Synchronous reluctance motor FC-MTPA control method based on current flux linkage model

Abstract

The invention relates to the field of motor driving control, and provides a control method and system for maximizing a torque current ratio of a synchronous reluctance motor (SynRM) under the constraint of a flux linkage circle, which can be called as flux linkage limited MTPA (Flux-Constrained MTPA, FC-MTPA), and is an efficiency optimal speed raising control strategy between the MTPA and the MTPV. Compared with an MTPA control method, the method has a larger speed increasing space, and compared with a maximum torque voltage ratio (MTPV) control method, efficiency optimization is considered. The main content of the invention is as follows: the method and the system for solving an FC-MTPA point on line by using a current flux linkage model and for control, the FC-MTPA method has a larger acceleration space than an MTPA method under the same voltage amplitude limit and considers the reduction of loss, and the online solving method does not need a large amount of storage data and is controllable in calculated amount. And the method can also be applied to a low-performance microcontroller (MCU).

Description

Technical Field

[0001] This invention belongs to the technical field of motor drive control, and more specifically, relates to a control method for a synchronous reluctance motor FC-MTPA based on a current flux linkage model. Background Technology

[0002] Synchronous reluctance motors have attracted widespread attention due to their high efficiency, low cost, and lack of permanent magnet dependence. To reduce losses and improve efficiency, MTPA control is widely used, while MTPV control is widely used to continue increasing speed after the voltage reaches its limit circle. For voltage-limited motors, flux-limited MTPA (FC-MTPA) is an effective method to achieve higher speeds while minimizing losses. For synchronous reluctance motors, at higher speeds, U... d ≈-ω e ψ q U q ≈ω e ψ d , ψ d 2 + ψ q 2 dc 2 / 3ω e 2 The flux linkage circle approximates the voltage circle; voltage limitation is equivalent to flux linkage amplitude limitation. Therefore, in optimizing the flux linkage circle, T... e / I s Optimizing T under the voltage limit circle e / I s They are consistent ​For the significant inductive nonlinearity of synchronous reluctance motors (SynRM), approximating the control model with only a fixed inductance results in large deviations. For control methods such as MTPA and MTPV, several common engineering implementation paths each have limitations: First, offline calibration lookup tables (LUTs) have low computational cost during runtime, but high initial mapping and maintenance costs, requiring resources to store lookup table data. Furthermore, when system parameters change irreversibly, the lookup table (LUT) needs to be manually recalculated; expanding the table size to adapt to changes in operating conditions increases storage requirements and workload. Second, online identification of electromagnetic parameters, based on analytical methods for MTPA and MTPV control, often ignores or simplifies electromagnetic phenomena such as saturation and mutual inductance, resulting in insufficient accuracy. Additionally, real-time inductance identification involves high computational costs and the risk of divergence. Third, for MTPA control, there are online small perturbation methods that inject small perturbations to search for extreme values ​​of the current angle. While this reduces model dependence, it introduces additional noise and ripple, and is difficult to implement on low-computing-power platforms. In summary, achieving a current distribution method on a low-cost MCU that can demonstrate nonlinear accuracy without increasing computing power and storage burden remains a key challenge for industrial application. This patent aims to balance speed-up potential and losses under voltage-constrained conditions, and designs a control strategy and system that reduces offline dependence while improving optimality and stability under controllable computing power. Summary of the Invention

[0003] To address the aforementioned deficiencies or improvement needs of existing technologies, this invention provides a synchronous reluctance motor FC-MTPA control method based on a current flux linkage model. The aim is to find the maximum torque-current ratio point (Flux-Constrained MTPA, FC-MTPA) on the flux linkage circle, thereby solving the technical problems of limited speed-up space in MTPA control and excessively high MTPV control losses under voltage constraints.

