An Improved Linear Active Disturbance Rejection Motor Control Method Based on Quasi-resonant Controller

Abstract

The present invention provides an improved linear anti-disturbance motor control method based on a quasi-resonant controller. First, a mathematical model of an AC motor current loop is established, and the total disturbance of the current loop is expanded into a new state variable; a linear extended state observer is constructed, and the frequency domain characteristics of the extended state observer are analyzed; the disturbance feedforward channel of the linear extended state observer is switched, the observation bandwidth of the linear extended state observer to the disturbance is expanded, and a low-pass filter is added in the error feedback channel to suppress the high-frequency noise in the observation system; a quasi-resonant controller is connected in parallel in the disturbance observation channel of the improved linear extended state observer to construct a quasi-resonant improved linear extended observer; the control rate is designed, and a quasi-resonant improved linear anti-disturbance controller is constructed on the basis of the quasi-resonant improved linear extended state observer. The present invention can suppress constant disturbances and high-frequency AC disturbances, and realize high-performance control of the motor system.

Description

Technical Field

[0001] The present invention relates to the technical field of motor control, and in particular to an improved linear auto-disturbance rejection motor control method based on a quasi-resonant controller. Background Art

[0002] In the field of motor control technology, the operating performance of the motor depends largely on the electric drive control algorithm. Compared with the traditional proportional-integral controller, the linear active disturbance rejection controller has stronger anti-disturbance capability and parameter robustness. It estimates the total disturbance in the controlled system through the extended state observer, and then compensates the total disturbance inside the controller to reduce the impact of the disturbance on the controlled system. This active anti-disturbance method greatly increases the anti-disturbance performance of the controlled system. In addition, the linear active disturbance rejection controller greatly reduces the difficulty of parameter adjustment by adopting the parameter adjustment strategy of the "bandwidth method", and makes the parameter adjustment reasonable. Therefore, the application of linear active disturbance rejection controller in practical engineering is becoming more and more extensive.

[0003] However, in high-precision motor control situations, harmonic currents are generated due to factors such as AC motor structural processing deviations, inverter dead zone effects, grid high harmonics, and current measurement errors, which increase motor losses and cause torque pulsation, affecting the stable operation of the motor. Among them, motor structural deviations and inverter dead zone effects are the main factors that generate harmonic currents, which will cause periodic disturbances of 6 times the fundamental frequency in the motor system, thereby generating 5th and 7th phase current harmonics. In order to eliminate this high-frequency AC disturbance, we can start from the perspective of the controller. However, the linear anti-disturbance controller only has a good suppression effect on DC disturbances and low-frequency disturbances, and has poor suppression ability for high-frequency AC disturbances. It cannot effectively eliminate high-order harmonic disturbances in the motor control system and cannot meet the control requirements of high-precision application sites. Therefore, it is very necessary to try to eliminate the high-frequency AC disturbances generated by 5th and 7th harmonic currents in the motor control system. At present, there is an urgent need for a controller that can suppress high-frequency AC disturbances and suppress the 5th and 7th harmonic disturbances in the phase current. Summary of the invention

[0004] In view of the problems existing in the high-precision control systems of existing AC motors, the present invention provides an improved linear anti-disturbance motor control method based on a quasi-resonant controller, which utilizes the zero-pole characteristics of an expanded observer to improve the observation bandwidth of the expanded observer, and designs a quasi-resonant controller to suppress high-frequency AC disturbances in the control system. The design is reasonable, solves the shortcomings of the existing technology, and has good effects.

[0005] In order to achieve the purpose of the invention, the following technical solutions are adopted:

[0006] An improved linear active disturbance rejection motor control method based on a quasi-resonant controller comprises the following steps:

[0007] S1. Establish a mathematical model of the AC motor current loop;

[0008] S2, expand the total disturbance of the current loop into a new state variable;

[0009] S3. Construct a linear extended state observer and analyze the frequency domain characteristics of the extended state observer;

[0010] S4, switching the disturbance feedforward channel of the linear extended state observer, expanding the observation bandwidth of the linear extended state observer to the disturbance, and adding a low-pass filter in the error feedback channel to suppress the high-frequency noise in the observation system;

[0011] S5. A quasi-resonant controller is connected in parallel in the disturbance observation channel of the improved linear expansion state observer to construct a quasi-resonant improved linear expansion observer to achieve accurate observation of high-frequency AC disturbances.

