Permanent magnet synchronous motor sensorless control method based on consistency of current transient and system disturbance

Abstract

The application discloses a permanent magnet synchronous motor position sensorless control method based on current transient and system disturbance consistency adjustment, generates a current transient index according to the discrete change of an actual stator current, generates a system disturbance transient index according to the total disturbance estimation of a linear extended state observer in the last sampling period; obtains a transient consistency coefficient by combining the two indexes, adjusts the current change rate compensation coefficient online between the preset upper and lower limits; after limiting the compensation amount, the compensation amount is superimposed to the alpha-beta axis current observation error, and a sliding mode injection signal is generated before a continuous sliding mode function, and the back electromotive force and the rotor position and speed are reconstructed. Compared with the prior art, the application makes the disturbance observation result reversely participate in the adjustment of the adaptive sliding mode observer, can enhance the transient compensation when the real load disturbance occurs, can inhibit the false enhancement and noise amplification when the speed instruction changes, sampling burrs or constant load steady state, and improves the estimation accuracy and anti-interference performance of the position sensorless control.

Description

Technical Field

[0001] This invention relates to the field of permanent magnet synchronous motor control technology, specifically to a sensorless control method for permanent magnet synchronous motors based on the coordinated adjustment of current transients and system disturbance consistency. Background Technology

[0002] Permanent magnet synchronous motors (PMSMs) are characterized by high power density, high operating efficiency, fast dynamic response, and compact structure, and are widely used in new energy vehicles, industrial servo systems, robotics, aerospace, and high-performance electric drives. Vector control of PMSMs typically requires obtaining rotor position and speed information. Traditional mechanical position sensors increase the size, cost, and installation complexity of the drive system and may affect system reliability under conditions such as high temperature, vibration, and dust. Therefore, sensorless control technology that uses stator voltage, current, and other electrical quantities to estimate rotor position and speed has become an important research direction in the field of PMSM control.

[0003] Existing sensorless control methods typically use sliding mode observers to estimate the motor's back electromotive force (EMF) to obtain rotor position and speed. However, traditional sliding mode observers often employ fixed sliding mode gain and sign switching functions, making it difficult to balance convergence speed and chatter suppression under different speed conditions. Furthermore, the back EMF filtering process is prone to phase lag, which reduces the accuracy of position and speed estimation when the speed changes or the load changes abruptly.

[0004] Furthermore, traditional speed loops often employ PI controllers, which have limited ability to suppress changes in motor parameters, load disturbances, and observation errors, easily leading to speed fluctuations and slowed dynamic response. Therefore, it is necessary to investigate a sensorless control method for permanent magnet synchronous motors that balances rotor state estimation accuracy with speed loop disturbance rejection performance. Summary of the Invention

[0005] Objective: To address the challenges of existing sliding mode observers using fixed current rate of change compensation, which struggle to distinguish between real system disturbances and rapid current changes caused by non-disturbance factors such as sampling glitches and switching noise, as well as the lack of a coordinated disturbance state adjustment channel between existing sliding mode observers and linear active disturbance rejection controllers, this invention provides a sensorless control method for permanent magnet synchronous motors based on the consistency adjustment of current transients and system disturbances. This method utilizes the total disturbance estimate output by a linear extended state observer to extract system disturbance transient indices and performs consistency judgment with stator current transient indices. Based on the consistency result, the current rate of change compensation coefficient is adjusted online, enhancing the compensation effect when real system transients occur and weakening it when the two types of transients are inconsistent. This improves the stability of back EMF reconstruction, rotor position and speed estimation, and speed loop disturbance rejection control.

[0006] Technical solution: This invention provides a sensorless control method for permanent magnet synchronous motors based on the coordinated adjustment of current transients and system disturbance consistency, comprising the following steps:

[0007] S1 collects and processes the three-phase stator current and DC bus voltage of the permanent magnet synchronous motor;

[0008] S2 calculates the stator current based on the current sampling time and the previous sampling time. The discrete change in shaft stator current is used to generate a current transient index; the system disturbance transient index is obtained by generating the total disturbance estimate of the linear extended state observer from the previous sampling period and delayed by at least one sampling period.