[0004] To achieve the above objectives, this invention provides a control method for a synchronous reluctance motor (FC-MTPA) based on a current flux linkage model, comprising the following steps: Obtain the d-axis and q-axis current and flux linkage data of the motor, and establish a current-flux linkage algebraic model containing self-axis saturation and cross-coupling terms; Establish a conditional MTPA solver: In the current-flux flux algebraic model, the stator flux amplitude ψ s As a constraint, the flux linkage angle γ is kept constant, and the torque-to-current ratio J(γ) = T is used as the optimization variable. e (γ) / I s (γ) is the objective function. The optimal flux linkage angle is obtained by performing the golden section method within the flux linkage circle constraint, where T... e (γ) is the electromagnetic torque, I s(γ) represents the current amplitude; the corresponding motor d-axis and q-axis currents I are calculated based on the optimal flux linkage angle. d I q With magnetic flux ψ s , to obtain (ψ s I d I q FC-MTPA point; Three sets of constant stator flux linkage amplitudes are selected and the conditional MTPA solver is called respectively to obtain three sets of FC-MTPA points, which are recorded as fixed table points in the FC-MTPA point lookup table. Select two sets of active stator flux linkage amplitudes and call the conditional MTPA solver respectively to obtain two sets of FC-MTPA points. Record these as active table points in the FC-MTPA point lookup table, and then use I... q The instruction value locates adjacent table points; In the speed loop interruption, according to real-time I q I is generated by linear interpolation between adjacent table points. dref The reference value is then filtered by a first-order low-pass filter and output to the current loop for control.

[0005] Furthermore, the stator flux linkage amplitude ψ s Maintaining the constant value as a constraint means that at any given moment during operation, the stator flux linkage amplitude ψ is kept constant. s >0 is considered an outer-layer given or constraint quantity, determined by the upper-layer strategy. To facilitate univariate optimization, this method considers each given ψ... s The solution is optimally independent and not coupled with other working points.

[0006] Furthermore, parameterizing the flux linkage angle and expressing the torque refers to using the flux linkage angle as the parameter. Decompose the magnetic flux linkage along the dq axis: ψ d =ψ s ·cosγ,ψ q =ψ s ·sinγ; according to ψ d and ψ q Based on the current-magnetic flux algebraic model I d (ψ d , ψ q ), I q (ψ d , ψ q (Obtained from tuning) Calculate the corresponding right-angle current I. d and I q Electromagnetic torque Current amplitude P represents the number of pole pairs of the motor.

[0007] Furthermore, with the torque-to-current ratio J(γ)=T e (γ) / Is The objective function is constructed using the torque-to-current ratio J(γ) = T. e (γ) / I s (γ) is the optimization objective (which can also be extended to an equivalent index of maximizing efficiency or minimizing loss). J depends only on γ and ψ. s This ensures that the optimization variable is a one-dimensional scalar, which facilitates stable implementation on low-computing-power platforms.

[0008] Furthermore, the golden section method is applied to the closed interval [0, π / 2], using the golden angle... Segmentation, determining internal sampling points a and b, and monotonically shrinking the interval by comparing J(a) and J(b); when the interval width Or the number of iterations reaches K max At that time, output γ according to the "accuracy standard achieved by taking the midpoint" criterion. m =(a+b) / 2 is used as the optimal flux linkage angle estimate in this state; to improve the neighborhood accuracy, one or two more derivative refinements can be performed to calculate dJ / dγ and perform a first-order update with a limited step size, and then γ is projected back to [a, b] to prevent out-of-bounds errors.

[0009] Based on the optimal flux linkage angle Calculate the corresponding right-angle current I d and I q , to obtain (ψ s I d I q The point at which the output flux linkage circle (voltage circle) is located is the point where ψ is obtained at the final angle γ. d =ψ s ·cosγ,ψ q =ψ s sinγ and I d I q And obtain (ψ) s I d I q The points; the above results can be directly used as entries for online interpolation, or generated instantly. dref Used for current loop control.

