Motor drive switching current surge modeling and model predictive control method

Abstract

The application discloses a motor driving switch current surge modeling and model predictive control method, and belongs to the technical field of power generation, power transformation or power distribution. The modeling method comprises the following steps: establishing a motor voltage equation and a magnetic motive force equation; calculating an expression of motor stator magnetic motive force about current and permanent magnet flux linkage; calculating an expression of stator magnetic resistance about stator inductance; simultaneously solving the magnetic circuit equation and the voltage equation, and substituting the magnetic motive force expression to calculate the stator flux linkage expression; substituting the stator flux linkage expression into the voltage equation, and replacing the equivalent magnetic induction coefficient before the voltage differential term with the differential magnetic induction coefficient to obtain a permanent magnet motor model considering the switch current surge. The modeling method effectively improves the motor modeling accuracy. The application also provides a model predictive control method and a measurement method for quantitatively describing the differential magnetic induction coefficient of the switch current surge, which can effectively improve the operation performance of the motor variable current control system and improve the practicability of the modeling method.

Description

Technical Field

[0001] The present invention relates to a motor system, and in particular discloses a motor drive switch current surge modeling and model prediction control method, belonging to the technical field of power generation, power transformation or power distribution. Background Art

[0002] Permanent magnet synchronous motors (PMSMs) have advantages such as simple structure, high power density, high efficiency, and excellent control performance. They are widely used in industrial fields such as energy production and electric transportation. An accurate motor model is the foundation for achieving high-performance control. Therefore, improving the PMSM system model and enhancing its modeling accuracy are crucial for enhancing the operational performance of PMSM drive systems.

[0003] Alternating magnetic flux generates eddy currents in the stator of a permanent magnet motor. The magnetomotive force generated by the eddy currents, in turn, changes the amplitude and phase of the source magnetic flux. This effect is known as eddy current reaction. The fundamental component of the magnetic flux density generates fundamental-frequency eddy currents in the stator core, thereby causing fundamental-frequency eddy current reaction. The high-frequency harmonic components of the magnetic flux density also generate high-frequency eddy currents in the stator core, causing high-frequency eddy current reaction. Furthermore, permanent magnet motor inverter control systems typically use a frequency converter to power the motor. The high-frequency switching action of the power electronic devices used in these converters causes high-frequency voltage and current variations, inducing high-frequency eddy currents in the core and generating switching current surges. However, existing permanent magnet motor models typically use fundamental-frequency iron loss resistors to equivalently model the iron loss caused by eddy currents. They fail to account for the high-frequency eddy current reaction caused by the high-frequency switching action of the converter power devices. As a result, the established motor models are generally limited to steady-state control such as efficiency optimization, making it difficult to accurately model the switching current surges caused by eddy current reaction, limiting the improvement of motor control performance. Summary of the Invention

[0004] The purpose of the present invention is to address the shortcomings of the above-mentioned background technology and provide a motor drive switching current surge modeling and model predictive control method to solve the technical problem that the existing motor modeling method does not take into account the switching current surge caused by eddy current reaction, thereby achieving the purpose of accurately establishing a motor model.

[0005] The present invention adopts the following technical solutions to achieve the above-mentioned purpose:

[0006] A method for modeling a motor drive switch current surge includes the following steps:

[0007] Step 1: Considering the eddy current reaction, a magnetic induction element is introduced into the motor magnetic circuit model, and the motor circuit equation and magnetic circuit equation are established. The introduced magnetic induction parameters are calculated based on the equivalent magnetic induction coefficient;

[0008] Step 2, calculate the expression of stator magnetomotive force with respect to current and permanent magnet flux;

[0009] Step 3, calculate the expression of stator magnetic resistance with respect to stator inductance;

[0010] Step 4: Combine the motor circuit equation and the magnetic circuit equation, and substitute the stator magnetomotive force expression with respect to the current and the permanent magnet flux linkage to calculate the stator flux linkage expression;

[0011] Step 5: Substitute the stator flux expression into the motor circuit equation to obtain the permanent magnet motor model;

[0012] Step 6: Replace the equivalent magnetic induction coefficient of the voltage differential term in the permanent magnet motor model obtained in step 5 with the differential magnetic induction coefficient that quantitatively describes the switching current surge, to obtain a permanent magnet motor model that takes the switching current surge into account.

