Lumped disturbance parameter observation and permanent magnet synchronous motor control method and system

Abstract

The present invention discloses a lumped disturbance parameter observation and permanent magnet synchronous motor control method and system. This control method samples motor phase current for each control cycle and designs a two-degree-of-freedom filter to filter out DC bias and high-frequency noise signals. Specifically, two low-pass filters are connected in parallel and interpolated. By adjusting their cutoff frequencies, high-frequency noise signals and DC bias are independently suppressed, achieving decoupling of the filtering effect. Based on the filtered current signal, a super-helical sliding mode observer is constructed, which includes an equivalent feedback gain and an exponential term, to observe the lumped disturbance parameter. The equivalent feedback gain has an integral form, which can effectively suppress high-frequency noise. The additional exponential term smoothes the dynamic convergence process of the observer. The lumped disturbance parameter observation value is then used for model-free current prediction of the motor. This method can effectively filter out DC bias and high-frequency noise in the current sampling signal, improving the observation and control accuracy of the super-helical sliding mode observer.

Description

Technical Field

[0001] The present invention belongs to the technical field of motor control, and more specifically, relates to a lumped disturbance parameter observation and permanent magnet synchronous motor control method and system. Background Art

[0002] Model predictive control (MPC) has attracted widespread attention in the control of permanent magnet synchronous motors (PMSMs) due to its simple design, superior dynamic performance, and ability to simultaneously optimize multiple control objectives. Its fast dynamic response is highly compatible with the high-precision, high-speed requirements of PMSMs. However, MPC is inherently a model-based control method, and its performance is highly dependent on the accuracy of the motor model. When motor parameters (such as inductance, resistance, and flux linkage) vary or mismatch, control errors increase significantly, leading to a decrease in system performance. This parameter sensitivity limits the robustness and reliability of MPC in practical applications.

[0003] Therefore, improving the parameter robustness of model-predictive current control methods in permanent magnet synchronous motors has become a key research direction. By introducing model-free predictive control strategies, the impact of model parameter mismatch on control performance can be effectively reduced, thereby enhancing system stability and adaptability. This not only broadens the application scope of model-predictive current control but also further unleashes the potential of permanent magnet synchronous motors in high-performance drive systems.

[0004] However, existing model-free predictive control for permanent magnet synchronous motors often employs a first-order hyperlocal model, and the calculation of the predicted current relies on current sampling and lumped disturbance parameters. Consequently, the DC bias and high-frequency noise of the current sampling, as well as the accuracy of the lumped disturbance parameter observation, directly impact the accuracy of the predicted current, and thus the motor control performance. Most existing methods directly utilize the sampled current or only perform low-pass filtering, failing to remove the effects of the DC bias. Furthermore, lumped disturbance observers can be constructed using sliding mode observers, state observers, and other techniques. However, model-free control methods based on conventional sliding mode observers or state observers still suffer from issues such as high-frequency noise and poor dynamic performance, requiring further optimization and improvement. Summary of the Invention

[0005] The calculation of predicted current in the model-free predictive control of permanent magnet synchronous motors based on a first-order hyperlocal model relies on current sampling and lumped disturbance parameters. Conventional model-free control methods based on sliding mode observers or state observers suffer from high-frequency noise and poor dynamic performance. The present invention provides a method and system for lumped disturbance parameter observation and permanent magnet synchronous motor control. A novel super-helical sliding mode observer is designed, which uses an equivalent feedback gain and exponential term to eliminate the influence of high-frequency noise on the lumped disturbance parameter observation results and smooth the dynamic convergence process. Finally, a first-order hyperlocal model is combined with current prediction to improve the parameter robustness and control accuracy of the permanent magnet synchronous motor predictive control.

[0006] To this end, the present invention provides the following technical solutions:

[0007] In one aspect, the present invention provides a permanent magnet synchronous motor control method, which includes performing the following steps:

[0008] S1. Get the αβ axis current i of the permanent magnet synchronous motor in the current control cycle αβ and speed ω r and the voltage vector u αβ ;

[0009] S2, based on αβ axis current i αβ and voltage vector u αβ , a super-helical sliding mode observer including an equivalent feedback gain and an exponential term is introduced, and the observed value of the current is obtained using the super-helical sliding mode observer and the lumped disturbance parameter observations ;

[0010] S3, using αβ axis current i αβ , voltage vector u αβ and the lumped disturbance parameter observations The model-free current prediction model of the permanent magnet synchronous motor is substituted into the predicted αβ axis current value in the next control cycle, which is used for optimal voltage vector screening. Then, the inverter switching signal is generated based on the optimal voltage vector and applied to the inverter of the permanent magnet synchronous motor for drive control.