[0010] Furthermore, during the power-on / reset phase, the parameters of the current-flux flux algebraic model are identified and solidified through injection experiments or existing data. Preferably, bipolar voltage pulses are applied sequentially to the d-axis, q-axis, and simultaneously to both axes while the rotor is stationary, and the flux flux (ψ) is obtained by integrating the voltage after resistance compensation. d , ψ q Collect scattered data (I) d I q , ψ d , ψ qThe "current-magnetic flux linkage" model is solved by writing it in linear least squares (or recursive least squares) form, with coefficients containing self-axis saturation and cross-coupling terms (such as ψ). d m ψ q n ψ d h ψ q j+2 (etc.); after the parameters converge, they are stored in non-volatile memory for rapid evaluation and calculation during runtime.

[0011] Furthermore, three sets of flux linkage amplitudes ψ were selected. min ψ mid ψ max (e.g., 1% / 50% / 100% of the rated flux), in each ψ s The conditional MTPA solver under the flux linkage circle (voltage circle) in the first part is called to obtain three sets (I d I q , ψ s And write it to "fixed entries"; fixed entries only participate in sorting and are fixed and will not be deleted during the operation.

[0012] Furthermore, in addition to the three fixed points, two more "active points" are maintained for local refinement and approximation of the optimal value; the real-time measured I... qref (After light filtering) This is used to locate adjacent left and right points within the table, forming a squeeze interval [L, R]. Where I qref The current command value for the q-axis output by the speed loop is L, and R is the two FC-MTPA points.

[0013] Furthermore, after determining the squeeze interval [L, R], the amplitude of the outer loop flux linkage is set to ψ. s_cmd =0.5 (ψ sL +ψ sR ), which serves as the amplitude for this round of FC-MTPA solution; this setting constrains the optimization point to the vicinity of the current working region, improving convergence stability and reducing sensitivity to noise; when the accuracy requirements are met, ψ s_cmd Keep it constant to avoid jitter. The accuracy requirements referred to include |I q_R - I q_L |or|ψ sR - ψ sL | When it is below the threshold, and I used as control qref With I used for interpolation q_R I q_L Near enough. Among them, I q_L I q_R Let ψ be the q-axis current at points L and R, which are FC-MTPA points. sL ψ sRLet L and R be the flux linkage amplitudes at the two FC-MTPA points.

[0014] Furthermore, press I during the speed loop interruption. qref to I dref Perform linear interpolation: I dref =I d_L + (I qref - I q_L ) / (I q_R - I q_L )·(I d_R - I d_L The output is then filtered using a first-order low-pass filter and sent to the current loop; the interrupt side only performs table reading and interpolation to ensure real-time performance. Among these, I... dref The d-axis current command value is obtained by interpolation using the FC-MTPA point lookup table, I d_L I d_R Let L and R be the d-axis currents at the two FC-MTPA points.

[0015] Furthermore, in the main loop / low-frequency thread, when the stability condition (such as |ΔI) is met... q When |<δ is repeated N times consecutively and cumulatively reaches M times, the magnetic flux domain golden section solver is invoked: with ψ fixed s_cmd Maximize J(γ) = T with γ as the sole variable. e / I s After iterating until |b-a|<ε, take the midpoint as the solution, or choose to perform two more derivative refinements to obtain γ; calculate (ψ s I d I q ) and write it to the "build table"; if the "build table" is full, then discard the old activity point with the weakest correlation / farthest distance to the current squeeze interval; then press I uniformly q Sort in ascending order, keeping a fixed point in the sorting process but not deleting any elements. Where ΔI... q The q-axis current refers to the change during a continuous cycle. δ, N, and M are manually given indicators. The "build table" is only used for updating and sorting and does not need to be controlled, so as to avoid reading the table while calculating and inserting new points.

[0016] Furthermore, the "active table" is a small control table (capacity >= 5: fixed point + two active points), while the "build table" is a background candidate database for new points. Whenever a new point is written to the "build table," the manager copies the information from the "build table" into the "active table," ensuring that the two active points can sandwich the current I. q Furthermore, writing to and reading from the table will not conflict; the control loop always generates I only from the "active table" through interpolation. dref This decouples optimization calculations from real-time interpolation.

