A position sensorless control method for a segmented linear permanent magnet synchronous motor

Abstract

The application discloses a kind of winding segmentation linear permanent magnet synchronous motor's position sensorless control method, first according to the coupling relationship of two adjacent stator and rotor, establish double-section composite equivalent model;Then MRAS reference model and adjustable model are constructed based on composite equivalent model;Third, the LESO observer of the dq axis lumped disturbance formed by parameter drift, inverter nonlinear error and load disturbance is established;Further, the disturbance estimation value output by LESO is fed forward to the adjustable model of MRAS, and a composite equivalent model LESO-MRAS observer is formed;Finally, the speed and position estimation value output by the observer is used to realize closed-loop position sensorless control.The application can significantly weaken the influence of position-sensitive parameters on position sensorless control, improve the speed, position estimation accuracy and system disturbance robustness in the full stroke range, especially in the transition area between sections.

Description

Technical Field

[0001] This invention belongs to the field of linear permanent magnet synchronous motor control technology, specifically relating to a sensorless control method for a winding segmented linear permanent magnet synchronous motor. Background Technology

[0002] Segmented linear permanent magnet synchronous motors have advantages such as high modularity, low copper loss over long stroke, easy maintenance and fault isolation, and have good application prospects in rail transit, logistics transportation, industrial automation and special linear drive equipment.

[0003] However, the segmented winding structure causes the mover to simultaneously engage with both the exiting and entering stator segments when crossing adjacent stator segments. Key electromagnetic parameters such as the direct-axis inductance, permanent magnet flux linkage, and back EMF coefficient exhibit significant nonlinear and time-varying characteristics with position. Traditional sensorless control methods based on single-segment fixed-parameter models, especially traditional MRAS (Model Reference Adaptive System) methods, typically assume that motor parameters are essentially constant within the observation interval. When these methods are directly applied to the transition region between segmented motor segments, model mismatch easily occurs between the reference model and the actual object, leading to speed estimation drift, increased position estimation error, and decreased control stability.

[0004] Existing technologies typically employ inter-segment signal splicing, zone switching criteria, or observation methods based on composite back EMF to improve inter-segment estimation performance. Inter-segment signal splicing methods, for example, superimpose or splice the observation signals or back EMF signals of adjacent stator segments when the mover crosses segments to maintain the continuity of position and velocity estimation, as seen in [R. Leidhold. Speed ​​Sensorless Control of a Long-Stator Linear Synchronous Motor Arranged Multiple Segments. IEEE Transactions on Industrial Electronics, 2007]. Zone switching criteria methods, on the other hand, determine whether the mover is in an intra-segment or inter-segment region and switch between different observation models, feedback signals, or control laws, as seen in [Z. Lin. Research on a New Sensorless Control of HighPower Segmented Permanent Magnet Linear Synchronous Motor by The Improved Sliding Mode Observer. ICEMS, 2021]. Observation methods based on composite back EMF construct position estimates by fusing the composite back EMFs of adjacent stator segments, as seen in [J. Liu. Study on the...]. [Position Estimation Method of Winding Segmented Permanent Magnet Linear Motor. IEEE Access, 2022]. However, these methods still suffer from problems such as insufficient overall model uniformity, sensitivity to parameter drift, and insufficient robustness under load disturbance conditions in the segment and inter-segment operating ranges after I / F (current-frequency control) is started and the load exceeds the preset threshold.

[0005] Therefore, it is necessary to propose a sensorless control method that can transform the time-varying coupling problem between segmented motors into a unified modeling and observation problem under the overall motor form, and further improve the system's ability to suppress parameter disturbances and external disturbances. Summary of the Invention

[0006] In view of the above, the present invention provides a sensorless control method for a segmented linear permanent magnet synchronous motor with windings. By constructing a dual-segment composite equivalent model, the inter-segment operating state of the segmented motor is equivalent to an overall motor model with approximately constant parameters. Based on the overall motor model, a LESO-MRAS observer is introduced to achieve high-precision and robust speed and position estimation throughout the entire stroke range, especially in the inter-segment transition region.