[0011] The mathematical model of the super-spiral sliding mode observer including the equivalent feedback gain and the exponential term is:

[0012] ;

[0013] Where α is a non-physical parameter, namely the voltage vector u αβ The coefficient of represents the observed value of the equivalent feedback parameter of the αβ axis, , Δi αβ The αβ axis current observation value represented by Relative to the αβ axis current sampling value i αβ error; sgn() is the sign function; k1, k3 and k4 are gain parameters, and t is time.

[0014] Preferably, the observed value of the equivalent feedback parameter of the αβ axis satisfy:

[0015] ;

[0016] Where k2 is the gain parameter.

[0017] Preferably, step S1 obtains the αβ axis current i of the permanent magnet synchronous motor in the current control cycle αβ After that, a two-degree-of-freedom filter is introduced to the current i αβ Perform pretreatment;

[0018] The two-degree-of-freedom filter includes a first low-pass filter and a second low-pass filter connected in parallel for subtraction, and the output signals of the first low-pass filter and the second low-pass filter are subtracted to obtain the output signal of the two-degree-of-freedom filter. The transfer function is expressed as:

[0019] ;

[0020] Where G, G LPF1 and G LPF2 Represent the transfer functions of the two-degree-of-freedom filter, the first low-pass filter, and the second low-pass filter respectively; ω c1 and ω c2 They represent the cutoff frequencies of the first and second low-pass filters respectively; s represents the differential operator.

[0021] Preferably, the cutoff frequency ω c1 and ω c2 The setting is 1.5ω r and 0.5ω r ,ω r is the rotation speed.

[0022] Preferably, the gain coefficients k1 and k2 satisfy:

[0023] ;

[0024] ;

[0025] Where Δi α , Δi β is the error Δi αβ The α-axis and β-axis components; for α and β axis components; Δm α, Δm β is the error Δm of the equivalent feedback signal αβ The α-axis and β-axis components, , is the actual value of the equivalent feedback signal, sup{·} is the supremum function; , m is the equivalent feedback parameter αβ The α and β axis components.

[0026] Among them, the model-free current prediction formula of permanent magnet synchronous motor is as follows:

[0027] ;

[0028] in, , , They represent the output voltage vector, current vector and observed lumped disturbance parameter vector of the inverter respectively; T s Indicates the control period.

[0029] The cost function for optimal voltage vector screening is expressed as follows:

[0030] ;

[0031] Among them, i αref and i βref Represent the α-axis component and β-axis component of the α-axis and β-axis stator current reference values ​​respectively; i α (t+1) and i β (t+1) represents the α-axis component and the β-axis component of the predicted current in the next control cycle t+1 respectively; g represents the cost function.

[0032] Preferably, the αβ axis current i of the permanent magnet synchronous motor in the current control cycle is αβ (t) and speed ω r (t) can be obtained by:

[0033] Collect the phase current i of the permanent magnet synchronous motor in the current control cycle abc , and convert it to the αβ axis to obtain the αβ axis current i αβ (t), i.e., the stator current;

[0034] Collect the position signal θ(t) of the permanent magnet synchronous motor in the current control cycle and Calculate the speed ω r (t).

[0035] In another aspect, the technical solution of the present invention provides a lumped disturbance parameter observation method applicable to a permanent magnet synchronous motor, which performs the following steps:

[0036] A super-helical sliding mode observer including equivalent feedback gain and exponential term is constructed;

[0037] Introducing the Lyapunov function to verify the stability of the super-helical sliding mode observer, and setting the gain parameters k1, k2, k3 and k4 in the super-helical sliding mode observer;

[0038] In each control cycle, the αβ axis current i of the permanent magnet synchronous motor is obtained αβ Then, the observed value of the current is identified based on the super spiral sliding mode observer. and the lumped disturbance parameter observations , for model-free current prediction of permanent magnet synchronous motors.