[0017] This invention completes the parameter tuning of the current flux linkage model within the MCU during the power-on phase. During the operation phase, under the condition of fixed stator flux linkage amplitude, the flux linkage angle is used as the optimization variable, and the maximum T on the flux linkage circle is achieved by combining lightweight search and interpolation table. e / I s FC-MTPA control is implemented, balancing accuracy, computational complexity, and reliability. The nonlinearity and cross-saturation characteristics of SynRM have been confirmed in published studies. The rational use of the current flux linkage model is crucial for efficient control; therefore, constructing a flux linkage model and obtaining the dq current distribution trajectory based on it is a reasonable approach.

[0018] The present invention also provides an electronic device, comprising: a computer-readable storage medium and a processor; The computer-readable storage medium is used to store executable instructions; The processor is used to read executable instructions stored in the computer-readable storage medium and execute the above-described method.

[0019] The present invention also provides a computer-readable storage medium storing computer instructions for causing a processor to perform the above-described method.

[0020] The present invention also provides a computer program product, including a computer program or instructions that, when executed by a processor, implement the above-described method.

[0021] In summary, compared with the prior art, the FC-MTPA control method provided by this invention has the following advantages: 1. Compared with offline table lookup, this invention does not require a large amount of resources to store the information of the lookup table; this method explicitly considers cross saturation and coupling in the flux linkage domain, and the current flux linkage model implicitly contains mutual inductance and saturation characteristics, which is close to the actual state of the motor; this method does not require a large amount of online calculation, nor does it require online estimation of the dq axis inductance. It obtains the MTPA point (FC-MTPA) under the flux linkage circle constraint by evaluating with finite function, with small jitter and controllable computational load, which is suitable for low computing power MCUs.

[0022] 2. Compared to the traditional MTPA method, which seeks the minimum current under fixed torque, this method, which maximizes the torque-to-current ratio under the flux linkage circle condition, has a larger speed-up potential under the same voltage limit. Compared to the method of maximizing torque under the voltage circle condition (MTPV), this method maximizes the torque-to-current ratio under the flux linkage circle condition, taking efficiency optimization into account. It is a method between MTPA and MTPV, balancing speed-up range and loss reduction. Attached Figure Description

[0023] Figure 1This is a flowchart of the synchronous reluctance motor FC-MTPA control method based on the current flux linkage model provided by the present invention; Figure 2 It is in ψ d ψ q In coordinate system, U d U q With ψ d ψ q Mapping relationship diagram; Figure 3 This is a schematic diagram of the single peak value of the torque-current ratio at the flux linkage angle γ. Figure 4 This figure shows the experimental results of the conventional MTPA method and the present invention under a 90 V voltage limit in this embodiment. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention. Furthermore, the technical features involved in the various embodiments of this invention described below can be combined with each other as long as they do not conflict with each other.

[0025] The flowchart of the entire control system is as follows: Figure 1 As shown, the system is logically divided into two subsystems: a "model fitting system" and an "online control system," coupled through a unified parameter interface. The former is responsible for self-tuning the flux linkage domain motor model during power-on or reset; the latter, during operation, uses low computational power to calculate the FC-MTPA point using the golden section of the flux linkage domain, interpolate to generate the current reference, and coordinate with the FOC current loop and speed loop. Through this decoupled design of "fitting first, then optimizing online," accuracy, stability, and real-time performance can be balanced on low-to-medium computational power MCUs such as the GD32.

