i dm , iqm The estimated value of ω e is the electric angular velocity, β 1 =2ω eso ,ω eso For the bandwidth of the expanded state observer, ψ f is the amplitude of the permanent magnet magnetic flux, L d , L q They are the d-axis inductance and q-axis inductance of the motor respectively;

[0012] The z1-z2 plane expansion state observer is:

[0013]

[0014] Among them, e z1 、e z2 For i z1m 、i z2m estimation error; for i z1m , i z2m The estimated value of R s For the stator resistance of the motor, L l For the leakage inductance of the motor, are the estimated values ​​of the error voltages generated by the d-axis error current and the q-axis error current, respectively. are the estimated values ​​of the error voltages generated by the z1-axis error current and the z2-axis error current respectively;

[0015] S3. Based on the obtained estimated value Obtain error current i de , i qe , i z1e , i z2e ;i z1e , i z2e They are the z1-axis error current and z2-axis error current of the sampled current in the rotating coordinate system respectively;

[0016] S4. Subtract the obtained error current from the sampled current in the rotating coordinate system to achieve error compensation of the sampled current.

[0017] Preferably, in S3, the compensation current i de , i qe , i z1e , i z2e for:

[0018]

[0019] s represents a pull-transformation expression.

[0020] As a preferred embodiment, the motor is a dual three-phase permanent magnet synchronous motor, and a four-phase current sensor is used, which is placed on the A phase, B phase, X phase, and Y phase of the motor to collect current. The mathematical model of the dual three-phase permanent magnet synchronous motor in the rotating coordinate system is obtained by using the Clarke-Park transformation according to the vector space decoupling method to determine the voltage u d 、u q 、u z1 、u z2 .

[0021] As a preferred embodiment, the mathematical model of the dual three-phase permanent magnet synchronous motor is:

[0022]

[0023] The beneficial effect of the present invention is that the extended state observer designed by the present invention can observe multiple harmonics, and the observation capability is only as large as ω eso The compensation method of the present invention is independent of the number and position of current sensors and has a wide range of applications. For the dual three-phase permanent magnet synchronous motor drive control system, the present invention implements current sampling compensation based on the current loop, suppresses torque pulsation, improves the current waveform, and improves the efficiency reduction problem caused by phase current with high harmonic content. Attached description of the drawings

[0024] Figure 1 The main power topology diagram of the electric drive control system of dual three-phase permanent magnet synchronous motor with a phase angle of 30°;

[0025] Figure 2 It is a sampling current error compensation method and control block diagram of a dual three-phase permanent magnet synchronous motor;

[0026] Figure 3 The control block diagram of the four-dimensional current loop for decoupling control;

[0027] Figure 4 It is a control block diagram of the speed ring;

[0028] Figure 5 It is a control block diagram of the q-axis sampling current compensation method;

[0029] Figure 6 Comparison waveforms of six-phase currents before and after the compensation method is added;

[0030] Figure 7 Add a comparison waveform of the speed before and after for the compensation method;

[0031] Figure 8 Comparison waveform of electromagnetic torque before and after adding compensation method. DETAILED DESCRIPTION

[0032] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. 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.

[0033] It should be noted that, in the absence of conflict, the embodiments of the present invention and the features in the embodiments may be combined with each other.

[0034] The present invention will be further described below in conjunction with the accompanying drawings and specific embodiments, but they are not intended to limit the present invention.

[0035] The sampling current compensation method based on the extended state observer in this embodiment is applied to the control of the dual three-phase permanent magnet synchronous motor. The control principle of the dual three-phase permanent magnet synchronous motor is as follows: Figure 2 As shown, it includes a speed loop, a current loop and a compensation method; the specific control process includes:

[0036] Four-phase current sensors are used, which are placed in phase A, phase B, phase X and phase Y respectively. The mathematical model of the dual three-phase permanent magnet synchronous motor in the rotating coordinate system is obtained by using Clarke-Park transformation according to the vector space decoupling method to obtain the sampling current and sampling voltage.

[0037] The control strategy is the vector control method of maximum torque current ratio, which is i for surface mounted motors. d =0 control; the sampling current error compensation scheme is based on the current loop, and the voltage equation considering the sampling current error is derived. The sampling error voltage is observed by an extended state observer in the dq and z1-z2 planes based on the sampling current and the sampling voltage. The error current is obtained according to the mathematical relationship between the error voltage and the error current, and the error current is subtracted from the sampling current for compensation.

[0038] After sampling current compensation, the dual three-phase permanent magnet synchronous motor is driven and controlled based on the double closed-loop control method of speed loop and current loop.