[0012] S6. Design the control rate and construct a quasi-resonant improved linear anti-disturbance controller based on the quasi-resonant improved linear extended state observer to achieve zero-error tracking of the command signal and suppress DC and AC disturbances in the system.

[0013] Furthermore, in S1, the current loop mathematical model of the AC motor is established as follows:

[0014]

[0015] Among them, i d 、i q are d-axis and q-axis currents, respectively, and u d 、u q are the d-axis and q-axis input voltages, respectively, L(i d )、L(i q ) is the coupling between the d-axis and q-axis currents, D(i d )、D(i q ) is the quantity related to the current loop output, γ d , γ q They are the d-axis and q-axis current loop modeling errors and external disturbances respectively.

[0016] Furthermore, in S2, the parts of the motor control system d and q axis current loops other than the input variables are attributed to the total disturbance, including the coupling terms between the d and q axis currents, the quantities related to the current loop output, the external disturbances and the modeling errors; the total disturbance is expanded into a new state variable, the current loop mathematical model is simplified, and the state space model after the upgrade is as follows:

[0017]

[0018] in:

[0019]

[0020] Among them, f d 、f q are the total disturbances of the d-axis and q-axis current loops respectively; z1 is the d-axis and q-axis currents, that is, [i d ,i q ] T ; is the first-order derivative of z1; z2 is the expanded state variable of the d and q axis current loops, that is, [f d ,f q ] T ; is the first-order derivative of z2; b is the gain of the d and q axis current loop, that is, [b d ,b q ] T ; u is the d and q axis input voltage, that is, [u d ,u q ] T .

[0021] Furthermore, in said S3, a linear extended state observer is designed using the new state space model constructed in step S2, and a deviation between the measured value and the estimated value of the system output is used for proportional feedback, so that the observer system is wirelessly close to the real system, thereby estimating the total disturbance of the system;

[0022] The linear extended state observer is designed as follows:

[0023]

[0024] Where e is the observation error of the linear expansion observer; is the observed value of d and q axis current, that is is the total disturbance observation value of the d and q axis current loops, that is, β1 and β2 are the feedback gains of the linear expansion observer; y is the current loop output of the motor control system;

[0025] By performing Laplace transformation on equation (4), we can obtain the transfer function between the total disturbance observed by the linear extended state observer and the observation error:

[0026]

[0027] Then, by combining equations (2), (4), and (5), we can obtain the transfer function between the total disturbance observed by the linear extended state observer and the actual total disturbance of the system:

[0028]

[0029] From equation (6), we know that the observation effect of the linear extended state observer on the total disturbance has a low-pass characteristic, and the parameters h1 and h2 determine the observation bandwidth of the linear extended state observer.

[0030] Furthermore, in the step S4, the disturbance feedforward channel of the linear extended state observer constructed in step S3 is switched, and a low-pass filter is added to the error feedback channel to form an improved linear extended state observer, and the transfer function between the total disturbance observed by the improved linear extended state observer and the actual total disturbance is obtained:

[0031]

[0032] Among them, ω L Add the cutoff frequency of the new low-pass filter.

[0033] By comparing equation (6) and equation (7), it can be seen that the improved linear extended state observer adds a zero and a pole. Adding a zero helps to broaden the observation bandwidth of the system, and adding a pole helps to enhance the observation system's ability to suppress high-frequency measurement noise while increasing the bandwidth.