[0013] As a further optimization scheme for the motor drive switching current surge modeling method, in step 1, the magnetic induction parameter introduced is expressed as Calculate, where is the magnetic induction caused by eddy current reaction, k is the equivalent magnetic induction coefficient, N a is the effective number of turns of each phase winding of the motor.

[0014] As a further optimization scheme for the motor drive switching current surge modeling method, the motor circuit equation and magnetic circuit equation established in step 1 are

[0015] Among them, u d 、u q are the d-axis and q-axis stator voltages, R s is the stator winding resistance, i d and i q are the d-axis and q-axis stator currents, ψ d and ψ q are the d-axis and q-axis stator flux respectively, ω is the electrical angular frequency, and are the d-axis and q-axis stator magnetomotive force respectively, and are the d-axis and q-axis magnetic resistances, Φ d and Φ q are the d-axis and q-axis magnetic fluxes respectively.

[0016] As a further optimization scheme for the motor drive switching current surge modeling method, the expression of the stator magnetomotive force calculated in step 2 with respect to current and permanent magnet flux is in, is the permanent magnet magnetomotive force, ψ f is the permanent magnet flux.

[0017] As a further optimization of the motor drive switching current surge modeling method, the stator reluctance calculated in step 3 with respect to the stator inductance is expressed as Among them, L d and L q are the d-axis and q-axis inductances respectively.

[0018] As a further optimization scheme for the motor drive switching current surge modeling method, the stator flux expression calculated in step 4 is

[0019] As a further optimization scheme for the motor drive switching current surge modeling method, the permanent magnet motor model obtained in step 5 is Among them, L dd , L qq , ψ f1 , ψ f2 , L dq , L d1 , L q1 For intermediate variables:

[0020] The permanent magnet motor model taking into account the switching current surge obtained in step S6 is: Where g is the differential magnetic induction coefficient that quantitatively describes the switching current surge.

[0021] A motor model predictive control method,

[0022] The surge current caused by the dq axis current minus the dq axis stator voltage differential term is defined as the dq axis virtual current;

[0023] Taking the dq-axis virtual current as the control object, the permanent magnet motor model taking into account the switching current surge is discretized into the state control equation;

[0024] The candidate voltage vectors are evaluated using the trajectory extrapolation cost function method, and the candidate voltage vector with the smallest cost function value is selected as the final output voltage vector.

[0025] As a further optimization scheme of the motor model predictive control method,

[0026] The state control equation is Among them, i vd (n+1), i vq (n+1)T s The predicted value of virtual current of d-axis and q-axis at the moment, i vd (n), i vq (n) are nT s Predicted values ​​of virtual currents on the d-axis and q-axis at time Ts is the discrete cycle of the controller, A p is the state matrix of the state control equation, B p is the input matrix of the state control equation, f ψp is the column vector related to the permanent magnet flux linkage of the permanent magnet motor model taking into account the switching current surge,

[0027] The specific method of evaluating the candidate voltage vector using the trajectory extrapolation value function is as follows: vd (n+2) and i vq (n+2) is the condition for the candidate voltage vector to be extrapolated within the allowable error range of each reference value. When the candidate voltage vector meets the extrapolation condition, the first value function g1 is used to evaluate all candidate voltage vectors that meet the extrapolation condition. When the candidate voltage vector does not meet the extrapolation condition, the second value function g2 is used to evaluate all candidate voltage vectors that meet the extrapolation condition.