[0009] S3 generates a transient consistency coefficient based on the current transient index and the delayed system disturbance transient index, and adjusts the current change rate compensation coefficient online based on the transient consistency coefficient.

[0010] S4 calculates the current change rate compensation amount based on the current change rate compensation coefficient and the stator current discrete change amount, and then adds the current change rate compensation amount after the limit to the current observation error to form the corrected current observation error.

[0011] S5 inputs the corrected current observation error into the adaptive sliding mode observer to generate... Shaft sliding mode injection signal, reconstruction The shaft back electromotive force is obtained, and the rotor electrical angle estimate and speed estimate for the current sampling period are obtained through a phase-locked loop;

[0012] The speed estimate of S6 in the current sampling period is used to estimate the total system disturbance through a linear extended state observer, generating... The shaft torque current reference value is used to generate system disturbance transient indices for use in the next sampling period based on the total disturbance estimate of the current sampling period.

[0013] The S7 performs current closed-loop control based on torque and current reference values, and drives the inverter through SVPWM to achieve sensorless control of the permanent magnet synchronous motor.

[0014] Furthermore, the linearly extended state observer uses the rotational speed estimate obtained by the adaptive sliding mode observer as the observation input, and its state equation is:

[0015] ;

[0016] ;

[0017] ;

[0018] in, This is the error in velocity observation; It is a speed state estimate; This is the estimate of the total system disturbance; The rotational speed estimate obtained by the adaptive sliding mode observer; for Actual shaft current; To control the nominal gain value; , This represents the gain of the linearly extended state observer.

[0019] Furthermore, the specific process for obtaining the system disturbance transient index is as follows:

[0020] First, calculate the dynamic disturbance component at the current sampling time based on the slowly varying baseline from the previous sampling time, and then update the slowly varying baseline, specifically as follows:

[0021] ;

[0022] ;

[0023] in, The slowly varying baseline for the total disturbance estimate; For dynamic disturbance components; The baseline tracking coefficient, and ;

[0024] Convert the dynamic disturbance component into an equivalent disturbance current: ;

[0025] Based on the equivalent disturbance current, generate the unfiltered system disturbance transient index:

[0026] ;

[0027] And low-pass filtering is applied to the unfiltered system disturbance transient indices:

[0028] ;

[0029] in, The dynamic disturbance dead zone threshold; The full-scale threshold for dynamic disturbance; The filter coefficients are the transient indexes of system disturbances. This means that the input value is restricted to the range of 0 to 1.

[0030] Furthermore, the specific current transient parameters are as follows:

[0031] Generate transient current indices based on discrete changes in stator current:

[0032] ;

[0033] in, This is the normalized reference value for the discrete variation of the stator current vector; , for Discrete variation of shaft stator current.

[0034] Furthermore, the transient consistency coefficient is:

[0035] ;

[0036] ;

[0037] in, This is a delayed transient indicator of system disturbances; The unfiltered transient consistency coefficient; For current transient indicators, The unfiltered transient consistency coefficient; The transient consistency coefficient after filtering; These are the consistency filter coefficients.

[0038] Furthermore, the current change rate compensation coefficient is adjusted online according to the following formula:

[0039] ;

[0040] in, This is the lower limit of the current change rate compensation coefficient; This is the upper limit of the current change rate compensation coefficient; the... The compensation amount for the rate of change of shaft current is:

[0041] ;

[0042] ;

[0043] in, The sampling period; This is the limit value for the current change rate compensation. and express Discrete variation of shaft stator current.

[0044] Furthermore, the sliding mode gain and sliding mode boundary layer width of the adaptive sliding mode observer are dynamically adjusted based on the filtered rotational speed estimate, as follows:

[0045] ;

[0046] ;

[0047] in, This is the lower limit of the sliding mode gain; This is the upper limit of the sliding mode gain; This is the filtered estimated rotational speed. The reference speed for adjusting the sliding mode gain; This is the initial boundary layer width; This is the boundary layer dynamic adjustment coefficient; and These are the lower and upper limits of the boundary layer width, respectively.