[0007] A sensorless control method for a segmented linear permanent magnet synchronous motor with windings includes the following steps: (1) By utilizing the complementary variation law of the direct and perpendicular axis inductance and the permanent magnet flux linkage, the equivalent inductance and equivalent flux linkage of two adjacent stator segments are determined, so that the operating state between motor segments is equivalent to the overall motor model with approximately constant parameters. (2) Construct a reference model and an adjustable model of MRAS based on the overall motor model; (3) Establish LESO (Linear Extended State Observer) to monitor parameter drift, inverter nonlinear error and dq axis lumped disturbance caused by external load disturbance in real time; (4) Feedforward the lumped disturbance estimate of the dq axis output by LESO to the adjustable model, and construct an adaptive error signal based on the output error between the reference model and the compensated adjustable model; (5) Based on the adaptive error signal, the estimated values ​​of electric angular velocity and electric angular position are updated using the adaptive law of MRAS, and then the estimated values ​​of the moving part linear velocity and position are calculated and applied to the closed-loop sensorless control system of the winding segmented linear permanent magnet synchronous motor.

[0008] Furthermore, the equivalent inductance and equivalent flux linkage of two adjacent stator segments in step (1) are expressed as follows: 2× L ss =2× L σ +Δ L ψ fs = ψ f in: L ss and ψ fs These are the equivalent inductance and equivalent flux linkage of two adjacent stator segments, respectively. L σ For motor leakage inductance, Δ L This represents the inductance difference between the fully coupled and fully uncoupled states. ψ f This represents the amplitude of the permanent magnet flux linkage of the motor in a fully coupled state.

[0009] Furthermore, the expression for the overall motor model in step (1) is as follows: u dΣ = R s i dΣ + L ss d i dΣ / d t - oh e L ss i qΣ u qΣ = R s i qΣ + L ss d i qΣ / d t + oh e L ss i dΣ + oh e ψ fs In the formula: u dΣ and u qΣ These are the total d-axis voltage and the total q-axis voltage of two adjacent stator segments, respectively. i dΣ and i qΣ These are the total d-axis current and the total q-axis current of two adjacent stator segments, respectively. oh e The electric angular velocity of the motor. R s This is the stator resistance of the motor. t Indicates time.

[0010] Furthermore, the expression for the reference model in step (2) is as follows:

[0011]

[0012] in: y r (t )for t The output of the reference model should be considered at all times. i dΣ ( t )and i qΣ ( t ) are respectively t The total d-axis current and the total q-axis current of two adjacent stator segments at any given time.

[0013] Furthermore, the expression for LESO in step (3) is as follows:

[0014] in: z 1 represents the system state variable. x The estimated value of 1, z 2 is for expanding state variables x The estimated value of 2, U For the equivalent input of LESO, when x 2 represents the d-axis lumped disturbance. f d hour, x 1= i dΣ , U = u dΣ - R s x 1+ oh e L ss i qΣ ;when x 2 represents the lumped disturbance along the q-axis. f q hour, x 1= i qΣ , U = u qΣ - R s x 1- oh e L ss i dΣ - oh e ψ fs ; b 0 is the input gain and b 0 = 1 / L ss , β1 and β 2 represents the observer gain. e obs For observation error, and They are respectively z 1 and z The first derivative of 2.

[0015] Furthermore, the adjusted model expression after compensation in step (4) is as follows:

[0016]

[0017]

[0018] in: y a ( t )for t The output of the time-adjustable model i and u These are the current state variables and the voltage state variables, respectively. and They are respectively i dΣ and i qΣ The set value, and They are respectively u dΣ and u qΣ The set value, and They are respectively and The estimated value, and They are respectively t time and The estimated value, and Lumped disturbances along the d-axis f d and q-axis lumped disturbance f q The estimated value.

