Permanent magnet synchronous motor current gain error compensation method based on rotating speed ripples

Abstract

A permanent magnet synchronous motor current gain error compensation method based on rotating speed ripples mainly aims at the problem of proportional gain error of three-phase output current sampling information in a permanent magnet synchronous motor driving system, and establishes a current equation from the angle of proportional error of three-phase current sampling. And analyzing the problem that the current has a negative sequence component. By analyzing the influence of the error sampling current on the system, it is found that the current negative sequence component can cause the speed to generate a frequency-doubled fluctuation signal, and the problem of proportional error sampling of the sampling current can be well solved. According to the invention, aiming at a rotating speed approximate equation obtained through analysis, a frequency-doubled disturbance component, namely proportional error information, in the rotating speed is extracted through a demodulation thought, and a compensation method is designed according to the error information.

Description

Technical Field

[0001] The present invention belongs to the technical field of power electronics and motor drive, and specifically relates to a current gain error compensation method for a permanent magnet synchronous motor based on speed ripple. Background Art

[0002] For permanent magnet synchronous motor drive systems based on vector control technology, accurate current measurement is a necessary condition for achieving high-performance control. However, imperfect components and non-ideal working conditions in the current measurement circuit, such as the nonlinearity of the current sensor, the inaccuracy of the resistors in the analog filter and the measurement resistor, and the quantization error of the analog-to-digital converter, can lead to current measurement errors. Current measurement errors in motor drive systems can reduce the overall control performance. Therefore, online compensation of output current sampling errors has become a hot topic of research.

[0003] The technical differences are as follows:

[0004] Technical comparison with patent CN113965129A "A method for compensating current measurement offset error in a permanent magnet synchronous motor control system"

[0005] Patent CN113965129A designs an error compensation strategy for the speed pulsation caused by the current measurement offset error of the permanent magnet synchronous motor control system. The core object is the speed pulsation problem caused by the current measurement offset error. We design an error compensation strategy for the speed double frequency pulsation problem caused by the current measurement gain error of the permanent magnet synchronous motor control system.

[0006] Technical comparison with patent CN116155159A "A method for compensating current measurement errors of permanent magnet synchronous motors"

[0007] In patent CN116155159A, the two-phase current of permanent magnet synchronous motor AB is sampled; the current sampling value of the current cycle is first compensated by the compensation value calculated in the previous cycle, and then the adaptive linear neural network is used to learn the two-phase current AB online, so as to extract the DC component and the amplitude of the phase current. However, we analyze the torque equation to find the double frequency pulsation problem caused by the current measurement gain error in the speed, and then design the compensation method by combining the use of low-pass filter and proportional differential controller to extract the pulsation component.

[0008] Technical comparison with patent CN116545328A "Complex vector control method and control system for permanent magnet synchronous motor of new energy vehicle"

[0009] In patent CN116545328A, the permanent magnet synchronous motor control performs high-speed double sampling or low-speed single sampling of the three-phase winding current of the motor stator in a stationary coordinate system, and then transforms the coordinates to a rotating two-phase coordinate system and makes a difference with the output current of the speed loop as the input signal of the complex vector decoupling controller to complete the decoupling; the position compensation control process improves the dynamic performance of the current loop by performing angle compensation for the delay problem existing in the sampled current signal. However, we solve the gain error problem that occurs when sampling the current. By sampling the speed signal, we analyze and extract the double frequency pulsating signal caused by the current measurement gain error in the speed, and then design a compensation method to compensate for the current sampling gain error of the permanent magnet synchronous motor.

[0010] In view of the above problems, the present application provides a current sampling gain error online compensation method. The present invention is simple to implement and does not require additional hardware. Summary of the invention

[0011] The purpose of the present invention is to overcome the deficiencies of the prior art. The present invention proposes a permanent magnet synchronous motor current gain error compensation method based on speed ripple, which solves the gain error problem existing in current sampling and improves the control performance of the permanent magnet synchronous motor drive system.