[0048] Furthermore, when obtaining the rotor electrical angle estimate and speed estimate for the current sampling period through the phase-locked loop, the phase-locked loop gain scheduling factor is also set based on the filtered speed estimate:

[0049] ;

[0050] in, This is the filtered estimated speed value. The base speed for phase-locked loop gain scheduling;

[0051] And adjust the phase-locked loop's proportional gain and integral gain online:

[0052] ;

[0053] ;

[0054] in, and These are the lower and upper limits of the proportional gain. and These are the lower and upper limits of the integral gain. and For proportional gain and integral gain.

[0055] Furthermore, the aforementioned The shaft torque current reference value is:

[0056] ;

[0057] ;

[0058] ;

[0059] ;

[0060] in, Based on the control quantity, Given a rotational speed; For speed error feedback gain; For acceleration feedback gain; For velocity state estimation; This is the estimate of the total system disturbance; This is an estimate of the acceleration. The filtered value of the acceleration estimate; These are the filter coefficients for the acceleration estimate; for Shaft current feedforward compensation; for Shaft torque current reference value and limit value; This is the nominal value for control gain.

[0061] Beneficial effects:

[0062] 1. This invention extracts the dynamic component of the total LESO disturbance estimate relative to the slowly varying baseline, and forms a system disturbance transient index through dead zone, normalization, and filtering. This reduces the impact of constant load, slowly varying parameters, and observation noise on disturbance transient judgment and improves the accuracy of load change identification.

[0063] 2. This invention constructs a current transient index based on the discrete change of stator current, and generates a transient consistency coefficient with the delayed system disturbance transient index, so as to distinguish between the real load disturbance and the rapid current change caused by speed command change, sampling glitch and switching ripple, etc., and reduce the risk of false triggering of transient compensation.

[0064] 3. The present invention adjusts the current change rate compensation coefficient between preset upper and lower limits based on the transient consistency coefficient, and applies the limited compensation amount to the current observation error of the sliding mode observer, so as to enhance transient compensation when a real disturbance occurs, while maintaining a low compensation intensity in the non-disturbance state, thus taking into account both dynamic compensation capability and steady-state noise immunity performance.

[0065] 4. This invention dynamically adjusts the sliding mode gain based on the filtered speed estimate and dynamically adjusts the sliding mode boundary layer width based on the deviation between the instantaneous speed and the filtered speed, which can take into account the observation convergence speed, steady-state accuracy and sliding mode chatter suppression effect under different speed conditions.

[0066] 5. This invention employs sliding mode injection differential feedforward and phase compensation to reduce information loss and phase lag caused by back EMF filtering, and sets the phase-locked loop gain scheduling factor according to the rotational speed, so that the phase-locked loop can take into account both dynamic tracking performance and steady-state noise immunity in different rotational speed ranges; limiting the integral update of the phase-locked loop under low back EMF conditions can suppress abnormal angle and rotational speed estimation.

[0067] 6. This invention utilizes the total system disturbance estimated by LESO for velocity loop compensation and introduces filtered acceleration information to improve dynamic damping; simultaneously, the dynamic information of LESO disturbance is used inversely to adjust the compensation intensity of ASMO, forming a bidirectional synergistic adjustment between ASMO and LADRC, thereby improving the estimation accuracy, dynamic response capability and disturbance rejection performance of the system. Attached Figure Description

[0068] Figure 1 This is a block diagram of the overall control system of the present invention.

[0069] Figure 2 This is a schematic diagram of the transient consistency fusion and compensation coefficient generation structure of the present invention.

[0070] Figure 3 This is a schematic diagram of the internal structure of the consistent collaborative adaptive sliding mode observer of the present invention.

[0071] Figure 4 This is a comparison chart of the electric angular velocity estimation of the traditional scheme and the improved method of the present invention.

[0072] Figure 5 This is a comparison diagram of the electric angular velocity estimation of fixed kdi and adaptive kdi according to the present invention.