[0019] Furthermore, the expression for the adaptive error signal in step (4) is as follows:

[0020] in: e ( t )for t The adaptive error signal at time t, and They are respectively t time i dΣ and i qΣ The set value.

[0021] Furthermore, in step (5), the adaptive law of MRAS updates the estimated values ​​of electric angular velocity and electric angular position using the following expression, and calculates the estimated values ​​of the mover's linear velocity and position:

[0022]

[0023]

[0024]

[0025] in: k p and k i These are proportional gain and integral gain, respectively. This is an estimated value for the electric angular velocity. This is an estimate of the electric angular position. e ( t )for t The adaptive error signal at time t, This is an estimate of the linear velocity of the moving part. This is the estimated position of the mover. t This represents the polar distance.

[0026] A computer device includes a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the above-described sensorless control method for a segmented linear permanent magnet synchronous motor with windings.

[0027] A computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-described sensorless control method for a segmented linear permanent magnet synchronous motor with windings.

[0028] Based on the above technical solution, the present invention has the following beneficial technical effects: 1. This invention establishes a composite equivalent model based on the complementary relationship between parameters of two adjacent segments, transforming the inter-segment operation problem of segmented motors into a unified modeling and unified observation problem under the overall motor form, thereby reducing the impact of position-sensitive parameters on the performance of the observer.

[0029] 2. This invention does not simply employ traditional MRAS or LESO alone, but rather eliminates the structural mismatch caused by time-varying inter-segment parameters through a composite equivalent model, and then uses LESO to compensate for the remaining lumped disturbances, forming a composite observation structure with clear hierarchy and logical closure.

[0030] 3. Compared with traditional constant-parameter MRAS, the present invention has higher speed and position estimation accuracy and shorter dynamic recovery time under parameter drift and load change conditions.

[0031] 4. This invention is applicable to the intra-segment operating region and inter-segment transition region of a segmented linear permanent magnet synchronous motor with windings. It has good engineering feasibility and scalability, and is particularly suitable for fields such as rail transit, industrial automation, logistics sorting, semiconductor manufacturing and special linear drive equipment. Attached Figure Description

[0032] Figure 1 This is a schematic diagram illustrating the construction of the two-segment composite equivalent model of the present invention.

[0033] Figure 2 This is a basic principle block diagram of MRAS.

[0034] Figure 3 This is a schematic diagram of the transfer function of a second-order LESO.

[0035] Figure 4 This is a system block diagram of the LESO-MRAS observer, a composite equivalent model of the present invention.

[0036] Figure 5 This is a block diagram of a sensorless control system for a segmented linear permanent magnet synchronous motor with windings throughout its entire stroke. Detailed Implementation

[0037] To describe the present invention in more detail, the technical solution of the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0038] This embodiment provides a sensorless control method for a segmented linear permanent magnet synchronous motor with windings. The specific implementation process is as follows: (1) Collect the data of the segmented linear permanent magnet synchronous motor currently participating in coupling. k Duan Dingzi and the first k +1 stator voltage signal, current signal, and mover operating status signal; based on the coupling relationship between adjacent stator and mover segments in the inter-segment transition region, a two-segment composite equivalent model is established, such as... Figure 1 As shown, the equivalent inductance is constructed by utilizing the complementary variation law of the direct and quadrature axis inductances and the permanent magnet flux of two adjacent segments. L ss and equivalent magnetic flux y fsThis makes the operating state between segmented motors equivalent to a whole motor model with approximately constant parameters.

[0039] The first to participate in coupling k The direct and quadrature axis inductance of the stator L k ( x ) and permanent magnet flux ψ fk ( x They are represented as follows: L k ( x )= L σ + α k ( x )Δ L ψ fk ( x )= α k ( x ) ψ f in: α k ( x ) is the first k The coupling coefficient between the stator and the mover. L σ For motor leakage inductance, Δ L This represents the inductance difference between the fully coupled and fully uncoupled states. ψ f This represents the amplitude of the permanent magnet flux linkage of the motor in a fully coupled state. x Indicates the position of the mover.