[0039] Specifically, this embodiment is designed based on the vector control method of the current loop and the speed loop. The three-phase windings of the 14kW dual three-phase permanent magnet synchronous motor with a phase shift angle of 30° are independently connected in star shape. The main circuit is a voltage source type six-phase inverter circuit with a bus voltage of 650V. Figure 1 Shown. The control policy is i d = 0 vector control strategy, the control system adopts the current loop and speed loop double closed loop control method, the speed loop output torque current i qref As the input given by the q-axis current loop, the excitation current i dGiven as 0, the z1-z2 plane current i of the dual three-phase motor z1 、i z2 Given to 0 to suppress harmonic current.

[0040] The Clarke transformation matrix based on the space vector decoupling method is:

[0041]

[0042] In this implementation, only the αβ plane is rotated, and the corresponding Park transformation matrix is:

[0043]

[0044] Where θ=ω e t,ω e is the electrical angular velocity, 0 is the zero matrix, and I is the 4th-order unit matrix. Ignoring the zero-sequence current plane, the voltage equation of the dual three-phase permanent magnet synchronous motor in the rotating coordinate system can be obtained as follows:

[0045]

[0046] Where, L d =L q = 3L m +L l , the electromagnetic torque expression in the rotating coordinate system is expressed as

[0047] T e =3p n [(L d -L q )i d i q +i q ψ f ] (4)

[0048] Adjust the current loop and speed loop: The current loop of this embodiment is a PI controller. According to the control requirements, it is adjusted to

[0049]

[0050] In the formula, k p_d , k i_d are the d-axis proportional and integral gains of the PI controller, k p_q , k i_q are the q-axis proportional and integral gains of the PI controller, k p_z1 , k i_z1 The z1 axis ratio and k of the PI controller are respectively p_z2 , k i_z2 are the z2-axis proportional and integral gains of the PI controller, ω cis the current loop bandwidth. Ignoring the delay and nonlinear factors of the inverter output, the open-loop transfer function of the current loop is set to a first-order system. The control block diagram of the four-dimensional current loop is as follows: Figure 3 Shown. C ix (s) is the mathematical expression of the current loop controller, and x is d, q, z1, and z2.

[0051] The speed loop is tuned as a second-order system. Ignoring the influence of the damping coefficient, the current loop is considered to be a well-tuned first-order system. At this time, the speed loop PI controller is tuned as

[0052]

[0053] In the formula, k p_s , k i_s are the proportional and integral gains of the PI controller, ω s For speed ring bandwidth, K T represents the torque coefficient, and J represents the moment of inertia. Substituting the speed loop bandwidth and the dual three-phase motor parameters into equation (6), we can obtain the control parameters of the speed loop PI controller. The control block diagram of the speed loop is as follows: Figure 4 shown.

[0054] The motor model considering the current measurement error is derived. The measured error current can be summarized as the offset error current and the gain error current, with the A phase current as the reference current.

[0055] i a =Icos(ωt+φ) (7)

[0056] The other five phase currents have similar forms, where I is the phase current amplitude, φ is the current phase angle, and

[0057]

[0058] The sampling current error can be expressed as

[0059]

[0060] In the formula, i ae 、i be 、i ce 、i xe 、i ye 、i ze For the sampling current error of each phase, ΔI a , ΔI b , ΔI x , ΔI y For each phase sampling bias current, i a 、i b 、i x 、i y To measure the actual current of the phase, ka , k b , k x , k y is the measured gain. The sampling current error includes the bias error current and the gain error current, which are analyzed separately below. Based on the vector space method, the Clarke-Park transformation of the sampling current can be obtained to obtain the current error expression in the dq and z1-z2 planes:

[0061]

[0062] Substituting the bias error current into the above formula can obtain

[0063]

[0064] In the formula, I off and Can be expressed as

[0065]

[0066] Substituting the gain error current into formula (9) can obtain

[0067]

[0068] In the formula, I scale1 , I scale2 and It can be expressed as

[0069]

[0070] From formula (4), we can see that since the surface mounted permanent magnet synchronous motor has L d =L q , so the electromagnetic torque output is from i q It is determined that the current bias error and gain error will produce q-axis current of fundamental frequency and twice the frequency, forming torque pulsation of fundamental frequency and twice the frequency. Only compensating the q-axis current can suppress the torque pulsation, but the error current of the d-axis, z1-axis and z2-axis currents is not compensated. In particular, the motor stator resistance and leakage inductance corresponding to the resistance-inductance of the z1-z2 plane are small, and the thermal effect generated by the harmonic current is more significant. It is necessary to adopt a suitable compensation method to suppress the harmonic current.