[0073] Figure 6 This is the curve showing the change of the transient consistency index and kdi in this invention. Detailed Implementation

[0074] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0075] like Figures 1 to 3 As shown in the figure, this embodiment discloses a sensorless control method for a permanent magnet synchronous motor based on the consistency adjustment of current transients and system disturbances. The method steps are as follows:

[0076] S1: Establish a mathematical model of the permanent magnet synchronous motor and construct a sensorless control system.

[0077] S1.1: Under the condition of neglecting magnetic circuit saturation, eddy current losses, and spatial harmonics, the permanent magnet synchronous motor in The stator current model in a two-phase stationary coordinate system is as follows:

[0078]

[0079]

[0080] in, , They are respectively shaft and Shaft-stator voltage; , They are respectively shaft and Shaft stator current;

[0081] Stator resistance; The stator equivalent inductance; , They are respectively shaft and Axis back electromotive force.

[0082] The relationship between back electromotive force and rotor electrical angle and rotor electrical angular velocity is as follows:

[0083]

[0084]

[0085] in, For permanent magnet flux linkage; The rotor electrical angle; ω is the rotor's electric angular velocity.

[0086] S1.2: The mechanical motion equation of the permanent magnet synchronous motor is:

[0087]

[0088] In Adoption When controlled in this manner, the electromagnetic torque is:

[0089]

[0090] And there are:

[0091]

[0092] in, It is the moment of inertia; The damping coefficient; This is the load torque; Electromagnetic torque; It is the extreme logarithm; It is the mechanical angular velocity; This is the q-axis current.

[0093] S1.3: Collect three-phase stator current , , Obtained by Clarke transform , ; Obtain the inverter output or reconstruct it from a voltage reference. , and will , , , Input is the adaptive sliding mode observer (ASMO). The ASMO outputs an estimate of the rotor electrical angle. and estimated electric angular velocity ;in, Used for Park transform and inverse Park transform. Alternatively, the mechanical speed estimate obtained from its conversion can be used as the speed feedback value of LADRC.

[0094] S2: Construct a current observer and generate stator current transient indices.

[0095] S2.1: As Figure 3 As shown, based on the permanent magnet synchronous motor Establish a current state observer using the shaft current model:

[0096]

[0097]

[0098] in, , They are respectively shaft and Shaft stator current observations; , They are respectively shaft and Shaft sliding mode injection signal.

[0099] Current observation error is defined as:

[0100]

[0101]

[0102] S2.2: Construct the discrete change in current using the difference between the stator current at the current sampling time and the previous sampling time:

[0103]

[0104]

[0105] To prevent artificial differential spikes from occurring when the system is put into operation due to the previous state not being initialized, the initial value of the previous sampling current is set to the actual sampling current when the control system is put into operation.

[0106] Generate transient current indices based on the discrete changes in the stator current vector:

[0107]

[0108] in, This serves as the normalized reference value for the discrete change in current. This indicates that the input is restricted to the range of 0 to 1. In this embodiment, .

[0109] S3: Adjust the current change rate compensation coefficient online based on the consistency between the current transient and the system disturbance transient.

[0110] S3.1: As Figure 2 As shown, the system disturbance transient index formed in the previous sampling period is input after a one-beat delay:

[0111]

[0112] Control system initialization command The current transient index is compared with the delayed system disturbance transient.

[0113] Multiply the indices to obtain the original transient consistency coefficient:

[0114]

[0115] This product relationship applies when rapid changes in current and system disturbances occur simultaneously. Only when the current change is mainly caused by sampling glitches, switching ripples, or measurement noise, and the system disturbance dynamics are not obvious, will the change become more pronounced. Keep it small.

[0116] S3.2: To suppress spikes in a single sampling period and smooth the changes in consistency adjustment, a first-order low-pass filter is applied to the original transient consistency coefficients:

[0117]

[0118]

[0119] in, The transient consistency coefficient after filtering; This is the time constant for the consistency filter. In this embodiment, .