[0034] Furthermore, in S5, a quasi-resonant controller is connected in parallel to the disturbance observation channel of the improved linear extended state observer to construct a quasi-resonant improved linear extended state observer, and its expression is as follows:

[0035]

[0036] Where R is a quasi-resonant controller, and its transfer function is:

[0037]

[0038] The transfer function between the observed total disturbance and the actual total disturbance of the quasi-resonant improved linear extended state observer is:

[0039]

[0040] in,

[0041]

[0042] The quasi-resonant improved linear extended state observer can accurately observe AC disturbances at a fixed high frequency, enhance the observer's observation ability of high-frequency disturbances, and thus help the system suppress high-frequency disturbances.

[0043] Further, in S6, a quasi-resonant improved linear active disturbance rejection controller is constructed based on the quasi-resonant improved linear extended state observer, and the control rate is designed to be:

[0044]

[0045] Among them, v is the control system command value, k p Improvement of linear ADRC gain for quasi-resonance;

[0046] Ultimately, the command signal can be tracked without static errors, and the DC and AC disturbances in the system can be suppressed, thus improving the robustness of the system.

[0047] The present invention has the following beneficial effects:

[0048] The present invention provides an improved linear auto-disturbance rejection motor control method based on a quasi-resonant controller, which improves the traditional linear auto-disturbance rejection controller in view of the insufficient suppression capability of the traditional linear auto-disturbance rejection controller on high-frequency AC disturbances. First, the disturbance feedforward channel of the extended state observer is switched, and then a low-pass filter is added to the error feedback channel. Finally, a quasi-resonant controller is added to the disturbance observation channel, and the high-frequency AC disturbance is observed while the constant disturbance is observed, thereby completing the suppression of the constant disturbance and the high-frequency AC disturbance, and realizing high-performance control of the motor system.

[0049] Compared with the traditional linear anti-disturbance controller, the method of the present invention can reduce the harmonic distortion of the stator current of the running AC motor by 3%, and greatly reduce the 5th and 7th harmonic contents of the stator current. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 The frequency domain characteristic Bode diagram of the disturbance observation effect of the linear extended state observer in the present invention;

[0051] Figure 2 It is a structural block diagram of the proportional integral linear extended state observer in the present invention;

[0052] Figure 3 A comparison diagram of the frequency domain characteristics of the disturbance observation effects of the proportional integral linear extended state observer and the linear extended state observer in the present invention;

[0053] Figure 4 It is a structural block diagram of the improved linear extended state observer in the present invention;

[0054] Figure 5 A comparison diagram of the frequency domain characteristics of the disturbance observation effects of the improved linear extended state observer and the proportional integral linear extended state observer in the present invention;

[0055] Figure 6 is the Bode diagram of the quasi-resonant controller in the present invention;

[0056] Figure 7 A quasi-resonant improved linear extended state observer in the present invention;

[0057] Figure 8A comparison diagram of the frequency domain characteristics of the disturbance observation effects of the quasi-resonant improved linear extended state observer and the improved linear extended state observer in the present invention;

[0058] Fig. 9 It is a structural block diagram of the first-order quasi-resonant improved linear active disturbance rejection controller in the present invention;

[0059] Fig.10 A simplified structural block diagram of the quasi-resonant improved linear active disturbance rejection controller in the present invention;

[0060] Fig.11 The frequency domain characteristics between the system output and the actual disturbance of the quasi-resonant improved linear active disturbance rejection controller in the present invention;

[0061] FIG. 12( a ) and FIG. 12( b ) are waveform diagrams and spectrum diagrams of the A-phase stator current controlled by a conventional linear active disturbance rejection controller;

[0062] FIG. 13( a ) and FIG. 13( b ) are waveform diagrams and spectrum diagrams of the stator current of phase A controlled by the improved linear active disturbance rejection controller based on quasi-resonance; DETAILED DESCRIPTION

[0063] The specific implementation of the present invention is further described below in conjunction with specific embodiments:

[0064] An improved linear anti-disturbance motor control method based on a quasi-resonant controller is applicable to AC asynchronous motors and permanent magnet synchronous motors. This embodiment takes the permanent magnet synchronous motor as the control object. The present invention is described in detail in conjunction with the accompanying drawings. The specific implementation steps are as follows:

[0065] S1. Establish a mathematical model of the permanent magnet synchronous motor current loop. In the vector control strategy, the mathematical model of the current loop generally establishes the nonlinear relationship between the differential of the d-axis and q-axis currents after vector transformation and the d-axis and q-axis input voltages.