[0071] Step 1: Sample the motor rotor position to obtain the current motor electrical angle θ, and then take the derivative of θ to obtain the current motor electrical angular velocity ω e According to the number of motor pole pairs, the current motor mechanical angular velocity ω can be obtained m ;

[0072] Four-phase current sensors with offset error and gain error are used to collect phase currents in phase A, phase B, phase X and phase Y respectively. Six-phase sampling currents are obtained according to Kirchhoff's current law. Clarke-Park transformation is used according to the vector space decoupling method to convert the six-phase currents in the natural coordinate system into four-dimensional currents in the dq and z1-z2 planes: i dm 、i qm 、i z1m 、i z2m ;i dm 、i qm are the d-axis current and q-axis current of the sampled current in the rotating coordinate system, i z1m 、i z2m They are respectively the z1-axis current and the z2-axis current of the sampled current in the rotating coordinate system;

[0073] Sampling the u output from the inverter ab 、u bc 、u ca 、u xy 、u yz 、u zx Line voltage, according to Kirchhoff's voltage law, the six-phase phase voltage is calculated. According to the vector space decoupling method, Clarke-Park transformation is used to convert the six-phase voltage in the natural coordinate system into four-dimensional voltage in the dq and z1-z2 planes: u d 、u q 、u z1 、u z2 ;u d and u q are the d-axis voltage and q-axis voltage of the sampled voltage in the rotating coordinate, u z1 、u z2 are the z1-axis voltage and z2-axis voltage of the sampled voltage in the z1-z2 plane respectively;

[0074] The sampled bus voltage is used as the reference bus voltage for PWM generation.

[0075] Step 2: Establish an extended state observer. The extended state observer in this embodiment is a four-dimensional current loop extended state observer, including a dq plane extended state observer and a z1-z2 plane extended state observer:

[0076] The sampled currents in the dq plane and the z1-z2 plane are re-expressed as

[0077]

[0078] In the formula, i dm 、i qm 、i z1m 、i z2mFor the sampling current after Clarke-Park transformation, i d 、i q 、i z1 、i z2 For the actual current after Clarke-Park transformation, i de 、i qe 、i z1e 、i z2e is the sampling current error after Clarke-Park transformation. Substituting equation (14) into equation (3) yields the voltage equation of the current sampling error:

[0079]

[0080] The deviation current is estimated by the linear extended state observer, and the above equation is rewritten as

[0081]

[0082] Build a DQ plane expansion state observer:

[0083]

[0084] Construct the z1-z2 plane expansion state observer:

[0085]

[0086] will u d 、u q 、u z1 、u z2 、i dm 、i qm 、i z1m 、i z2m and the error current i de 、i qe Input to the dq plane extended state observer and the z1-z2 plane extended state observer to obtain the estimated value

[0087] Step 3: Based on the estimated value obtained The error current is obtained. The disturbance observed by the extended state observer in the dq plane and z1-z2 plane is

[0088]

[0089] The compensation current i can be obtained from formula (19). de 、i qe 、i z1e 、i z2e for:

[0090]

[0091] s represents a pull-transformation expression.

[0092] In formula (17), u d,q The sampled line voltage is obtained through coordinate transformation, i dm,qm The sampled four-phase current is obtained through coordinate transformation, ω e From the motor position sensor sampling, i de,qe The output compensation current of the compensation method is obtained. The motor parameters are known and the extended state observer of the dq plane can be constructed. Equation (18) does not have the current coupling term of the dq plane, and the required system state parameters and motor parameters are similar to those of equation (16).

[0093] Step 4:

[0094] The estimated current error is subtracted from the sampled current in the rotating coordinate system to achieve sampled current error compensation. The compensated current suppresses the current pulsation in the dq plane and the z1-z2 plane, and is used as the feedback current of the current loop to control the motor current. The output given voltage of the current loop is transformed by the inverse Clarke-Park transformation based on vector space decoupling to obtain the six-phase given phase voltage u abc 、u xyz ,u abc 、u xyz After SPWM modulation, a driving signal is sent to control the power module of the six-phase inverter to generate waves. The inverter outputs six-phase voltage to achieve the drive of dual three-phase motors. The block diagram of the sampling current error compensation method of the q-axis extended state observer is shown in Figure 5 The comparison waveforms of the six-phase currents before and after the compensation method is added are shown in Figure 6 As shown in the figure, the comparison waveform of the rotation speed before and after the compensation method is added is as follows Figure 7 As shown in the figure, the comparison waveforms of the electromagnetic torque before and after the compensation method is added are as follows: Figure 8 shown.

[0095] Although the present invention is described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the present invention. It should therefore be understood that many modifications may be made to the exemplary embodiments and that other arrangements may be devised without departing from the spirit and scope of the present invention as defined by the appended claims. It should be understood that the various dependent claims and features described herein may be combined in a manner different from that described in the original claims. It should also be understood that features described in conjunction with individual embodiments may be used in other described embodiments.