[0120] S3.3: Adjust the current change rate compensation coefficient online based on the filtered transient consistency coefficient:

[0121]

[0122] in, and These are the lower and upper limits of the current change rate compensation coefficient, respectively. In this embodiment, , .

[0123] S3.4: Generate compensation amount based on the stator current discrete change rate:

[0124]

[0125]

[0126]

[0127]

[0128] in, As the compensation limit value, this embodiment takes... The compensation amount after limiting is added to the current observation error to form the correction input of the continuous sliding mode function:

[0129]

[0130]

[0131] Through the above processing, the current change rate compensation intensity is increased when the two types of transient information have high consistency, and the compensation coefficient is brought close to the lower limit when the two types of transient information have low consistency, thereby reducing noise differential amplification and unnecessary overcompensation.

[0132] S4: Generate sliding mode injection signal based on corrected current observation error, and estimate rotor electrical angle and speed.

[0133] S4.1: Dynamically adjust the sliding mode gain based on the filtered velocity estimate:

[0134]

[0135] in, and These are the lower and upper limits of the sliding mode gain, respectively. This is the filtered velocity estimate; This serves as the speed adjustment reference value. In this embodiment, , , .

[0136] S4.2: Dynamically adjust the boundary layer width based on the deviation between the PLL instantaneous velocity estimate and the filtered velocity estimate.

[0137]

[0138] in, The reference boundary layer width; This is a dynamic adjustment coefficient; and These are the lower and upper limits of the boundary layer width, respectively. In this embodiment, , , , .

[0139] S4.3: The sign function is replaced by a continuous saturation function, which is defined as follows:

[0140]

[0141] The sliding mode injection signal is:

[0142]

[0143]

[0144] The current observer is updated using the forward Euler method:

[0145]

[0146]

[0147] S4.4: Perform a first-order low-pass filter on the sliding mode injection signal to obtain the back electromotive force filter value:

[0148]

[0149]

[0150]

[0151] In this embodiment, the back EMF low-pass filter cutoff frequency To reduce the dynamic lag caused by low-pass filtering, a sliding mode injection difference feedforward correction is introduced:

[0152]

[0153]

[0154] in, As the lead correction factor, in this embodiment, we take... .

[0155] S4.5: Calculate the back electromotive force angle based on the compensated back electromotive force estimate:

[0156]

[0157] Calculate the phase compensation angle based on the estimated filter velocity and the low-pass filter time constant:

[0158]

[0159]

[0160] S4.6: Employ speed-dependent dynamic PLL parameters. Let:

[0161]

[0162] The proportional gain and integral gain of the PLL are as follows:

[0163]

[0164]

[0165] In this embodiment, , , , , The PLL angle error signal is:

[0166]

[0167] When the back electromotive force amplitude reaches a preset threshold, update the PLL integral state, instantaneous velocity, and angle:

[0168]

[0169]

[0170]

[0171] The speed output uses a first-order filter:

[0172]

[0173] S4.7: Calculate the back electromotive force amplitude:

[0174]

[0175] when When, perform the above PLL error correction; when At this time, pause the error correction of the PLL integral state, let the instantaneous velocity adopt the filtered velocity estimate, and update the angle recursively according to the following formula:

[0176]

[0177]

[0178] S5: Use LADRC to estimate the total system disturbance, generate the q-axis current reference value, and form the system disturbance transient index for the next sampling period.

[0179] S5.1: As Figure 1 and Figure 2 As shown, the speed estimate output by ASMO / PLL is used as the LADRC speed feedback signal, and the actual q-axis current is used as the signal. As the LESO control input, the speed loop of the permanent magnet synchronous motor can be written as:

[0180]

[0181] in, This includes the total system disturbance, which incorporates load torque, friction, parameter deviations, etc. This is the nominal value for control gain.

[0182] The velocity observation error of LESO is:

[0183]

[0184] The LESO equation of state is:

[0185]

[0186]

[0187] in, For velocity state estimation; This is the estimate of the total system disturbance; , This is the LESO gain. The sampling period is used in the discrete implementation. The above state equations are recursively derived. To avoid outliers affecting the controller, It can be limited to a range corresponding to the preset maximum equivalent disturbance current.