[0028] i vd (n+2), i vq (n+2)T s The predicted values ​​of the d-axis and q-axis virtual currents at the moment, T is the maximum number of sampling cycles to maintain the d-axis virtual current and the q-axis virtual current within the allowable error range under the continuous action of the current candidate vector, n switch is the switching function of the converter, S x (n) is the x-phase nT s The switching function at the moment, S x (n+1) is the x phase (n+1)T s The switching function at the moment, when l=vd, i vd max 、i vd min They are the maximum and minimum values ​​allowed for the d-axis virtual current in the control, respectively. When l = vq, i vq max 、i vq min are the maximum and minimum values ​​allowed for the q-axis virtual current in control respectively.

[0029] A measurement method for the differential magnetic inductance that quantitatively describes the switching current surge in a permanent magnet motor model taking into account the switching current surge.

[0030] Collect the stator current waveform and stator voltage mutation value during current transient switching under different stator voltages;

[0031] The composite expression of exponential function and linear function is used to fit the stator current waveform during current transient switching and calculate the stator current mutation value;

[0032] Taking the stator voltage mutation value as the independent variable and the stator current mutation value as the dependent variable, linear fitting is performed on different groups of stator voltage mutation values ​​and stator current mutation values. The slope of the fitted straight line is the measurement result of the differential magnetic induction coefficient that quantitatively describes the switching current surge.

[0033] The present invention adopts the above technical solution and has the following beneficial effects:

[0034] (1) Based on the vector magnetic circuit theory, the present invention establishes a permanent magnet motor model that takes into account the switching current surge, and explains the physical principle of the switching current surge, that is, the transient change of the stator terminal voltage caused by the switching action of the inverter causes the switching secondary eddy current induced in the stator to produce a reaction, which corresponds to the newly added stator terminal voltage differential term in the motor model, thereby improving the accuracy of the motor model.

[0035] (2) The present invention provides a method for measuring the differential magnetic induction coefficient that quantitatively describes the switching current surge in a motor model taking into account the switching current surge. It only requires setting the inverter to operate in a specific voltage output mode, and no additional hardware circuit is required. The switching current surge differential magnetic induction coefficient of different motors can be easily obtained, ensuring the practicality of the present invention.

[0036] (3) The permanent magnet motor model taking into account the switching current surge established based on the vector magnetic circuit theory of the present invention has higher modeling accuracy because it takes into account the influence of the switching secondary eddy current reaction, and can reduce the current online prediction error. Based on the newly established model, a model predictive control method is provided, which discretizes the permanent magnet motor model taking into account the switching current surge into a state control equation, defines the current value after deducting the surge current from the dq axis current as a virtual current, and implements model predictive control with the virtual current as the control object. It can reduce the switching frequency while keeping the current harmonic performance unchanged, reduce the current prediction error, and improve the operating performance of the motor system. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0038] Figure 1 It is the d-axis equivalent circuit diagram corresponding to the existing vector magnetic circuit permanent magnet motor model.

[0039] Figure 2is the q-axis equivalent circuit diagram corresponding to the existing vector magnetic circuit permanent magnet motor model.

[0040] Figure 3 is the permanent magnet synchronous motor modeling method flow chart of the application.

[0041] Figure 4 is the d-axis equivalent circuit diagram corresponding to the permanent magnet motor model of the application.

[0042] Figure 5 is the q-axis equivalent circuit diagram corresponding to the permanent magnet motor model of the application.

[0043] Figure 6 is the motor variable current control system topology structure diagram of the embodiment of the application.

[0044] Figure 7 is the switching current surge waveform experimental result diagram of the embodiment of the application.

[0045] Figure 8 is the differential magnetic induction coefficient measurement principle diagram provided by the application for quantitatively describing the switching current surge.

[0046] Figure 9 is the measurement result diagram of the differential magnetic induction coefficient in the motor model of the application for quantitatively describing the switching current surge.

[0047] Figure 10 is a predictive control block diagram based on the existing vector magnetic circuit motor model or the motor model of the application.