[0066] The permanent magnet synchronous motor is a complex nonlinear system, so the following assumptions need to be made when analyzing the mathematical model:

[0067] (1) Ignore magnetic circuit saturation and core eddy current loss;

[0068] (2) The three-phase stator windings are symmetrically distributed, with a phase difference of 120°;

[0069] (3) The rotor has no damping winding;

[0070] Based on the above assumptions, the stator voltage equations under the d and q axis coordinates are obtained:

[0071]

[0072] Among them, u d 、uq are the d-axis and q-axis components of the stator voltage respectively; i d 、i q are the d-axis and q-axis components of the stator current respectively; ψ d , q are the d-axis and q-axis components of the magnetic flux respectively; L d , L q are the d-axis and q-axis components of the inductance respectively; R s is the stator resistance; ω e is the rotor angular velocity.

[0073] According to formula (1), the mathematical model of the d-axis and q-axis current loops can be obtained as follows:

[0074]

[0075] S2, expand the total disturbance of the current loop into a new state variable;

[0076] From equation (2), it can be seen that the input of the current loop in the motor control system is the voltage and the output is the current. Therefore, in the mathematical model of the current loop, all quantities that are not related to the input can be called total disturbances, including coupling terms between the d and q axes, quantities related to the current, modeling errors, and external disturbances. The total disturbance is expanded into a new state variable to simplify the mathematical model of the current loop.

[0077] Since the designs of d-axis and q-axis controllers are similar, taking the d-axis as an example, the mathematical model of the d-axis current loop can be expressed as:

[0078]

[0079] Among them, b0 is the approximate control gain of the d-axis current loop model, and f is the total disturbance of the d-axis current loop. Their specific contents are shown in formula (4):

[0080]

[0081] Where γ is the d-axis current loop modeling error and external disturbance.

[0082] Let x1 = i d , x2 = f, the d-axis current loop state equation after the expanded state variable is obtained, as shown in formula (5):

[0083]

[0084] Where y is the d-axis current loop output.

[0085] S3. Construct a linear extended state observer (ESO) for equation (5) and analyze the frequency domain characteristics of the extended state observer;

[0086] The current loop model is a first-order system, so the second-order linear extended state observer is established as follows:

[0087]

[0088] Where e is the error between the output of the system observed by the linear extended state observer and the actual output of the system, are the estimated values ​​of the state variables x1 and x2 respectively, and β1 and β2 are the feedback gains of the extended state observer.

[0089] From the mathematical model of the linear extended state observer, we can get:

[0090]

[0091] Performing Laplace transform on equation (7) yields:

[0092]

[0093] Substituting formula (8) into formula (5) yields:

[0094]

[0095] Combining formula (6), we can get:

[0096]

[0097] According to formula (10), the transfer function between the observation error of the linear extended state observer and the actual total disturbance of the system can be obtained:

[0098]

[0099] According to formula (6), the transfer function between the total disturbance observed by the linear extended state observer and the observer observation error can be obtained:

[0100]

[0101] Combining equations (11) and (12), we can obtain the transfer function between the total disturbance observed by the observer and the actual disturbance of the system:

[0102]

[0103] From equation (13), it can be seen that the observation effect of the linear extended state observer on disturbance has a low-pass characteristic, and its Bode diagram is as follows: Figure 1 As shown, the performance of the observer can be adjusted by adjusting the parameters β1 and β2. In order to ensure the stability and excellent performance of the system, the parameter determination method adopted by the present invention is to set β1 = 2ω0, β2 = ω0 2 , where ω0 is the bandwidth of the linear expansion observer.