[0012] To achieve the above object, the technical solution adopted by the present invention is:

[0013] A method for compensating a current gain error of a permanent magnet synchronous motor based on speed ripple, characterized in that the method comprises the following steps:

[0014] Step 1: Establish the ab two-phase current equation of the three-phase permanent magnet synchronous motor drive system considering the sampling current error and the ideal sampling current;

[0015] Step 2: Combined with the restriction that the zero-sequence current is zero, the c-phase current is obtained, and the CLARK transformation is used to implement the three-phase transformation to express the measured current and the actual current equation in the stationary coordinate system. The relationship between the measured current and the actual current is analyzed by connecting the two equations;

[0016] Step 3: Combine the current equation under the complex vector to calculate the complex vector representation of the actual current and analyze that the proportional error of the sampled current will cause a negative sequence current component;

[0017] Step 4: Combining the torque equation of the permanent magnet synchronous motor with the actual current equation calculated in step 4, it is analyzed that the existence of the negative sequence component of the current in the torque will cause the fluctuation of the double frequency;

[0018] Step 5: Combining the motor mechanical motion equation and the torque equation in step 5, the proportional error information is extracted through demodulation, and the corresponding compensation method is designed. A permanent magnet synchronous motor current gain error compensation method based on speed ripple is designed.

[0019] As a further improvement of the present invention, the step 1 is specifically as follows:

[0020] In a three-phase permanent magnet synchronous motor drive system, two-phase current is generally sampled. Taking phase a and phase b as an example, considering the sampling error, the sampled current can be described as:

[0021]

[0022] where i o_m ,i b_m Represents the measured current, i a ,i b Indicates the real current, K a ,K b represents the coefficient gain, a 0 , b 0 Indicates the bias. Ideally, K ab =1,a 0 =b 0 =0.

[0023] As a further improvement of the present invention, the step 2 is specifically as follows:

[0024] According to the limitation of zero sequence current being zero, we can get:

[0025]

[0026] According to (1) and (2), the measured current in the two-phase stationary coordinate system can be expressed as:

[0027]

[0028] The measured current in the two-phase stationary coordinate system can be expressed as:

[0029]

[0030] According to (3) and (4), we can get:

[0031]

[0032] and:

[0033]

[0034] in Define a vector.

[0035] As a further improvement of the present invention, the step three is specifically as follows:

[0036] Define the vector:

[0037]

[0038] in 88 Represent the amplitude of the real current and the measured current, 0, C 0, C77 Respectively represent the phase of the real current and the measured current;

[0039] Combining (5) and (7), we can obtain:

[0040]

[0041] in,

[0042]

[0043] From observation (8), it can be seen that the current sampling error will cause the real current to produce negative sequence components and offsets, that is, the real current has harmonics at this time.

[0044] As a further improvement of the present invention, the step 4 is specifically as follows:

[0045] Consider the impact on the system, where ω c Represents the frequency of the current, and the flux linkage is:

[0046]

[0047] At this time, the motor torque can be expressed as:

[0048]

[0049] Among them, θ p =θ cm -θ r ,θ p , recorded as the power angle, ideally θ p =90°,θ cm Denoted as the rotor current angle, θ r Denote as the rotor position angle;

[0050] As a further improvement of the present invention, the step five is specifically as follows:

[0051] According to the motor speed equation, the friction coefficient is very small compared to the moment of inertia and can be ignored:

[0052]

[0053] Combining (11) and (12), the velocity response signal is:

[0054]

[0055] Observation (13) shows that when there is a proportional error in current sampling, the speed will produce a 2-fold frequency fluctuation. When there is a bias error in current sampling, the speed will produce a 1-fold frequency fluctuation. The amplitude of the 2-fold frequency fluctuation signal contains proportional error information. The proportional error information can be extracted through the demodulation idea:

[0056]

[0057] Design compensation method according to (14):

[0058]

[0059] The demodulation compensation method is implemented by combining a low-pass filter, a PI controller and a sign function. For the parameter design of the low-pass filter, in theory, as long as the bandwidth is not higher than twice the base frequency, it is 10 here. For the design of the PI controller and the sign function, the error gain is compensated. Assuming that the speed is forward, is positive, that is, K a >K b 'At this time n <0, the extracted gain error information y<0, it can be seen that K is needed at this time a Reduce, K b Increase, thereby achieving compensation for the error gain, so: At this point, correct gain compensation is successfully achieved, k p , k i The coefficients are 5 and 20 respectively.