[0040] When the moving part is in the first k Section and the k When there are +1 stator segments, the coupling coefficients of two adjacent stator segments satisfy: α k ( x )+ α k+1 ( x )=1 Therefore, the equivalent inductance of two adjacent stator segments L ss and equivalent magnetic flux ψ fs satisfy: 2× L ss =[ L k (x )+ L k+1 ( x )] = [ L σ + α k ( x )Δ L+L σ + α k+1 ( x )Δ L ] = 2× L σ +Δ L ψ fs = ψ fk ( x )+ ψ f(k+1) ( x ) =α k ( x ) ψ f + α k+1 ( x ) ψ f= ψ f Therefore, it can be seen that the first [participating in coupling] k Section and the k The equivalent inductance of segment +1 L ss and equivalent magnetic flux ψ fs It can be considered a constant, which is beneficial for control.

[0041] Based on the composite equivalent model of two adjacent segments, the total current component and the differential current component are defined. When the differential current component is suppressed to a sufficiently small size, the two-segment composite model and the overall motor model are approximately consistent in the inter-segment region. Thus, an MRAS reference model can be constructed on the basis of a unified model.

[0042] The current state variable consists of the total current component and the differential current component based on the dq-axis currents of two adjacent stator segments. The total current component and the differential current component of two adjacent stator segments are defined as follows: i dΣ = i dk + i d(k+1) i qΣ = i qk + i q(k+1) i dΔ = i dk - i d(k+1) i qΔ = i qk - i q(k+1) In the formula: i dΣ , i qΣ For the first k Section and the k +1 segment of stator dq axis total current, i dΔ , i qΔ For the first k Section and the k +1 segment of stator dq axis differential current i dk 、i qk For the first k The dq axis current of the stator segment i d(k+1) 、i q(k+1) For the first k +1 segment of stator dq axis current.

[0043] The total current loop is used to meet the thrust output requirements, based on the output of the velocity loop. , As a reference value for the total current loop, according to , Generate total voltage command; differential current loop with , Given a value, according to , The differential voltage compensation is output through the PI regulator, and then allocated according to the total voltage command and the differential voltage command to obtain the first... k Section and the k +1 stator voltage command, thereby making the stator current of adjacent stator segments tend to be consistent.

[0044] By controlling the stator currents of adjacent segments to remain consistent, the differential current component is suppressed to zero, ensuring the overall motor model holds true in the inter-segment region. This is achieved when the differential current component satisfies the condition through dual-segment synchronous control. i dΔ =0= i qΔ When = 0, the segmented linear permanent magnet synchronous motor with windings can be equivalent to a whole motor model in the transition region between segments, and its dq-axis equivalent voltage equation is: u dΣ = R s i dΣ + L ss d i dΣ / d t - oh e L ss i qΣ u qΣ = R s i qΣ + L ss d i qΣ / d t + oh e L ss i dΣ + oh e ψ fs In the formula: u dΣ , u qΣ For the first k Section and the k +1 segment dq Total shaft voltage, oh e Electric angular velocity, R s This is the stator resistance.

[0045] (2) Construct the reference model and adjustable model of MRAS based on the overall motor model.

[0046] like Figure 2As shown, MRAS includes a reference model, an adjustable model, and an adaptive control law. The reference model is constructed using the current equation of a composite equivalent model. The adjustable model uses the electric angular velocity to be estimated as the parameter to be estimated, and constructs an adaptive error signal through the error between the output of the reference model and the output of the adjustable model. e ( t ), and update the estimated values ​​of electric angular velocity, electric angular position, and linear velocity.