[0048] Figure 11 is the stator current prediction error experimental result diagram using the existing vector magnetic circuit motor model for control.

[0049] Figure 12 is the stator current prediction error experimental result diagram using the motor model of the application for control.

[0050] Figure 13 is the stator current waveform diagram using the existing vector magnetic circuit motor model for control.

[0051] Figure 14 is the stator current waveform spectrum diagram using the existing vector magnetic circuit motor model for control.

[0052] Figure 15 is the stator current waveform diagram using the motor model of the application for control.

[0053] Figure 16 is the stator current waveform spectrum diagram using the motor model of the application for control. DETAILED DESCRIPTION

[0054] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present invention.

[0055] like Figure 1 and Figure 2 The equivalent circuit diagram of the existing vector magnetic circuit motor model shown in Figure 1 includes the stator winding resistance R s , the equivalent resistance ωL caused by eddy current reaction dq , d-axis inductance L considering eddy current reaction d1 , coupled potential E considering eddy current reaction dv , d-axis stator voltage u d , q-axis inductance L considering eddy current reaction q1 , coupled potential E considering eddy current reaction qv and q-axis stator voltage u q According to the equivalent circuit structure, the motor model can be obtained as shown in formula (1):

[0056]

[0057] Among them, i d and i q are the d-axis and q-axis stator currents, ψ d and ψ q are the d-axis and q-axis stator flux, ω is the electrical angular frequency, L dq , ψ f1 , ψ f2 is the intermediate variable,

[0058]

[0059] E dv =-ωL q1 i q +ωψ f1 +L dq di q / dt

[0060] E qv =ωL d1 i d +ωψ f2 -L dq di d / dt

[0061]

[0062] like Figure 3As shown, a motor drive switching current surge modeling and model predictive control method includes the following 6 steps.

[0063] S1, establish motor circuit equation and magnetic circuit equation, respectively as shown in formula (2) and (3):

[0064]

[0065] Among them, and The d-axis and q-axis stator magnetic motive force, and The d-axis and q-axis reluctance, Φ d andΦ q The d-axis and q-axis flux, The magnetic induction caused by eddy current reaction, k is the equivalent magnetic induction coefficient, N a The effective number of turns of each phase winding of the motor.

[0066] S2, calculate the expression of stator magnetic motive force about current and permanent magnet flux linkage. The expression of motor armature winding magnetic motive force is:

[0067]

[0068] Among them, the permanent magnet magnetic motive force Can be expressed as:

[0069]

[0070] Among them, ψ f Permanent magnet flux linkage.

[0071] S3, the expression of stator reluctance about stator inductance is:

[0072]

[0073] Among them, L d And L q The d-axis and q-axis inductance.

[0074] S4, the magnetic circuit equation and the motor circuit equation are solved, and the expression of stator flux linkage is calculated by substituting the expression of stator magnetic motive force:

[0075]

[0076] S5, substitute the expression of stator flux linkage into the voltage equation, which can be obtained:

[0077]

[0078] Among them, L dd , L qqis an intermediate variable,

[0079]

[0080] S6, considering that the voltage differential term involves the switching sub-voltage transient process, the equivalent magnetic induction coefficient k of the voltage differential term is replaced by the differential magnetic induction coefficient g that quantitatively describes the switching current surge. The differential magnetic induction coefficient g can quantitatively describe the voltage differential term du d / dt,du q / dt and current differential term di d / dt,di q The mathematical relationship between / dt is used to obtain the permanent magnet motor model taking into account the switching current surge:

[0081]

[0082] Comparing equations (1) and (10), it can be found that the eddy current reaction makes the motor model more complicated, and the differential term of the stator voltage appears, which means that theoretically the current can be discontinuous in time, which is the direct cause of the switching current surge.