[0060] The beneficial effects of the present invention are:

[0061] When the system is in steady state, the three-phase current of the motor with sampling current gain error is compensated and controlled, and the asymmetry problem of the three-phase current is finally eliminated, and the torque pulsation and speed pulsation are eliminated, so that the motor drive system can run stably. The present invention is simple to implement and does not add additional hardware costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0062] Figure 1 This is a system principle block diagram of the entire application;

[0063] Figure 2 A flow chart of current sampling gain error compensation for the system operation of this application;

[0064] Figure 3 The detailed operating principle diagram of the current sampling gain error compensation of this application;

[0065] Figure 4A phase diagram of various physical quantities in the derivation process of the principle of this application;

[0066] Figure 5 This is a comparison diagram of the torque waveform and demodulated signal waveform of the running example of this application;

[0067] Figure 6 This is a speed waveform diagram of the running example of this application;

[0068] Figure 7 Gain error extraction and disturbance compensation diagram for the running example of this application;

[0069] Figure 8 The three-phase current waveform diagram of the initial stage of demodulation control compensation of the running example of this application;

[0070] Fig. 9 This is a three-phase current waveform diagram at the later stage of demodulation control compensation of the running example of this application. DETAILED DESCRIPTION

[0071] The present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments:

[0072] The operating example of the present invention is a permanent magnet synchronous motor drive system in which the sampling of the a-phase current has a sampling gain error of 1.2 times.

[0073] Figure 1 The operation flow chart of the whole system, the system is first powered by the power supply. The inverter generates the required three-phase AC power to the permanent magnet synchronous motor according to the control signal of the controller and the power supply of the power supply. The permanent magnet synchronous motor outputs the speed and the electromagnetic torque of the motor. Considering that the torque cannot be accurately measured in the actual system, the present invention uses the output speed, the speed setting and the demodulated cosine signal to demodulate the double frequency error component in the current sampling, that is:

[0074]

[0075] The sampled current signal is compensated according to the demodulated error signal, and the corresponding control method is designed, namely:

[0076]

[0077] The signal obtained by current sampling is then transmitted to the controller for analysis and control after control compensation, and finally a new control signal is output to the power supply and inverter to form a closed-loop control.

[0078] Figure 2This is the system operation current sampling gain error compensation flow chart: the system first obtains the current through current sensor sampling, assuming that there is a gain error in the current information obtained at this time; then, the motor speed can be obtained through the encoder. Because of the gain error, the motor speed has a double frequency fluctuation at this time; then, according to the characteristics of the double frequency disturbance of the speed, the disturbance signal is extracted through the idea of ​​demodulation; finally, the corresponding compensation strategy is designed according to the disturbance signal to compensate and correct the sampled three-phase current error information.

[0079] Figure 3 This is a detailed operating principle diagram of current sampling gain error compensation. The disturbance signal can be extracted through the speed signal combined with the idea of ​​demodulation. The sampled current signal can be compensated and corrected through the PI regulator and the corresponding control method, and finally the disturbance component of the torque speed is eliminated and then stabilized.

[0080] Figure 4 It is a theoretical verification before running the example. First, verify whether the equation derivation is correct. In the three-phase permanent magnet synchronous motor drive system, two-phase current is generally sampled. Taking the sampling of two phases ab as an example, considering the sampling error, the sampled current can be described as:

[0081]

[0082] Among them. _mb m represents the measured current, ab represents the real current. K, K represents the coefficient gain, (0 1 b0 represents bias. Ideally, K = 1, @o = b0 = 0

[0083] The zero-sequence current limitation and coordinate transformation are as follows:

[0084] According to the limitation of zero sequence current being zero, we can get:

[0085]

[0086] According to (18) and (19), the measured current in the two-phase stationary coordinate system can be expressed as:

[0087]

[0088] The measured current in the two-phase stationary coordinate system can be expressed as:

[0089]

[0090] According to (20) and (21), we can get:

[0091]

[0092] and:

[0093]

[0094] in Define the vector:

[0095]

[0096] Among them I s , I s_m Represent the amplitude of the real current and the measured current, θ c θ cm Respectively represent the phase of the real current and the measured current.