Claims

1. A sampling current compensation method based on an extended state observer, characterized in that: include: S1, collect the current and voltage signals of the motor to obtain the current i dm 、i qm 、i z1m 、i z2m and voltage u d 、u q 、u z1 、u z2 ; i dm 、i qm are the d-axis current and q-axis current of the sampled current in the rotating coordinate system, i z1m 、i z2m are the z1 axis current and z2 axis current of the sampled current in the rotating coordinate system, u d and u q are the d-axis voltage and q-axis voltage of the sampled voltage in the rotating coordinate, u z1 、u z2 are the z1-axis voltage and z2-axis voltage of the sampled voltage in the z1-z2 plane respectively; S2, u d 、u q 、u z1 、u z2 、i dm 、i qm 、i z1m 、i z2m and the error current i obtained by current observation de 、i qe Input to the extended state observer to get the estimated value i de 、i qe They are the d-axis error current and q-axis error current of the sampling current in the rotating coordinate system respectively; The extended state observer is a four-dimensional current loop extended state observer, including a dq plane extended state observer and a z1-z2 plane extended state observer. The dq plane extended state observer is: where e d 、e q are i dm 、i qm The estimation error of i dm 、i qm The estimated value of ω e is the electrical angular velocity, β1=2ω eso ,ω eso is the bandwidth of the extended state observer, ψ f is the permanent magnet flux amplitude, L d , L q They are the d-axis inductance and q-axis inductance of the motor respectively; The expanded state observer in the z1-z2 plane is: Among them, e z1 、e z2 are i z1m 、i z2m The estimation error of for i z1m 、i z2m The estimated value, R s is the stator resistance of the motor, L l is the leakage inductance of the motor, are the estimated values ​​of the error voltages generated by the d-axis error current and the q-axis error current, respectively. are the estimated values ​​of the error voltages generated by the z1-axis error current and the z2-axis error current respectively; S3. Based on the estimated value Get the error current i de 、i qe 、i z1e 、i z2e ;i z1e 、i z2e They are the z1-axis error current and z2-axis error current of the sampled current in the rotating coordinate system respectively; S4. Subtract the obtained error current from the sampled current in the rotating coordinate system to achieve error compensation of the sampled current.

2. The sampling current compensation method based on the extended state observer according to claim 1 is characterized in that: In S3, the compensation current i de 、i qe 、i z1e 、i z2e for: s represents a pull transformation expression.

3. The sampling current compensation method based on extended state observer according to claim 1, characterized in that: The motor is a dual three-phase permanent magnet synchronous motor. A four-phase current sensor is used, which is placed on the A phase, B phase, X phase, and Y phase of the motor to collect current. The mathematical model of the dual three-phase permanent magnet synchronous motor in the rotating coordinate system is obtained using the Clarke-Park transformation according to the vector space decoupling method to determine the voltage u d 、u q 、u z1 、u z2 .

4. The sampling current compensation method based on extended state observer according to claim 3 is characterized in that: The mathematical model of dual three-phase permanent magnet synchronous motor is:

5. A computer-readable storage device storing a computer program, characterized in that: When the computer program is executed by a processor, the steps of the sampling current compensation method based on an extended state observer according to any one of claims 1 to 4 are implemented.

6. A sampling current compensation device based on an extended state observer, comprising a storage device, a processor, and a computer program stored in the storage device and executable on the processor, characterized in that: The processor executes the computer program to implement the steps of the sampling current compensation method based on the extended state observer according to any one of claims 1 to 4.

7. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the sampling current compensation method based on an extended state observer as claimed in any one of claims 1 to 4 are implemented.

8. The control method based on the sampling current compensation method based on the extended state observer according to claim 1, characterized in that: Use the speed loop to get the torque current i qref , as the input given of the q-axis current in the current loop, based on the sampling current compensation method based on the extended state observer described in claim 1, the error-compensated sampling current is obtained as the sampling current of the current loop, and the output given voltage of the current loop is based on the inverse Clarke-Park transformation of the vector space decoupling to obtain the three-phase voltage u of the two sets of windings abc 、u xyz ,u abc 、u xyz After SPWM modulation, a driving signal is sent out to control the power module of the six-phase inverter to generate waves. The six-phase inverter outputs six-phase voltage to realize the drive of the dual three-phase permanent magnet synchronous motor.

9. The control method according to claim 8, characterized in that: The PI controller of the current loop is: In the formula, k p_d , k i_d are the d-axis proportional and integral gains of the PI controller, k p_q , k i_q are the q-axis proportional and integral gains of the PI controller, k p_z1 , k i_z1 are the z1-axis ratio and k of the PI controller respectively. p_z2 , k i_z2 are the z2-axis proportional and integral gains of the PI controller, ω c is the current loop bandwidth.

10. According to the control method of claim 8, the PI controller of the speed loop is: In the formula, k p_s , k i_s are the proportional and integral gains of the PI controller, ω s is the speed loop bandwidth, ω c is the current loop bandwidth, K T represents the torque coefficient and J represents the moment of inertia.

Images