[0188] S5.2: Due to It simultaneously includes steady-state load, friction, parameter deviation, and dynamic disturbance, and cannot be directly applied. Determine if the system is in a transient state. Use a slowly varying baseline to extract the dynamic components of the system disturbance.

[0189]

[0190]

[0191]

[0192] in, The slowly varying baseline for the total disturbance estimate; For the dynamic components of system disturbance; This is the baseline tracking time constant.

[0193] Convert the dynamic components of the system disturbance into an equivalent disturbance current:

[0194]

[0195] Normalization is performed using the dead zone and full-scale threshold:

[0196]

[0197] in, The system disturbance dynamic dead zone threshold, This is the dynamic full-scale threshold for system disturbance.

[0198] The normalized result is low-pass filtered to form the system disturbance transient index:

[0199]

[0200]

[0201] in, The time constant for filtering the system disturbance transient index. Formed by unit delay module And input the consistency adjustment module in the next sampling period.

[0202] S5.3: Based on the given speed and LESO velocity state estimator Calculate the speed tracking error:

[0203]

[0204] Acceleration estimates are calculated based on the velocity loop model:

[0205]

[0206] Low-pass filtering is applied to the acceleration estimate:

[0207]

[0208]

[0209] The speed state error feedback control quantity is:

[0210]

[0211] Based on the LESO total disturbance estimate, disturbance compensation is performed to obtain the q-axis torque current reference value:

[0212]

[0213] in, For speed error feedback gain; For acceleration feedback gain; This is the q-axis current feedforward. Limit the q-axis current reference value; .

[0214] S6: Based on the q-axis torque current reference value output by LADRC Input q-axis current closed-loop controller, d-axis current reference value set to By estimating the electrical angle right , Perform the Park transform to obtain the actual current component. , The d-axis and q-axis current errors are respectively:

[0215]

[0216] The current error is obtained after PI adjustment. , Then, the estimated electrical angle output from the ASMO / PLL was used. Perform an inverse Park transformation to obtain the voltage reference value in the α−β coordinate system. , The SVPWM modulation module generates inverter switching signals based on the voltage reference value, and the inverter outputs three-phase voltage to drive the PMSM. The system realizes a complete closed-loop control process from ASMO position observation-free, LADRC speed disturbance compensation, current closed-loop control to SVPWM inverter drive.

[0217] In this embodiment, a sensorless control model of a permanent magnet synchronous motor is built using the MATLAB / Simulink platform. The core modules of the simulation model include: a permanent magnet synchronous motor module, a three-phase inverter module, an SVPWM modulation module, a current closed-loop control module, an adaptive sliding mode observer (ASMO) module, a dynamic phase-locked loop (PLL) module, and a linear active disturbance rejection speed controller (LADRC) module containing a linear extended state observer (LESO). The motor parameters simulated in this embodiment are shown in Table 1 below:

[0218] Table 1 PMSM Parameter Table

[0219] Stator resistance 0.11Ω Stator inductor 0.00097H Permanent magnet flux 0.1119Wb Moment of inertia 0.0016 kg·m² Damping coefficient 0.0002024 N·m·s Extreme logarithm 4

[0220] The sampling time was set to Ts = 1 × 10⁻⁶ s, and the total simulation time was 3 s. The motor was started under no-load conditions, and the reference speed was set to 1500 r / min. At 1 s of simulation time, the speed was reduced to 800 r / min; at 1.5 s, the load torque was increased by 10 N·m; at 2 s, the speed was increased to 1200 r / min; and at 2.5 s, the load torque was increased again by 10 N·m, with the simulation time remaining at 3 s. The advantages of this invention were verified from aspects such as rotor electrical angle estimation error, electrical angular velocity tracking, α-axis back electromotive force estimation, and adaptive parameter changes. The core verification results are as follows:

[0221] See Figure 4 Electrical angular velocity tracking: The actual electrical angular velocity and the estimated electrical angular velocity of this invention have a higher degree of overlap, and the steady-state fluctuation is smaller. It can maintain good tracking performance under conditions of speed variation and sudden load changes. This improves the overall estimation accuracy of electrical angular velocity and the system's disturbance rejection performance.