[0083] like Figure 4 and Figure 5 The figure shows the equivalent circuit diagram corresponding to the motor model of the present invention. The d-axis circuit includes the stator resistance R s , equivalent resistance ωL dq , equivalent inductance (R s k+1)L d1 、Electric potential ωψ f1 -ωL q1 i q , potential (R s k+1)L qq di q / dt-L d1 gdu d / dt-L qq gdu q / dt. The q-axis circuit includes the stator resistance R s , equivalent resistance ωL dq , equivalent inductance (R sk +1)L q1 、Electric potential ωψ f2 +ωL d1 i d and potential-(R s k+1)L dd di d / dt+L dd gdu d / dt-L q1 gdu q / dt. It can be seen that Figure 4 and Figure 5 The equivalent circuits of the motor model of the present invention and the existing vector magnetic circuit model are significantly different. The coupling relationship between the dq axes is more complicated, and a voltage differential term is added.

[0084] like Figure 6 The topology diagram of the three-phase permanent magnet motor drive system in this embodiment is shown in FIG. , which is composed of a three-phase two-level converter and a three-phase permanent magnet motor. Figure 6 As shown in the right figure, there are eight voltage vectors, six of which are effective voltage vectors, and the remaining two are zero voltage vectors. By stipulating that the upper arm on is marked as "1" and the lower arm on is marked as "0," a three-digit sequence can represent eight different voltage vectors. For example, the vector "100" indicates that the upper arm of phase A is on and the lower arm is off, while both phases B and C are in the switching state with the lower arm on and the upper arm off.

[0085] like Figure 7 The following figure shows the experimental results of switching current surges caused by eddy current reaction. The waveform is the motor phase A current waveform, measured by a high-bandwidth current probe. The converter is set to a special operating mode, repeatedly outputting the vector sequence 000-100-000-011-000, while the motor remains stationary. The corresponding vector sequence changes the stator voltage of the motor phase A in the following pattern: 0, 2U dc / 3, 0, -2U dc / 3, 0. From Figure 7 It can be seen from the experimental waveform of the A-phase current that the current change trend is generally slow change (corresponding to vector 000) - rapid increase (corresponding to vector 100) - slow change (corresponding to vector 000) - rapid decrease (corresponding to vector 011) - slow change (corresponding to vector 000). Specifically, as Figure 7 As shown in the right figure, when the current switches from a rapidly rising state to a slowly changing state, there is an extremely short transient change process. This phenomenon is called a switching current surge, which is caused by the reaction of eddy currents generated by the voltage mutation caused by the switching action of the power device. In the motor model established by the present invention, the switching current surge is reflected in the voltage differential term.

[0086] like Figure 8 The present invention provides a method for measuring the differential magnetic induction coefficient that quantitatively describes the switching current surge. As shown in the motor model of formula (10), the value of the current transient change is proportional to the value of the voltage transient change. Therefore, in Figure 7 When the switch state switches from vector 100 to 000, the current will first experience a rapid transient state, and then approximately maintain a constant slope state change. By fitting the current waveform using a composite objective function of an exponential function and a linear function, the current increment during rapid changes can be determined. The fitting objective function expression is shown in Equation (11):

[0087]

[0088] where λ1~λ4 are coefficients to be fitted, the current transient increment Δi A The expression is:

[0089] Δi A = λ4-i A (0) (12)

[0090] Figure 8 The fitting result of the transient condition is that the current increment is -0.059 A, and the voltage increment Δu A is -66.7 V. By changing the DC bus voltage value, the voltage increment can be changed, and the current waveform data is measured again to fit the corresponding current increment, so that a set of voltage and current increment data can be obtained. According to the motor model shown in equation (10), the set of voltage and current increment data should be linear and the slope is the switch current surge differential magnetic induction coefficient to be solved.

[0091] As Figure 9 shown is the fitting result of the voltage and current increment when the DC bus voltage in this embodiment changes from 80 V to 180 V. From the result, it can be seen that the data has good linearity, which verifies the correctness of the motor model, and the measured switch current surge differential magnetic induction coefficient value is 0.000799 Ω -1 . The measurement method of the equivalent magnetic induction coefficient k has been reported in documents and published patents, and the measured value of the equivalent magnetic induction coefficient k in this embodiment is 0.001773 Ω -1 .