[0104] S4, switching the disturbance feedforward channel of the linear extended state observer, expanding the observation bandwidth of the linear extended state observer to the disturbance, and adding a low-pass filter in the error feedback channel to suppress the high-frequency noise in the observation system;

[0105] First, in order to remove the coupling between the parameters of the linear ADRC and broaden the observer bandwidth, the observation disturbance output channel of the linear extended state observer is switched. Its structural block diagram is shown in Figure 2 As shown. Since the observed disturbance and the observed error are in a proportional-integral relationship, it can be called a proportional-integral linear extended state observer, and its state space expression is:

[0106]

[0107] By analyzing equation (14) according to the method from equation (7) to equation (13) in S3, the transfer function between the observed disturbance and the actual disturbance of the proportional integral linear extended state observer can be obtained:

[0108]

[0109] It can be seen that Equation (15) has one more zero point than Equation (13), so its frequency domain amplitude will increase at the zero point β2 / β1 and decrease at the bandwidth frequency point, thus widening the bandwidth of the observer. Compared with the Bode diagram of the conventional linear extended state observer, Figure 3 shown.

[0110] pass Figure 3 It can be seen that due to the addition of a zero point in the system, the observer has insufficient ability to suppress high-frequency noise, which easily introduces high-frequency noise into the system. For this reason, the present invention adds a first-order low-pass filter in the error feedback channel and suppresses high-frequency noise by reasonably adjusting the cutoff frequency of the low-pass filter. The structural block diagram is shown in FIG. Figure 4 As shown. The spatial expression of the improved linear extended state observer is:

[0111]

[0112] Among them, ω L is the cutoff frequency of the low-pass filter, e L is the observer observation error after being filtered by a low-pass filter.

[0113] By analyzing equation (16) according to the method from equation (7) to equation (13) in step 3, the transfer function between the observed disturbance and the actual disturbance of the improved linear extended state observer can be obtained:

[0114]

[0115] in,

[0116]

[0117] G c1 (s) represents the transfer function between the observer observation disturbance and the filtered observation error, G L (s) represents the transfer function of the low-pass filter.

[0118] The frequency domain characteristics of the improved linear extended state observer and the proportional integral linear extended state observer are compared. Figure 5 It can be seen that the bandwidth of the improved linear extended state observer is the same as that of the proportional integral extended state observer, but the improved linear extended state observer has a better suppression effect on high-frequency noise than the proportional integral extended state observer.

[0119] S5. A quasi-resonant controller is connected in parallel in the disturbance observation channel of the improved linear expansion state observer to construct a quasi-resonant improved linear expansion observer to achieve accurate observation of high-frequency AC disturbances.

[0120] Because the bandwidth of the improved linear extended state observer cannot be infinite, some high-frequency disturbances cannot be observed, and the motion system often generates high-frequency disturbances due to the inverter dead zone effect and the odd harmonics of the power grid, mainly the 5th and 7th harmonics. Therefore, in order to suppress the AC disturbance in the motion system, the present invention introduces a quasi-resonant controller into the improved linear extended state observer by using the inner membrane principle to form a quasi-resonant improved linear extended state observer.

[0121] The transfer function of the quasi-resonant controller is as follows:

[0122]

[0123] Among them, ω r is the resonant frequency, ω c is the damping coefficient, k r is the gain factor at the resonant frequency.

[0124] The Bode diagram of a quasi-resonant controller is shown in Figure 6 As shown, by adjusting ω in the quasi-resonant controller r ,ω c and k r The three parameters can change the resonant frequency, damping at the resonant frequency and resonant gain of the quasi-resonant controller respectively.

[0125] The quasi-resonant controller is introduced into the improved linear extended state observer to form a quasi-resonant improved linear extended state observer. Its structure diagram is shown in the figure below. Figure 7 As shown. The spatial expression of the quasi-resonant linear extended state observer is:

[0126]

[0127] Where R is a quasi-resonant controller, and the expression is shown in formula (19).