[0047] Based on the above two equations, a reference model for MRAS can be established:

[0048]

[0049] Redefining state variables i and u The adjustable models are shown below:

[0050]

[0051]

[0052] in: y r ( t () represents the output of the reference model. y a ( t ) represents the output of the adjustable model.

[0053] After considering lumped disturbances such as parameter drift, inverter error, and load disturbance, the current state equation of the composite equivalent model can be reconstructed as:

[0054] Among them, the voltage drop due to resistance change, the induced electromotive force deviation caused by inductance change, the back electromotive force error caused by flux linkage deviation, the inverter voltage error, and the external load disturbance are defined as the lumped disturbance of the dq axis, denoted as . f d and f q The corresponding estimated value is and .

[0055] (3) Establish a linear extended state observer (LESO) to monitor parameter drift, inverter nonlinear error and dq axis lumped disturbance caused by external load disturbance in real time.

[0056] To improve the robustness of traditional MRAS under parameter perturbations and external disturbances, this invention further constructs a lumped disturbance model along the d-q axis, and designs the following for the d-axis and q-axis respectively. Figure 3 The second-order LESO shown extracts the lumped disturbance consisting of parameter drift, inverter nonlinearity error and load disturbance in real time, and outputs the disturbance estimate.

[0057] Taking the q-axis as an example, the system state variable x 1= i qΣ Define extended state variables x 2= f q Its derivative is h ( t Then its second-order extended state equation is as follows:

[0058] The corresponding second-order LESO observer is:

[0059] in: z 1 represents the current x The estimated value of 1, z 2 is for the disturbance x The estimated value of 2, b 0 is the input gain and b 0 = 1 / L ss , β 1 and β 2 represents the observer gain. e obs For observation error, U This is the equivalent input for LESO.

[0060] Perform a Laplace transform on the above equation:

[0061] eliminate Z 2. Sorted out Z 1(s) and Z 2(s) about input Y (s) and U The transfer function of (s):

[0062] The characteristic equation of the observer is:

[0063] To simplify the parameter tuning process, all poles of the observer are configured at the same location in the left half-plane. s =- oh obs Among them oh obs Defined as observer bandwidth, i.e., let:

[0064] By comparing the coefficients, the gain calculation formula can be obtained:

[0065] The advantage of this configuration method is that its physical meaning is clear, requiring only adjustment. oh obs One parameter. Wherein oh obs The sampling frequency is determined based on the current sampling frequency, noise level, and current loop bandwidth. A safe range, typically higher than the current loop bandwidth but lower than the sampling angular frequency, is usually chosen. When the noise level is high, the sampling frequency can be appropriately reduced. oh obs When the disturbance changes rapidly, appropriately increase oh obs .

[0066] (4) The disturbance estimate output by LESO is used as a feedforward term to compensate the MRAS adjustable model, so that the adjustable model dynamically matches the actual system operating state, forming a composite equivalent model LESO-MRAS observer, such as Figure 4 As shown, the observer uses a composite equivalent model to eliminate the structural model mismatch caused by the time-varying parameters between segments, and then uses LESO to compensate for the remaining lumped disturbances. Therefore, it can simultaneously take into account the consistency of modeling between segments and strong disturbance robustness.

[0067] During the motor startup phase, an initial alignment and I / F open-loop startup method is adopted, maintaining a current closed loop and a speed open loop. The current power supply stator segment is determined based on the known initial segment number, mechanical limit, or initial alignment result. When the mover speed reaches the preset switching threshold and the observer meets the convergence criterion, the startup control mode switches to a sensorless closed-loop control mode based on the composite LESO-MRAS observer. The preset switching threshold is determined based on the back EMF amplitude, current sampling signal-to-noise ratio, and observer convergence; in this embodiment, the preset switching threshold is 0.1 m / s.