[0026] This invention provides a control method for a synchronous reluctance motor (FC-MTPA) based on a current flux linkage model, comprising the following steps: Obtain the d-axis and q-axis current and flux linkage data of the motor, and establish a current-flux linkage algebraic model containing self-axis saturation and cross-coupling terms; Establish a conditional MTPA solver: In the current-flux flux algebraic model, the stator flux amplitude ψ s As a constraint, the flux linkage angle γ is kept constant, and the torque-to-current ratio J(γ) = T is used as the optimization variable. e (γ) / I s (γ) is the objective function. The optimal flux linkage angle is obtained by performing the golden section method within the flux linkage circle constraint, where T... e (γ) is the electromagnetic torque, I s(γ) represents the current amplitude; the corresponding motor d-axis and q-axis currents I are calculated based on the optimal flux linkage angle. d I q With magnetic flux ψ s , to obtain (ψ s I d I q FC-MTPA point; Three sets of constant stator flux linkage amplitudes are selected and the conditional MTPA solver is called respectively to obtain three sets of FC-MTPA points, which are recorded as fixed table points in the FC-MTPA point lookup table. Select two sets of active stator flux linkage amplitudes and call the conditional MTPA solver respectively to obtain two sets of FC-MTPA points. Record these as active table points in the FC-MTPA point lookup table, and then use I... q The instruction value locates adjacent table points; In the speed loop interruption, according to real-time I q I is generated by linear interpolation between adjacent table points. dref The reference value is then filtered by a first-order low-pass filter and output to the current loop for control. Example 1 Specifically, during the power-on / reset phase, including (1) Input and operating condition organization: The inputs include three-phase current sampling, bus voltage, position signal / zero-speed lockout status, and upper-level limiting and pulse parameters. The fitting phase is implemented in a stationary or low-speed state. The controller applies symmetrical bipolar voltage pulses in the d-axis and q-axis directions according to a predefined script to ensure that the flux linkage trajectory is uniformly covered within a considerable range.

[0027] (2) Flux Estimation and Preprocessing: After subtracting the stator resistance voltage drop from the phase voltage and transforming it to dq coordinates, discrete integration is performed using the stator voltage model to obtain (ψ). d , ψ q To suppress drift and noise, zero-bias correction and mild IIR filtering are introduced; and robust removal is performed for obvious outliers (such as lost samples).

[0028] (3) Current-magnetic flux algebraic model: use: I d (ψ d , ψ q )=A_d(ψ d , ψ q )·ψ d ; I q (ψ d , ψ q )=A_q(ψ d , ψ q )·ψq ; The structure, in which the coefficient functions A_d and A_q are represented by a family of power and polynomial functions containing self-axis saturation and cross-coupling, allows nonlinearity and coupling to be captured in the flux linkage domain without explicit calculation of mutual inductance.

[0029] For example: A_d = a_d + b_d |ψ d | m + (c / (j+2))|ψ d | h |ψ q | j+2 ; A_q = a_q + b_q |ψ q | n + (c / (h+2))|ψ q | j |ψ d | h+2 ; m, n, h, j are nonnegative constants or fixed a priori; a_d, b_d, a_q, b_q, c are parameters to be tuned. This structure can simultaneously reflect inductor nonlinearity and cross-saturation effect; the accuracy of this model has been proven in published studies.

[0030] (4) Parameter solution: The scattered points are regressed into a system of linear / quasi-linear equations, and the undetermined parameters are estimated using least squares or recursive least squares (RLS).

[0031] (5) Quality control and physical consistency inspection: After the fitting is completed, check the following items one by one: ① Monotonic / bounded within the feasible region; ② The correlation between the model and the fitted data meets the requirements; ③ The partial derivative values ​​are stable; if not, automatically revert to a lower-order model or narrow the feasible region and refit.

[0032] (6) Parameter freezing and interface output: Qualified parameters are written to non-volatile storage and exported through a unified interface: the coefficient sets of A_d and A_q, and the effective ψ. s Range and fitting error indicators; the runtime only reads data and does not update online to ensure repeatability.

[0033] Specifically, the operational phase includes: (1) Stratification and time scale: The current loop operates at 5–20 kHz during sampling interruptions, and the speed loop operates at 0.5–2 kHz; the FC-MTPA solver and table management operate in a low-frequency schedule of 50–200 Hz and are triggered only when conditions such as “current stability” are met, thus ensuring a real-time safety boundary.

[0034] (2) FC-MTPA table and construction table: To reduce the frequency of online solutions, maintain "at least three fixed points (ψ)". min ψ mid ψ max The system consists of a small table (capacity >= 5) with two active points and a larger build table. Fixed points are sorted but not deleted, serving as global anchors and end-area protection; active points are locally encrypted as the working point changes, improving interpolation accuracy and convergence speed.