[0128] By analyzing equation (20) according to the method from equation (7) to equation (13) in step 3, the transfer function between the observed disturbance and the actual disturbance of the quasi-resonant improved linear extended state observer can be obtained:

[0129]

[0130] in,

[0131]

[0132] The frequency domain characteristics of the quasi-resonant improved linear extended state observer and the improved linear extended state observer are compared. Figure 8 It can be seen that the quasi-resonant improved linear extended state observer proposed in the present invention can observe disturbances at the resonant frequency beyond the bandwidth.

[0133] S6. Design the control rate and construct a quasi-resonant improved linear anti-disturbance controller based on the quasi-resonant improved linear extended state observer to achieve zero-static error tracking of the command signal, suppress DC and AC disturbances in the system, and improve the robustness of the system.

[0134] According to the conventional method, a quasi-resonant improved linear active disturbance rejection controller is designed based on the quasi-resonant improved extended state observer, and its control rate is designed as follows:

[0135]

[0136] Among them, u d0 for:

[0137]

[0138] Among them, r is the d-axis current loop command value, k p is the proportional coefficient of the control rate.

[0139] By combining equations (20), (23) and (24), we can get the quasi-resonant improved linear active disturbance rejection controller. In summary, the block diagram of the quasi-resonant improved linear active disturbance rejection controller based on the first-order system is as follows: Fig. 9 shown.

[0140] After Laplace transforming equation (20), we can get:

[0141]

[0142] Among them, G L (s) and G c1 (s) is as shown in formula (18) and formula (22).

[0143] Substituting equation (25) into equation (23), we can obtain

[0144]

[0145] make

[0146]

[0147] According to equations (26) and (27), the simplified block diagram of the quasi-resonant improved linear active disturbance rejection controller can be obtained as follows: Fig.10 shown.

[0148] in accordance with Fig.10 The relationship between the system output and the actual disturbance can be obtained.

[0149]

[0150] Among them, G p (s) is the system object transfer function, which can be expressed as a first-order inertia link:

[0151]

[0152] Among them, a0 and b are system object parameters.

[0153] The frequency domain characteristics between the system output and the actual disturbance under the control of the quasi-resonant improved linear active disturbance rejection controller are as follows: Fig.11 As shown. It can be seen that the system output at the resonant frequency can have a good suppression effect on disturbances.

[0154] Experimental results: Figure 12 shows the A-phase stator current waveform when the current loop is controlled by a conventional linear anti-disturbance controller at constant speed, and the spectrum after fast Fourier transformation. From Figure 12(b), it can be seen that the harmonic distortion of the A-phase stator current is 10.38%, and the 5th and 7th harmonic contents are relatively high, which can easily affect the stability of the system. It can be seen that the conventional linear anti-disturbance controller is not capable of suppressing high-order harmonic disturbances.

[0155] Figure 13 is a waveform diagram of the A-phase stator current when the current loop is controlled by the quasi-resonant improved linear anti-disturbance controller proposed by the present invention and is running at a constant speed, as well as a spectrum diagram after fast Fourier transformation. It can be seen that the method of the present invention can effectively suppress the 5th and 7th harmonics compared with the conventional linear anti-disturbance controller, and its harmonic distortion is greatly reduced to 7.31%. This experimental result further verifies that the quasi-resonant linear anti-disturbance controller has a good ability to suppress high-frequency AC disturbances.

[0156] Of course, the above description is not a limitation of the present invention, and the present invention is not limited to the above examples. Changes, modifications, additions or substitutions made by technicians in this technical field within the essential scope of the present invention should also fall within the protection scope of the present invention.