[0092] As Figure 10 shown, in this application, an embodiment of the model-based motor control algorithm based on the vector magnetic circuit theory considering the eddy current reaction permanent magnet motor model comprises: a speed closed-loop PI regulator, a model predictive control module, a three-phase two-level inverter, a three-phase permanent magnet synchronous motor, a coordinate transformation module, a control delay compensation current prediction, a position encoder and a speed calculation module. This embodiment is used to illustrate that the model of the application has higher current prediction accuracy and can provide better motor control performance. Among them, the existing vector magnetic circuit motor model and the motor model considering switch current surge established by the application are discretized using the first-order Euler method, and are arranged into state control equations. The discrete state space model of the existing vector magnetic circuit motor model is:

[0093]

[0094] where T s is the controller discrete period, A c is the state matrix of the state control equation of the existing vector magnetic circuit motor model, and Bc is the input matrix of the state control equation of the existing vector magnetic circuit motor model, f ψc is the column vector related to the permanent magnet flux linkage of the existing vector magnetic circuit motor model. The specific expression of the related variables is:

[0095]

[0096] Among them, L d2 , L q2 is an intermediate variable.

[0097] The discrete state space model of the motor model shown in formula (10) of the present invention is shown in formula (15), with the dq axis virtual current i vd and i vq For state variables:

[0098]

[0099] Among them, i vd (n+1), i vq (n+1) is (n+1)T s The predicted value of virtual current of d-axis and q-axis at the moment, i vd (n), i vq (n) is nT s Predicted values ​​of virtual currents on the d-axis and q-axis at time T s is the controller discrete cycle, A p is the state matrix of the motor model state control equation of the present invention, B p is the input matrix of the motor model state control equation of the present invention, f ψp is the column vector related to the permanent magnet flux linkage of the motor model of the present invention, A p 、B p 、f ψp The specific expression is:

[0100]

[0101] In formula (15), i vd (n) and i vq The expression for (n) is:

[0102]

[0103] Actual (n+1)T s The predicted dq axis current at the moment can be calculated as:

[0104]

[0105] Among them, i d (n+1), i q (n+1) is (n+1)Ts The predicted values ​​of d-axis and q-axis current at time i d (n), i q (n) is nT s The predicted values ​​of d-axis and q-axis current at the moment, u d (n),u q (n) is nT s Predicted values ​​of d-axis and q-axis voltage at time u d (n-1), u q (n-1) is (n-1)T s Predicted values ​​of the d-axis and q-axis voltages at the moment.

[0106] Considering one-beat delay compensation, the voltage vector determined in this cycle is actually from (n+1)T s The control goal is to make (n+1)T s Moment i vd and i vq Track the corresponding reference value. s Candidate voltage vector u at time d_c and u q_c Under the action of (n+2)T s Moment i vd and i vq The predicted value of can be calculated as:

[0107]

[0108] The conditions under which the candidate voltage vector can be extrapolated are defined as follows: i vd (n+2) and i vq (n+2) is within the allowable error range of the corresponding variable reference value. The principle of selecting the value function of trajectory extrapolation model predictive control is as follows: if there is a candidate vector that can be extrapolated, then use the value function g1 to evaluate all candidate voltage vectors that can be extrapolated, and select the candidate voltage vector with the smallest value as the final output voltage vector.

[0109]

[0110] Among them, T is the ability to maintain i under the continuous action of the candidate vector vd and i vq The maximum number of sampling cycles that does not exceed the allowable error range. switch is the switching function of the two-level converter, which is expressed as:

[0111]

[0112] Among them, S xis the switching function of phase A, phase B, or phase C. When the upper bridge arm of the phase is on and the lower bridge arm is off, the switching function is equal to 1; when the lower bridge arm of the phase is off and the upper bridge arm is on, the switching function is equal to 0.