[0039] In a third aspect, the technical solution of the present invention further provides a control system based on the above-mentioned permanent magnet synchronous motor control method, comprising:

[0040] Feedback module, used to obtain the αβ axis current i of the permanent magnet synchronous motor in the current control cycle αβ and speed ω r and the voltage vector u αβ ;

[0041] Lumped disturbance parameter observation module, based on αβ axis current i αβ and voltage vector u αβ , a super-helical sliding mode observer including an equivalent feedback gain and an exponential term is introduced, and the observed value of the current is obtained using the super-helical sliding mode observer and the lumped disturbance parameter observations ;

[0042] Current prediction module, used to use the αβ axis current i αβ , voltage vector u αβ and the lumped disturbance parameter observations Substitute the model-free current prediction model of the permanent magnet synchronous motor to predict the αβ axis current prediction value in the next control cycle;

[0043] Optimal vector selection module, used for optimal voltage vector screening;

[0044] The control module is used to generate an inverter switching signal based on the optimal voltage vector and apply it to the inverter of the permanent magnet synchronous motor to achieve drive control.

[0045] In a fourth aspect, the technical solution of the present invention further provides a permanent magnet synchronous motor system, comprising:

[0046] A permanent magnet synchronous motor, an inverter and a control system for the permanent magnet synchronous motor are provided. The control system is connected to the permanent magnet synchronous motor and the inverter respectively. The midpoints of the three-phase bridge arms of the inverter are respectively connected to the three-phase windings of the permanent magnet synchronous motor.

[0047] In a fifth aspect, the technical solution of the present invention further provides a readable storage medium storing a control program, wherein the control program is called by a processor to implement:

[0048] Get the αβ axis current i of the permanent magnet synchronous motor in the current control cycle αβ and speed ω r and the voltage vector u αβ ;

[0049] Based on the αβ axis current i αβ and voltage vector u αβ , a super-helical sliding mode observer including an equivalent feedback gain and an exponential term is introduced, and the observed value of the current is obtained using the super-helical sliding mode observer and the lumped disturbance parameter observations ; Among them, the mathematical model of the super-helical sliding mode observer including the equivalent feedback gain and exponential term is:

[0050] ;

[0051] Where α is a non-physical parameter, namely the voltage vector u αβ The coefficient of represents the observed value of the equivalent feedback parameter of the αβ axis, , Δi αβ The αβ axis current observation value represented by Relative to the αβ axis current sampling value i αβ The error; sgn() is the sign function; k1, k3 and k4 are gain parameters, t is time;

[0052] Using the αβ axis current i αβ , voltage vector u αβ and the lumped disturbance parameter observations The model-free current prediction model of the permanent magnet synchronous motor is substituted into the predicted αβ axis current value in the next control cycle, which is used for optimal voltage vector screening. Then, the inverter switching signal is generated based on the optimal voltage vector and applied to the inverter of the permanent magnet synchronous motor to realize drive control.

[0053] Through the above technical solutions conceived by the present invention, the following progress can be achieved:

[0054] (1) The current filtering method based on a two-degree-of-freedom filter provided by the present invention utilizes the differences between two low-pass filters for filtering. Compared with the filtering method based on a second-order generalized integrator, the method provided by the present invention can achieve independent suppression of high-frequency noise signals and DC bias by adjusting the cutoff frequencies of the two low-pass filters, thus achieving decoupling of the filtering effect, effectively filtering out the DC bias and high-frequency noise of the sampled current, and effectively improving the accuracy of the lumped disturbance observer and controller.

[0055] (2) The super spiral sliding mode observer provided by the present invention is:

[0056] , its equivalent feedback gain has the form of integration , which can effectively suppress high-frequency noise; compared with the traditional super-helical sliding mode observer, the additional exponential term It can make the dynamic convergence process of the observer smoother.

[0057] (3) The model-predictive current control method for a permanent magnet synchronous motor provided by the present invention can effectively suppress the interference of high-frequency noise during the observation of lumped disturbance parameters. In addition, directly predicting the current using the observed lumped disturbance parameters can effectively avoid the impact of flux parameter offset on the predictive control accuracy, effectively improving the parameter robustness of the permanent magnet synchronous motor predictive control. BRIEF DESCRIPTION OF THE DRAWINGS

[0058] Figure 1 A Bode diagram of the transfer function of a two-degree-of-freedom filter of a permanent magnet synchronous motor provided by an embodiment of the present invention;

[0059] Figure 2 A flow chart of a model-free current prediction control method for a permanent magnet synchronous motor provided by an embodiment of the present invention;

[0060] Figure 3 A block diagram of a model-free current predictive control method for a permanent magnet synchronous motor provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0061] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only intended to illustrate the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below may be combined with each other as long as they do not conflict with each other.

[0062] In the present invention, the terms "first", "second", etc. (if any) in the present invention and the drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence.