[0068] An adaptive error signal is constructed based on the output error between the reference model and the compensated adjustable model. The adaptive law of the MRAS is then used to update the estimated electric angular velocity and electric angular position. The adaptive error signal of the MRAS is constructed using the cross-product of the output current of the reference model and the output current of the compensated adjustable model. This facilitates the design of a one-dimensional adaptive law, which can be used as the adjustment signal for the MRAS observer to indicate the current electric angular velocity estimation status, as detailed below:

[0069] in: e ( t The error signal reflects the phase difference between the reference current vector and the estimated current vector in the dq plane.

[0070] (5) Obtain the estimated linear velocity and position of the mover based on the updated electrical angular velocity and electrical angular position estimates, and use the estimates for closed-loop sensorless control of the winding segmented linear permanent magnet synchronous motor.

[0071] The adaptive law of MRAS is based on the error signal. e ( t Real-time adjustment of the estimated electric angular velocity value This gradually reduces the error, thereby achieving convergence of the electric angular velocity estimation, as detailed below:

[0072]

[0073]

[0074]

[0075] in: k p and k i These are proportional gain and integral gain, respectively. This is an estimate of the electric angular position. This is an estimate of the linear velocity of the moving part. This is the estimated position of the mover. t This represents the polar distance.

[0076] With the first k Duan Dingzi and the first k Based on the mechanical boundary position of the +1 segment stator, set the switching hysteresis window Δ x h In this embodiment, Δ is taken. x h =20mm, then there is a 20mm hysteresis protection zone between the adjacent section entering the standby power supply state and the previous section exiting the power supply state, as long as the position estimation error is 20mm. e x If the following expression is satisfied, the estimated position will not cross two adjacent switching thresholds, thus preventing erroneous segment switching.

[0077]

[0078] Considering control cycle and switching delay T dIf the following formula is further satisfied, false switching due to delay can be avoided.

[0079]

[0080] In this embodiment, the positionless closed-loop switching speed is taken as 0.1 m / s, even according to... T d =2ms, the position offset caused by the delay is only 0.2mm, which is far less than the 10mm switching margin, so it will not cause incorrect segment switching.

[0081] Furthermore, to avoid abrupt changes in the dq coordinate orientation during position-free closed-loop entry, an electrical angle error constraint is set before the observer enters the closed-loop. Due to the polar distance of the prototype in this embodiment... t =12mm, the electrical angle and the displacement of the moving part satisfy... i e =π x / t When cutting in, the initial electrical angle error must meet the following requirements:

[0082] That is, a 30° electrical angle corresponds to the following position error constraint:

[0083] Therefore, in this invention, the initial segment number and initial angle reference are provided by mechanical limiting, initial alignment, and I / F open-loop process during the startup phase. Once the velocity reaches a preset threshold and the observer meets the aforementioned convergence conditions, the composite LESO-MRAS position-free closed loop is entered, and the estimated position or estimated velocity participates in subsequent inter-segment switching. The aforementioned timing relationships and error constraints can avoid the circular dependency problem caused by generating switching signals from estimated velocity.

[0084] The full-stroke sensorless control system for a segmented linear permanent magnet synchronous motor mainly includes a speed outer loop, a dual-segment current synchronous controller, two sets of SVPWM (space vector pulse width modulation) and inverter drive modules, an odd-even segment switching switch, a current sampling and coordinate transformation unit, and a composite equivalent model LESO-MRAS sensorless controller, such as... Figure 5 As shown, given speed reference value With estimated speed Together, they form the outer speed loop input, from which the speed controller generates the current command required for the inner loop; the dual-stage current synchronous controller further outputs two sets of voltage commands in the dq coordinate system. and After coordinate transformation, the following is obtained: and And through SVPWM1 and SVPWM2 respectively, switching pulses T1 and T2 are generated to drive inverter 1 and inverter 2. The mover is in the first position.k Within a segment, it is mainly composed of the first k Stator power supply; mover enters the first stage k Section and the k When switching to segment +1, the first k Section and the k +1 segment is powered simultaneously, and the composite equivalent model and differential current synchronous control are activated; after the mover leaves the switching area, the first k Section 1 disconnects power supply, the first k +1 segment enters single-segment control. This control system has only two sets of inverters, controlling the odd-numbered segment and the even-numbered segment respectively. It employs a switch logic based on the position of the mover to determine the odd / even segment grouping power supply. The outputs of the two inverters are connected to the corresponding primary stator segments via odd-numbered and even-numbered segment switching switches. Current sampling 1 and current sampling 2 respectively collect the three-phase currents of the two stator segments. i abc1 and i abc2 Then, by successively transforming, the current in the stationary coordinate system is obtained. i αβ1 , i αβ2 and the current in the rotating coordinate system i dq1 , i dq2 The composite equivalent model of the LESO-MRAS sensorless controller receives voltage. u dq1 , u dq2 and current i dq1 , i dq2 Output estimated speed and estimated location ,in Used to generate position switching signals and drive the on / off switching of each stator segment. Feedback is sent to the outer speed loop, thereby achieving full-stroke closed-loop sensorless control.