[0222] See Figure 5 k di Parameter variation: Adaptive k di The proposed scheme reduces the electrical angular velocity estimation error during velocity changes and load abrupt changes without increasing steady-state estimation oscillations. The root mean square error of the electrical angular velocity estimation across the entire time domain is reduced by approximately 7.24%; the root mean square error in the local interval of the second load abrupt change is reduced by approximately 5.84%. This demonstrates that the adaptive k-axis... di The method can improve the accuracy of electric angular velocity estimation and the dynamic performance of the observer during the load change process of permanent magnet synchronous motor.

[0223] See Figure 6 When the rotational speed changes, although the transient current index changes, the transient mechanical disturbance index does not increase synchronously. Therefore, the consistency coefficient remains essentially zero, and the adaptive k... di Maintain a low baseline value. During two load surges, the current transient index and the mechanical disturbance transient index rise synchronously, and the consistency coefficient increases rapidly, causing k to... di From 0.2×10 −6  s increased to approximately 1.0 × 10 −6  s, thereby enhancing the transient compensation of the rate of change of current during load disturbances; after the disturbance weakens, k di The consistency coefficient gradually decreases. The results show that the proposed method can distinguish between speed command changes and load disturbances, and adjust the transient compensation intensity online according to the degree of disturbance consistency, thereby enhancing dynamic compensation capability while avoiding steady-state noise amplification.

[0224] The foregoing description of the embodiments enables those skilled in the art to make or use the present invention. Various modifications to the embodiments will be readily apparent to those skilled in the art. The general principles of the invention may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention should not be limited to the embodiments shown herein, but should cover the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A sensorless control method for a permanent magnet synchronous motor based on coordinated regulation of current transients and system disturbance consistency, characterized in that, Includes the following steps: S1 collects and processes the three-phase stator current and DC bus voltage of the permanent magnet synchronous motor; S2 calculates the stator current based on the current sampling time and the previous sampling time. The discrete change in shaft stator current is used to generate a current transient index; the system disturbance transient index is obtained by generating the total disturbance estimate of the linear extended state observer from the previous sampling period and delayed by at least one sampling period, with the initial system disturbance transient index being 0; S3 generates a transient consistency coefficient based on the current transient index and the delayed system disturbance transient index, and adjusts the current change rate compensation coefficient online based on the transient consistency coefficient. S4 calculates the current change rate compensation amount based on the current change rate compensation coefficient and the stator current discrete change amount, and then adds the current change rate compensation amount after the limit to the current observation error to form the corrected current observation error. S5 inputs the corrected current observation error into the adaptive sliding mode observer to generate... Shaft sliding mode injection signal, reconstruction The shaft back electromotive force is obtained, and the rotor electrical angle estimate and speed estimate for the current sampling period are obtained through a phase-locked loop; The speed estimate of S6 in the current sampling period is used to estimate the total system disturbance through a linear extended state observer, generating... The shaft torque current reference value is used to generate system disturbance transient indices for use in the next sampling period based on the total disturbance estimate of the current sampling period. The S7 performs current closed-loop control based on torque and current reference values, and drives the inverter through SVPWM to achieve sensorless control of the permanent magnet synchronous motor.

2. The sensorless control method for a permanent magnet synchronous motor based on the coordinated adjustment of current transient and system disturbance consistency as described in claim 1, characterized in that, The linear extended state observer uses the rotational speed estimate obtained by the adaptive sliding mode observer as the observation input, and its state equation is: ； ； ； in, This is the error in velocity observation; It is a speed state estimate; This is the estimate of the total system disturbance; The rotational speed estimate obtained by the adaptive sliding mode observer; for Actual shaft current; To control the nominal gain value; , This represents the gain of the linearly extended state observer.