[0097] Combining (22) and (24), we can obtain:

[0098]

[0099] in,

[0100]

[0101] From observation (25), it can be seen that the current sampling error will cause the real current to produce negative sequence components and offsets, that is, the real current has harmonics at this time.

[0102] Step 4: Analyze the influence of current sampling gain error on the system through torque equation:

[0103] Consider the impact on the system, where ω c Represents the frequency of the output current. The flux response is:

[0104]

[0105] At this time, the motor torque can be expressed as:

[0106]

[0107] Among them, θ p =θ cm - 0r ,θ p Denoted as power angle, θ cm Denoted as the rotor current angle, θ r Denote as the rotor position angle, see the specific phase diagram Figure 4 .

[0108] It can be seen that when only the gain sampling error is considered in the steady state, T e Will produce double frequency fluctuations, T e There is no one-fold frequency fluctuation, and the amplitude of the double-frequency fluctuation signal contains gain error information.

[0109] So we can directly verify that when only the gain sampling error is considered in the steady state, Te and sin(-2θ r -θ p +ψ 1 ) The phases are the same or 180° apart. The simulation results are shown in Figure 5 .from Figure 5 It can be seen that after the system reaches a steady state, T e The phase difference with sin(-20,-0,+) is 180°, which is consistent with expectations.

[0110] Since the torque signal cannot be directly measured to get an accurate value in the actual system, the motor speed equation is used (the friction coefficient is very small compared to the moment of inertia and can be ignored):

[0111]

[0112] It can be seen that the motor speed and torque differ by 90° in steady state, so in actual situations, the speed can be used instead of the torque to identify the sign of the double frequency component generated by the gain disturbance error.

[0113] Combining (28) and (29), the velocity response signal is:

[0114]

[0115] Observation (30) shows that when there is a gain error in current sampling, the speed will produce a double frequency fluctuation, and when there is an offset error in current sampling, the speed will produce a single frequency fluctuation. And the amplitude of the double frequency fluctuation signal contains the gain error information. The gain error information is extracted through the demodulation idea:

[0116]

[0117] According to (31), the compensation method is designed:

[0118]

[0119] Therefore, based on the above theoretical analysis, a simulation model is built and run for verification. Figure 6 From the waveform diagram of the forward rotation speed during example operation, it can be seen that the actual speed can eventually coincide with the given speed, which indicates that the present invention can effectively eliminate the influence of the current sampling gain error on the speed and torque fluctuations, thus verifying the correctness of the present invention.

[0120] Figure 7 It is the waveform diagram of the motor speed signal, demodulated extracted signal cos(-20,-0,+) and the extracted gain error information. Figure 7 It can be seen that k can be well extracted by using the demodulated signal, speed information and low-pass filter. n ψf To compensate the current gain error disturbance, and finally achieve good disturbance elimination and suppression, that is, to make the double frequency disturbance component k n ψ f sin(-2θ r -θ p +ψ 1 ) tends to 0.

[0121] Figure 8 The three-phase current waveform at the beginning of disturbance compensation when the sampling current has gain disturbance error. Figure 8 It can be seen that at the beginning, when the current has a gain error and is not fully compensated, the three-phase current is asymmetric, which is related to Figure 6 The speed waveforms correspond to each other, and the corresponding speed will fluctuate, so it is necessary to compensate for the current with gain error in the sampling current.

[0122] The compensation method is designed according to formula (32). Fig. 9 This is the three-phase current waveform after gain disturbance compensation of the sampled current. It can be seen that the current can gradually tend to three-phase symmetry, and the compensation effect is very good. Correspondingly, Figure 6 The speed waveform gradually becomes stable and no longer fluctuates.

[0123] The above description is only a preferred embodiment of the present invention and does not constitute any other form of limitation to the present invention. Any modification or equivalent change made based on the technical essence of the present invention still falls within the scope of protection required by the present invention.