[0035] (3) Golden section solver for flux linkage domain: Figure 2 Shown in ψ d ψ q In coordinate system, U d U q With ψ d ψ q The mapping relationship diagram shows that the flux linkage circle can be approximated as the voltage circle. Therefore, optimization on the flux linkage circle is consistent with optimization on the voltage limit circle.

[0036] like Figure 3 As shown, given the flux linkage magnitude, the objective function J(γ)=T e / I exhibits a single-peak characteristic as the flux linkage angle γ changes, and the peak point can be determined using mathematical methods such as the golden section. Given ψ s Let γ∈[0,π / 2] be the unique optimization variable. From ψ d = ψ s cosγ、ψ q = ψ s sinγ, calculated by the model I d (γ), I q (γ), then obtain T e (γ)=(3p / 2)(ψ d I q - ψ q I d ) and I s (γ) = , with J(γ)=T e / I s The goal is to maximize the value. Iterative contraction is performed on the [a, b] = [0, π / 2] using the golden ratio, and "voltage limiting, current limiting, and ψ" are implemented when the boundary is exceeded. s The feasible region projection of the "upper limit". When |b - a| < ε or the number of iterations reaches K. max Stop immediately, and if necessary, perform 1-2 additional derivative refinements.

[0037] (4) Interval squeeze and outer ring ψ sGiven: According to the measured I q Determine the squeeze interval [L, R] in the small table, and the outer ring ψ s_cmd Take both ends ψ s The median value, and set the debouncing logic: when |I q_R - I q_L |or|ψ sR - ψ sL | Freeze when below the threshold ψ s_cmd To avoid back-and-forth jitter, the golden ratio is only performed within a stable window to reduce current ripple.

[0038] (5) Speed ​​loop interruption side interpolation output: The velocity loop interrupt does not participate in optimization; it only performs linear interpolation: I dref =I d_L +(I q - I q_L ) / (I q_R - I q_L )·(I d_R - I d_L ), then a first-order low-pass filter, which will pass I dref with I qref Inputting FOC current loop ensures hard real-time performance and determinism.

[0039] (6) Table maintenance and eviction strategies: The optimized thread generates a new point (I) after each golden section. d I q , ψ s Write to the build table; when the build table is full, old points are eliminated according to the criterion of "weakest correlation with the current work area and farthest distance"; then the build table is mapped and synchronized into a small table with fixed capacity, fixed points are only sorted and not deleted, and active points cover local intervals.

[0040] (7) Abnormal rollback and re-entry: If a sensor malfunction is detected, the golden section and table writing processes are paused, and only the interpolation output is retained; if the fault is cleared or the data stabilizes, the process re-enters: reselects the squeeze interval and resets ψ. s And restore optimization. If necessary, revert to a conservative strategy (such as I). d =I q (Or a fixed weak magnetic angle) to ensure safety and continuous torque.

[0041] (1) Golden section convergence threshold: ε is 5×10 -3 Up to 1×10 -2 rad, K max Take 8 to 10; (2) I dref Interpolation filtering: Idref A first-order low-pass filter is used, with filter coefficients α ranging from 0.1 to 0.3; (3) Stability check: |ΔI q If |<δ is determined to be stable after N consecutive tests, δ≈0.1 A, N≈10; (4) Table capacity: The small table should have at least 5 points (three fixed and two active). The table capacity can be increased appropriately according to the RAM resource configuration. On a 5.5 kW synchronous reluctance motor platform, sampling was performed at 5 kHz, the speed loop at 0.5 kHz, and the current loop at 5 kHz. After power-on, sample acquisition and fitting were completed, followed by initialization of three fixed points, where ψ was selected. min =0.01 Wb、ψ mid =0.5 Wb、ψ max =1 Wb. Figure 4 The experimental results show that, under the same bus voltage limit (90V), both methods accelerate to the voltage limit and trigger software protection. The left figure shows the traditional MTPA method, where the software protection is triggered after the speed reaches 665 rpm. The right figure shows the FC-MTPA method obtained from the flux linkage domain, where the software protection is triggered after the speed reaches 968 rpm. It can be seen that this method has a higher speed-up margin than the traditional MTPA method.