[0085] Under constant load conditions, this invention can achieve smooth estimation of mover velocity and position in intra- and inter-segment regions. Under parameter drift and external load abrupt changes, traditional MRAS exhibits significant estimation bias, while the composite equivalent model LESO-MRAS observer of this invention can still maintain a small steady-state error and a fast dynamic recovery speed.

[0086] To objectively verify the improvement of this invention compared to existing technologies, we conducted comparative experiments on a prototype linear permanent magnet synchronous motor with the same winding and the same control platform, using the traditional MRAS method and the composite equivalent model LESO-MRAS method of this invention. The relevant performance comparisons are shown in Tables 1 and 2. Under no-load cross-segment conditions, the mover runs stably at 0.4 m / s and crosses adjacent stator segments; under load disturbance conditions, the mover runs at a constant speed of 0.5 m / s and is suddenly subjected to a 20 N resistance load.

[0087] Table 1: Performance Comparison of the Invention and Traditional MRAS under Unloaded Segment Operation Conditions

[0088] Table 2: Performance Comparison of the Invention and Traditional MRAS under Sudden Load Segmentation Conditions

[0089] In summary, this invention transforms the complex time-varying problem of segmented motor operation between segments into a unified modeling and unified observation problem under the overall motor form by using a "composite equivalent model + LESO-MRAS observer", thereby improving the overall performance of sensorless control.

[0090] The above description of the embodiments is provided to enable those skilled in the art to understand and apply the present invention. Those skilled in the art can readily make various modifications to the above embodiments and apply the general principles described herein to other embodiments without creative effort. Therefore, the present invention is not limited to the above embodiments, and any improvements and modifications made to the present invention by those skilled in the art based on the disclosure thereof should be within the scope of protection of the present invention.

Claims

1. A sensorless control method for a segmented linear permanent magnet synchronous motor with windings, characterized in that, Includes the following steps: (1) By utilizing the complementary variation law of the direct and perpendicular axis inductance and the permanent magnet flux linkage, the equivalent inductance and equivalent flux linkage of two adjacent stator segments are determined, so that the operating state between motor segments is equivalent to the overall motor model with approximately constant parameters. (2) Construct a reference model and an adjustable model of MRAS based on the overall motor model; (3) Establish LESO to monitor the dq axis lumped disturbances caused by parameter drift, inverter nonlinearity error and external load disturbances in real time; (4) Feedforward the lumped disturbance estimate of the dq axis output by LESO to the adjustable model, and construct an adaptive error signal based on the output error between the reference model and the compensated adjustable model; (5) Based on the adaptive error signal, the estimated values ​​of electric angular velocity and electric angular position are updated using the adaptive law of MRAS, and then the estimated values ​​of the moving part linear velocity and position are calculated and applied to the closed-loop sensorless control system of the winding segmented linear permanent magnet synchronous motor.