Claims

1. An improved linear active disturbance rejection motor control method based on a quasi-resonant controller, characterized in that: The following steps are involved: S1. Establish a mathematical model of the AC motor current loop; S2, expand the total disturbance of the current loop into a new state variable; S3. Construct a linear extended state observer and analyze the frequency domain characteristics of the extended state observer; S4, switching the disturbance feedforward channel of the linear extended state observer, expanding the observation bandwidth of the linear extended state observer to the disturbance, and adding a low-pass filter in the error feedback channel to suppress the high-frequency noise in the observation system; S5. Connect a quasi-resonant controller in parallel in the disturbance observation channel of the improved linear expansion state observer to construct a quasi-resonant improved linear expansion observer to achieve accurate observation of high-frequency AC disturbances. S6. Design the control rate, and construct a quasi-resonant improved linear anti-disturbance controller based on the quasi-resonant improved linear extended state observer to achieve zero-static error tracking of the command signal and suppress DC and AC disturbances in the system; In S1, the current loop mathematical model of the AC motor is established as: Among them, i d 、i q are d-axis and q-axis currents, respectively, and u d 、u q are the d-axis and q-axis input voltages, respectively, L(i d )、L(i q ) is the coupling between the d-axis and q-axis currents, D(i d )、D(i q ) is the quantity related to the current loop output, γ d , γ q They are the d-axis and q-axis current loop modeling errors and external disturbances respectively; In S2, the parts of the motor control system d and q axis current loops other than the input variables are attributed to the total disturbance, including the coupling terms between the d and q axis currents, the quantities related to the current loop output, the external disturbances and the modeling errors; the total disturbance is expanded into a new state variable, the current loop mathematical model is simplified, and the state space model after the upgrade is as follows: in: Among them, f d 、f q are the total disturbances of the d-axis and q-axis current loops respectively; z1 is the d-axis and q-axis currents, that is, [i d ,i q ] T ; is the first-order derivative of z1; z2 is the expanded state variable of the d and q axis current loops, that is, [f d ,f q ] T ; is the first-order derivative of z2; b is the gain of the d and q axis current loops, that is, [b d ,b q ] T ; u is the d and q axis input voltage, that is, [u d ,u q ] T ; In said S3, a linear extended state observer is designed using the new state space model constructed in step S2, and a proportional feedback is performed using the deviation between the measured value and the estimated value of the system output, so that the observer system is infinitely close to the real system, thereby estimating the total disturbance of the system; The linear extended state observer is designed as follows: Where e is the observation error of the linear expansion observer; is the observed value of d and q axis current, that is is the total disturbance observation value of the d and q axis current loops, that is, h1 and h2 are the feedback gains of the linear expansion observer; y is the current loop output of the motor control system; By performing Laplace transformation on equation (4), we can obtain the transfer function between the total disturbance observed by the linear extended state observer and the observation error: Then, by combining equations (2), (4), and (5), we can obtain the transfer function between the total disturbance observed by the linear extended state observer and the actual total disturbance of the system: From equation (6), we know that the observation effect of the linear extended state observer on the total disturbance has a low-pass characteristic, and the parameters h1 and h2 determine the observation bandwidth of the linear extended state observer; In step S4, the disturbance feedforward channel of the linear extended state observer constructed in step S3 is switched, and a low-pass filter is added to the error feedback channel. The observation error e of the linear extended state observer is obtained after passing through the low-pass filter. L , constitute an improved linear extended state observer, and obtain the transfer function between the total disturbance observed by the improved linear extended state observer and the actual total disturbance: Among them, ω L Add the cutoff frequency of the new low-pass filter; In S5, a quasi-resonant controller is connected in parallel to the disturbance observation channel of the improved linear extended state observer to construct a quasi-resonant improved linear extended state observer, and its expression is as follows: Where R is a quasi-resonant controller, and its transfer function is: Among them, k r is the gain coefficient at the resonant frequency, ω c is the damping coefficient, ω r is the resonant frequency; The transfer function between the observed total disturbance and the actual total disturbance of the quasi-resonant improved linear extended state observer is: in, In S6, a quasi-resonant improved linear active disturbance rejection controller is constructed based on the quasi-resonant improved linear extended state observer, and the control rate is designed to be: Among them, v is the control system command value, k p Improvement of linear ADRC gain for quasi-resonance; Ultimately, the command signal can be tracked without static errors, and the DC and AC disturbances in the system can be suppressed, thus improving the robustness of the system.

Images