Claims

1. A method for compensating current gain error of a permanent magnet synchronous motor based on speed ripple, characterized in that: The method comprises the following steps: Step 1: Establish the ab two-phase current equation of the three-phase permanent magnet synchronous motor drive system considering the sampling current error and the ideal sampling current; Step 2: Combined with the restriction that the zero-sequence current is zero, the c-phase current is obtained, and the CLARK transformation is used to implement the three-phase transformation to express the measured current and the actual current equation in the stationary coordinate system. The relationship between the measured current and the actual current is analyzed by connecting the two equations; Step 3: Combine the current equation under the complex vector to calculate the complex vector representation of the actual current and analyze that the proportional error of the sampled current will cause a negative sequence current component; Step 4: Combining the torque equation of the permanent magnet synchronous motor and the actual current equation calculated in step 4, it is analyzed that the existence of the negative sequence component of the current in the torque will cause the fluctuation of the double frequency; Step 5: Combining the motor mechanical motion equation and the torque equation in step 5, the proportional error information is extracted through demodulation, and the corresponding compensation method is designed. A permanent magnet synchronous motor current gain error compensation method based on speed ripple is designed.

2. The method for compensating a current gain error of a permanent magnet synchronous motor based on speed ripple according to claim 1, characterized in that: The step 1 is specifically as follows: In a three-phase permanent magnet synchronous motor drive system, two-phase current is generally sampled. Taking phase a and phase b as an example, considering the sampling error, the sampled current can be described as: where i a_m ,i b_m Represents the measured current, i a ,i b Indicates the real current, K a , K b represents the coefficient gain, a0, b0 represent the bias, and ideally K ab =1, a0=b0=0.

3. The method for compensating a current gain error of a permanent magnet synchronous motor based on speed ripple according to claim 1, characterized in that: The step 2 is specifically as follows: According to the limitation of zero sequence current being zero, we can get: According to (1) and (2), the measured current in the two-phase stationary coordinate system can be expressed as: The measured current in the two-phase stationary coordinate system can be expressed as: According to (3) and (4), we can get: and: in Define a vector.

4. The method for compensating a current gain error of a permanent magnet synchronous motor based on speed ripple according to claim 1, characterized in that: The step three is specifically as follows: Define the vector: Among them I s , I s_m Represent the amplitude of the real current and the measured current, θ c ,θ cm Respectively represent the phase of the real current and the measured current; Combining (5) and (7), we can obtain: in, From observation (8), it can be seen that the current sampling error will cause the real current to produce negative sequence components and offsets, that is, the real current has harmonics at this time.

5. The method for compensating a current gain error of a permanent magnet synchronous motor based on speed ripple according to claim 1, characterized in that: The step 4 is specifically as follows: Consider the impact on the system, where ω c Represents the frequency of the current, and the flux linkage is: At this time, the motor torque can be expressed as: Among them, θ p =θ cm -θ r ,θ p Denoted as the power angle, ideally θ p =90°,θ cm Denoted as the rotor current angle, θ r Recorded as the rotor position angle.

6. The method for compensating a current gain error of a permanent magnet synchronous motor based on speed ripple according to claim 1, characterized in that: The step five is specifically as follows: According to the motor speed equation, the friction coefficient is very small compared to the moment of inertia and can be ignored: Combining (11) and (12), the velocity response signal is: Observation (13) shows that when there is a proportional error in current sampling, the speed will produce a 2-fold frequency fluctuation. When there is a bias error in current sampling, the speed will produce a 1-fold frequency fluctuation. The amplitude of the 2-fold frequency fluctuation signal contains proportional error information. The proportional error information can be extracted through the demodulation idea: Design compensation method according to (14): The demodulation compensation method is implemented by combining a low-pass filter, a PI controller and a sign function. For the parameter design of the low-pass filter, in theory, as long as the bandwidth is not higher than twice the base frequency, it is 10 here. For the design of the PI controller and the sign function, the error gain is compensated. Assuming that the speed is forward, is positive, that is, K a >K b , at this time k n <0, the extracted gain error information y<0, it can be seen that K is needed at this time a Reduce, K b Increase, thereby achieving compensation for the error gain, so: At this point, correct gain compensation is successfully achieved, k p , k i The coefficients are 5 and 20 respectively.

Images