2. The sensorless control method for a segmented linear permanent magnet synchronous motor with windings according to claim 1, characterized in that, The equivalent inductance and equivalent flux linkage of two adjacent stator segments in step (1) are expressed as follows: 2× L ss =2× L σ +D L ψ fs = ψ f in: L ss and ψ fs These are the equivalent inductance and equivalent flux linkage of two adjacent stator segments, respectively. L σ For motor leakage inductance, Δ L This represents the inductance difference between the fully coupled and fully uncoupled states. ψ f This represents the amplitude of the permanent magnet flux linkage of the motor in a fully coupled state.

3. The sensorless control method for a segmented linear permanent magnet synchronous motor with windings according to claim 2, characterized in that, The expression for the overall motor model in step (1) is as follows: u dΣ = R s i dΣ + L ss d i dΣ / d t - ω e L ss i qΣ u qΣ = R s i qΣ + L ss d i qΣ / d t + ω e L ss i dΣ + ω e ψ fs In the formula: u dΣ and u qΣ These are the total d-axis voltage and the total q-axis voltage of two adjacent stator segments, respectively. i dΣ and i qΣ These are the total d-axis current and the total q-axis current of two adjacent stator segments, respectively. ω e The electric angular velocity of the motor. R s This is the stator resistance of the motor. t Indicates time.

4. The sensorless control method for a segmented linear permanent magnet synchronous motor with windings according to claim 3, characterized in that, The expression for the reference model in step (2) is as follows: in: y r ( t )for t The output of the reference model should be considered at all times. i dΣ ( t )and i qΣ ( t ) are respectively t The total d-axis current and the total q-axis current of two adjacent stator segments at any given time.

5. The sensorless control method for a segmented linear permanent magnet synchronous motor with windings according to claim 3, characterized in that, The expression for LESO in step (3) is as follows: in: z 1 represents the system state variable. x The estimated value of 1, z 2 is for expanding state variables x The estimated value of 2, U For the equivalent input of LESO, when x 2 represents the d-axis lumped disturbance. f d hour, x 1= i dΣ , U = u dΣ - R s x 1+ ω e L ss i qΣ ;when x 2 represents the lumped disturbance along the q-axis. f q hour, x 1= i qΣ , U = u qΣ - R s x 1- ω e L ss i dΣ - ω e ψ fs ; b 0 is the input gain and b 0 = 1 / L ss , β 1 and β 2 represents the observer gain. e obs For observation error, and They are respectively z 1 and z The first derivative of 2.

6. The sensorless control method for a segmented linear permanent magnet synchronous motor with windings according to claim 3, characterized in that, The expression for the compensated adjustable model in step (4) is as follows: in: y a ( t )for t The output of the time-adjustable model i and u These are the current state variables and the voltage state variables, respectively. and They are respectively i dΣ and i qΣ The set value, and They are respectively u dΣ and u qΣ The set value, and They are respectively and The estimated value, and They are respectively t time and The estimated value, and Lumped disturbances along the d-axis f d and q-axis lumped disturbance f q The estimated value.

7. The sensorless control method for a segmented linear permanent magnet synchronous motor with windings according to claim 6, characterized in that, The expression for the adaptive error signal in step (4) is as follows: in: ε ( t )for t The adaptive error signal at time t, and They are respectively t time i dΣ and i qΣ The set value.

8. The sensorless control method for a segmented linear permanent magnet synchronous motor with windings according to claim 7, characterized in that, In step (5), the adaptive law of MRAS updates the estimated values ​​of electric angular velocity and electric angular position using the following expression, and calculates the estimated values ​​of the mover's linear velocity and position: in: k p and k i These are proportional gain and integral gain, respectively. This is an estimated value for the electric angular velocity. This is an estimate of the electric angular position. ε ( τ )for τ The adaptive error signal at time t, This is an estimate of the linear velocity of the moving part. This is the estimated position of the mover. τ This represents the polar distance.

Images