[0113] When there is no candidate vector that can be extrapolated, the cost function g2 is used to evaluate all candidate vectors, and the candidate voltage vector with the smallest value is selected as the final output voltage vector.

[0114]

[0115] Among them, i vdmax and i vdmin I vd The maximum and minimum values ​​allowed for the variable, and their average value is equal to i vd Reference value, i vqmax and i vqmin I vq The maximum and minimum values ​​allowed for the variable, and their average value is equal to i vq Reference value.

[0116] The experimental results of the current prediction accuracy of the existing model and the model of the present invention are as follows: Figure 11 and Figure 12 The parameters of the permanent magnet motor in the experiment are as follows: the number of pole pairs is 10, the q-axis inductance is 21.2mH, the d-axis inductance is 16.9mH, the permanent magnet flux is 0.1936Wb, the stator resistance is 0.85Ω, and the sampling frequency is 15kHz. Figure 11 The waveform corresponding to the left half of the time axis in the left figure is the current prediction error waveform of the existing model, and the waveform corresponding to the right half of the time axis is the current prediction error waveform of the model of the present invention. The right figure is the enlarged waveform of the left figure. id and e iq The experimental results show that the error of the motor dq axis current prediction by the present invention is significantly smaller than that by the existing model, which verifies the effectiveness of the present invention.

[0117] The experimental results after the existing model and the model of the present invention are used to predict the control are as follows Figures 13 to 16 As shown. Among them, Figure 13 and Figure 14 The following are the experimental waveforms and spectrum results of the motor stator current when predictive control is performed using the existing model. Under the current operating conditions, the motor phase current THD value is 5.47%, and the average switching frequency of the converter is 1870Hz. Figure 15 and Figure 16 The following are the experimental waveforms and spectrum results of the motor stator current when the model of the present invention is used for predictive control. Under the same working conditions, the motor phase current THD value is 5.49% and the average switching frequency of the converter is 1781Hz. Figure 13-14 and Figure 15-16 The results show that using the existing model reduces the switching frequency by 5%, to 89Hz, while maintaining essentially unchanged phase current THD. Therefore, the experimental results demonstrate that the permanent magnet motor model and model predictive control algorithm, which account for switching current surges, can effectively improve the performance of motor control systems.

[0118] Throughout this specification, references to terms such as "one embodiment," "example," or "specific example" indicate that the specific features, structures, materials, or characteristics described in conjunction with that embodiment or example are included in at least one embodiment or example of the present invention. In this specification, schematic representations of these terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

[0119] The basic principles, main features, and advantages of the present invention are shown and described above. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions are merely illustrative of the principles of the present invention. Various changes and modifications may be made to the present invention without departing from the spirit and scope of the present invention, and such changes and modifications fall within the scope of the invention as claimed.