[0042] Example 2 The present invention also relates to an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the method described above.

[0043] The electronic device can be a desktop computer, laptop, handheld computer, or cloud server, etc. The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The memory can be used to store computer programs and / or modules. The processor performs various functions of the electronic device by running or executing the computer programs and / or modules stored in the memory, and by accessing data stored in the memory.

[0044] Example 3 The present invention also relates to a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the method described above.

[0045] Specifically, the memory may include high-speed random access memory, as well as non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital (SD) cards, flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0046] Example 4 This invention provides a computer program product or computer program that includes computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the steps of the method described in the above embodiments of this invention.

[0047] The technical features of the embodiments described above can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification. It should be noted that the terms "in one embodiment," "for example," and "again" in this invention are intended to illustrate the invention and are not intended to limit the invention.

[0048] The embodiments described above are merely examples of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. A control method for a synchronous reluctance motor (FC-MTPA) based on a current flux linkage model, characterized in that, Includes the following steps: Obtain the d-axis and q-axis current and flux linkage data of the motor, and establish a current-flux linkage algebraic model containing self-axis saturation and cross-coupling terms; Establish a conditional MTPA solver: In the current-flux flux algebraic model, the stator flux amplitude ψ s As a constraint, the flux linkage angle γ is kept constant, and the torque-to-current ratio J(γ) = T is used as the optimization variable. e (γ) / I s (γ) is the objective function. The optimal flux linkage angle is obtained by performing the golden section method within the flux linkage circle constraint, where T... e (γ) is the electromagnetic torque, I s (γ) represents the current amplitude; the corresponding motor d-axis and q-axis currents I are calculated based on the optimal flux linkage angle. d I q With magnetic flux ψ s , to obtain (ψ s I d I q FC-MTPA point; Three sets of constant stator flux linkage amplitudes are selected and the conditional MTPA solver is called respectively to obtain three sets of FC-MTPA points, which are recorded as fixed table points in the FC-MTPA point lookup table. Select two sets of active stator flux linkage amplitudes and call the conditional MTPA solver respectively to obtain two sets of FC-MTPA points. Record these as active table points in the FC-MTPA point lookup table, and then use I... q The instruction value locates adjacent table points; In the speed loop interruption, according to real-time I q I is generated by linear interpolation between adjacent table points. dref The reference value is filtered by a first-order low-pass filter and output to the current loop for control. The two sets of active meter points are updated in real time when the update conditions are met, so as to achieve accurate interpolation of the current command.

2. The synchronous reluctance motor FC-MTPA control method as described in claim 1, characterized in that, It also includes optimizing efficiency on a flux circle with a constant flux amplitude, with the goal of maximizing the torque-to-current ratio.

3. The synchronous reluctance motor FC-MTPA control method as described in claim 1, characterized in that, The optimization using the golden section method includes: shrinking the magnetic flux angle γ in the interval [0, π / 2] using the golden section method; when the width of the squeeze interval is less than Or the number of iterations reaches K max When, the output γ m =(a+b) / 2 is used as the optimal flux linkage angle estimate for this state, where K max Let [a, b] be a constant, and let [a, b] be the squeezing interval determined by the golden ratio.

4. The synchronous reluctance motor FC-MTPA control method as described in claim 1, characterized in that, According to real-time I qref I is generated by linear interpolation between adjacent table points. dref Reference values ​​include: determining the squeeze interval [L, R] based on adjacent table points, and linear interpolation: I dref =I d_L +(I qref - I q_L ) / (I q_R - I q_L )·(I d_R - I d_L ), where I qref I is the q-axis current command value. dref The d-axis current command value is obtained by interpolation through the FC-MTPA point lookup table, where L and R are two FC-MTPA points, and I... d_L I d_R Let I be the d-axis current at the two FC-MTPA points L and R. q_L I q_R Let L and R be the q-axis currents at the two FC-MTPA points.

5. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 4.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 4.

7. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or instructions are executed by a processor, they implement the steps of the method as described in any one of claims 1 to 4.

Images