3. The sensorless control method for a permanent magnet synchronous motor based on the coordinated adjustment of current transient and system disturbance consistency according to claim 2, characterized in that, The specific process for obtaining the system disturbance transient index is as follows: First, calculate the dynamic disturbance component at the current sampling time based on the slowly varying baseline from the previous sampling time, and then update the slowly varying baseline, specifically as follows: ； ； in, The slowly varying baseline for the total disturbance estimate; For dynamic disturbance components; The baseline tracking coefficient, and ; Convert the dynamic disturbance component into an equivalent disturbance current: ; Based on the equivalent disturbance current, generate the unfiltered system disturbance transient index: ； And low-pass filtering is applied to the unfiltered system disturbance transient indices: ； in, The dynamic disturbance dead zone threshold; The full-scale threshold for dynamic disturbance; The filter coefficients are the transient indexes of system disturbances. This means that the input value is restricted to the range of 0 to 1.

4. The sensorless control method for a permanent magnet synchronous motor based on the coordinated adjustment of current transient and system disturbance consistency according to claim 1, characterized in that, The specific transient current parameters are as follows: Generate transient current indices based on discrete changes in stator current: ； in, This is the normalized reference value for the discrete variation of the stator current vector; , for Discrete variation of shaft stator current.

5. The sensorless control method for a permanent magnet synchronous motor based on the coordinated adjustment of current transient and system disturbance consistency according to claim 1, characterized in that, The transient consistency coefficient is: ； ； in, This is a delayed transient indicator of system disturbances; The unfiltered transient consistency coefficient; For current transient indicators, The unfiltered transient consistency coefficient; The transient consistency coefficient after filtering; These are the consistency filter coefficients.

6. The sensorless control method for a permanent magnet synchronous motor based on the coordinated adjustment of current transient and system disturbance consistency according to claim 5, characterized in that, The current change rate compensation coefficient is adjusted online according to the following formula: ； in, This is the lower limit of the current change rate compensation coefficient; This is the upper limit of the current change rate compensation coefficient; the... The compensation amount for the rate of change of shaft current is: ； ； in, The sampling period; This is the limit value for the current change rate compensation. and express Discrete variation of shaft stator current.

7. The sensorless control method for a permanent magnet synchronous motor based on the coordinated adjustment of current transient and system disturbance consistency according to claim 1, characterized in that, The sliding mode gain and sliding mode boundary layer width of the adaptive sliding mode observer are dynamically adjusted based on the filtered rotational speed estimate, as follows: ； ； in, This is the lower limit of the sliding mode gain; This is the upper limit of the sliding mode gain; This is the filtered estimated rotational speed. The reference speed for adjusting the sliding mode gain; This is the initial boundary layer width; This is the boundary layer dynamic adjustment coefficient; and These are the lower and upper limits of the boundary layer width, respectively.

8. The sensorless control method for a permanent magnet synchronous motor based on the coordinated adjustment of current transient and system disturbance consistency according to claim 1, characterized in that, When obtaining the rotor electrical angle estimate and speed estimate for the current sampling period through the phase-locked loop, the phase-locked loop gain scheduling factor is also set based on the filtered speed estimate: ； in, This is the filtered estimated speed value. The base speed for phase-locked loop gain scheduling; And adjust the phase-locked loop's proportional gain and integral gain online: ； ； in, and These are the lower and upper limits of the proportional gain. and These are the lower and upper limits of the integral gain. and For proportional gain and integral gain.

9. The sensorless control method for a permanent magnet synchronous motor based on the coordinated adjustment of current transient and system disturbance consistency according to claim 1, characterized in that, The The shaft torque current reference value is: ； ； ； ； in, Based on the control quantity, Given a rotational speed; For speed error feedback gain; For acceleration feedback gain; For velocity state estimation; This is the estimate of the total system disturbance; This is an estimate of the acceleration. The filtered value of the acceleration estimate; These are the filter coefficients for the acceleration estimate; for Shaft current feedforward compensation; for Shaft torque current reference value and limit value; To control the nominal gain value; This represents the actual q-axis current.

Images