Claims

1. A method for modeling current surge of a motor drive switch, characterized in that: The steps include: Step 1: Considering the eddy current reaction, introduce magnetic induction elements into the motor magnetic circuit model, and establish the motor circuit equation and magnetic circuit equation. The introduced magnetic induction parameters are calculated based on the equivalent magnetic induction coefficient. The established motor circuit equation and magnetic circuit equation are: Among them, u d 、u q are the d-axis and q-axis stator voltages, R s is the stator winding resistance, i d and i q are the d-axis and q-axis stator currents, ψ d and ψ q are the d-axis and q-axis stator flux respectively, ω is the electrical angular frequency, and are the d-axis and q-axis stator magnetomotive force respectively, and are the d-axis and q-axis magnetic resistances, Φ d and Φ q are the d-axis and q-axis magnetic fluxes respectively, The introduced magnetic induction parameter is expressed as Calculate, where is the magnetic induction caused by eddy current reaction, k is the equivalent magnetic induction coefficient, N a is the effective number of turns of each phase winding of the motor; Step 2: Calculate the expression of stator magnetomotive force with respect to current and permanent magnet flux: in, is the permanent magnet magnetomotive force, ψ f is the permanent magnet flux; Step 3, calculate the expression of stator reluctance with respect to stator inductance: Among them, L d and L q are the d-axis and q-axis inductances respectively; Step 4: Combine the motor circuit equation and the magnetic circuit equation, and substitute the stator magnetomotive force expression with respect to the current and permanent magnet flux to calculate the stator flux expression: Step 5: Substitute the stator flux expression into the motor circuit equation to obtain the permanent magnet motor model: Among them, L dd , L qq , ψ f1 , ψ f2 , L dq , L d1 , L q1 For intermediate variables: Step 6: Replace the equivalent magnetic inductance of the voltage differential term in the permanent magnet motor model obtained in step 5 with the differential magnetic inductance that quantitatively describes the switching current surge, and obtain a permanent magnet motor model that takes the switching current surge into account: Where g is the differential magnetic induction coefficient that quantitatively describes the switching current surge.

2. The motor drive switch current surge modeling method according to claim 1, characterized in that: The measurement method of the differential magnetic inductance for quantitatively describing the switching current surge in step 6 is: Collect the stator current waveform and stator voltage mutation value during current transient switching under different stator voltages; The composite expression of exponential function and linear function is used to fit the stator current waveform during current transient switching and calculate the stator current mutation value; Taking the stator voltage mutation value as the independent variable and the stator current mutation value as the dependent variable, linear fitting is performed on different groups of stator voltage mutation values ​​and stator current mutation values. The slope of the fitted straight line is the measurement result of the differential magnetic induction coefficient that quantitatively describes the switching current surge.

3. A motor model predictive control method, characterized in that: The surge current caused by the dq axis current minus the dq axis stator voltage differential term is defined as the dq axis virtual current; Taking the dq-axis virtual current as the control object, the permanent magnet motor model taking into account the switching current surge obtained in claim 1 is discretized into a state control equation; The candidate voltage vectors are evaluated using the trajectory extrapolation cost function method, and the candidate voltage vector with the smallest cost function value is selected as the final output voltage vector.

4. A motor model predictive control method according to claim 3, characterized in that: The state control equation is Among them, i vd (n+1), i vq (n+1)T s The predicted value of virtual current of d-axis and q-axis at the moment, i vd (n), i vq (n) are nT s Predicted values ​​of virtual currents on the d-axis and q-axis at time T s is the discrete cycle of the controller, A p is the state matrix of the state control equation, B p is the input matrix of the state control equation, f ψp is the column vector related to the permanent magnet flux linkage of the permanent magnet motor model taking into account the switching current surge, The specific method of evaluating the candidate voltage vector using the trajectory extrapolation value function is as follows: vd (n+2) and i vq (n+2) is the condition for the candidate voltage vector to be extrapolated within the allowable error range of each reference value. When the candidate voltage vector meets the extrapolation condition, the first value function g1 is used to evaluate all candidate voltage vectors that meet the extrapolation condition. When the candidate voltage vector does not meet the extrapolation condition, the second value function g2 is used to evaluate all candidate voltage vectors that meet the extrapolation condition. (n+2)T s The predicted values ​​of the d-axis and q-axis virtual currents at the moment, T is the maximum number of sampling cycles to maintain the d-axis virtual current and the q-axis virtual current within the allowable error range under the continuous action of the current candidate vector, n switch is the switching function of the converter, S x (n) is the x-phase nT s The switching function at the moment, S x (n+1) is the x phase (n+1)T s The switching function at the moment, when l=vd, i vdmax 、i vdmin They are the maximum and minimum values ​​allowed for the d-axis virtual current in the control, respectively. When l = vq, i vqmax 、i vqmin are the maximum and minimum values ​​allowed for the q-axis virtual current in control respectively.

Images