[0063] Before explaining the technical solution of the present invention in detail, the existing model-free current prediction method is analyzed and explained. In the following description, unless otherwise specified, the basic principle adopted by the symbols is:

[0064] The superscript “^” indicates an observed quantity;

[0065] The subscript “αβ” represents the component of the αβ axis, and the subscripts “α” and “β” represent the α component and β component of the corresponding αβ axis parameters, respectively;

[0066] To this end, in response to technical problems such as high-frequency noise and poor dynamic performance in existing model-free control methods, the technical solution of the present invention provides a lumped disturbance parameter observation and permanent magnet synchronous motor control method and system, optimizes the current filtering method, the lumped disturbance parameter observation model and the permanent magnet synchronous motor control method, and its overall idea is to: design a two-degree-of-freedom filter, and filter out the DC bias and high-frequency noise in the sampled current by reasonably setting the cutoff frequency of the two-degree-of-freedom filter, thereby effectively improving the accuracy of the lumped disturbance observer and controller; design a super-helical sliding mode observer, and eliminate the influence of high-frequency noise on the lumped disturbance parameter observation results through equivalent feedback gain and exponential terms, and make the dynamic convergence process smoother; combine the first-order super-local model for current prediction, and improve the parameter robustness and control accuracy of the permanent magnet synchronous motor predictive control.

[0067] About 2-DOF filters:

[0068] The present invention proposes a method for determining the αβ axis current i of a permanent magnet synchronous motor. αβ (t) Perform preprocessing, that is, design a two-degree-of-freedom filter and set the cutoff frequency of the two-degree-of-freedom filter to filter out the DC bias and high-frequency noise in the sampled current.

[0069] Among them, the two-degree-of-freedom filter includes a first low-pass filter and a second low-pass filter that are connected in parallel for difference (that is, a signal is first connected in parallel to obtain two filtered signals, and then the signal of the first filter is subtracted from the signal of the second filter to obtain the final result). The output signal of the two-degree-of-freedom filter is subtracted from the output signal of the first low-pass filter and the second low-pass filter. Its Bode diagram is as follows Figure 1 As shown, the transfer function is expressed as:

[0070] ;

[0071] Where, G, G LPF1 and G LPF2 Represent the transfer functions of the two-degree-of-freedom filter, the first low-pass filter, and the second low-pass filter respectively; ω c1 and ω c2They represent the cutoff frequencies of the first and second low-pass filters respectively; s represents the differential operator.

[0072] The current filtering method based on the two-degree-of-freedom filter provided in this embodiment, as shown by G(0)=0, achieves the suppression of DC bias and high-frequency noise by making a difference in the parallel connection of two low-pass filters, eliminates the DC bias and high-frequency noise in the current sampling process, and provides a guarantee for the accuracy of the subsequent observer and controller. In addition, compared with the traditional filtering method based on the second-order generalized integrator, the designed method can be used to independently adjust ω c1 and ω c2 To achieve decoupling and suppression of DC bias and high frequency noise, where the cutoff frequency ω c1 and ω c2 The setting is 1.5ω r and 0.5ω r ,ω r is the speed. The angular frequency of the motor current ω s =p·ω r , p is the number of motor pole pairs. When the medium frequency ω c1 and ω c2 When the center frequency of the two-degree-of-freedom filter is set to be equal to the angular frequency ω, a large amount of DC bias and high-frequency noise can be filtered out, and the sampled fundamental current is not affected.

[0073] The super spiral sliding mode observer is constructed to identify the observed value of the current and the lumped disturbance parameter observations .

[0074] Among them, the stator current vector i is established ab The current prediction equation for the state variable is:

[0075] ;

[0076] Where i αβ is the αβ axis component of the current, F ab Represents the αβ axis component of the lumped disturbance parameter, α is a non-physical parameter, u αβ Represents the αβ-axis components of the voltage vector.

[0077] Secondly, a super-helical sliding mode observer including equivalent feedback gain and exponential term is constructed. The mathematical model is:

[0078] ;

[0079] Among them, the superscript “^” indicates the observed value of the corresponding variable; , The αβ axis current observation value represented by Relative to the actual value i αβ Error; represents the observed value of the equivalent feedback parameter of the αβ axis; Represents the αβ axis component of the lumped disturbance parameter observation value; k1, k2, k3 and k4 are all gain parameters and are positive numbers; sgn() is the sign function, is an exponential term; α is a non-physical parameter, which is based on the motor model according to The voltage vector u is derived formally αβ The value of α is related to the motor parameters, and the specific value is related to the motor model used. For example, in some embodiments, the permanent magnet motor model considering the disturbance is derived as follows: ,in, represents the stator inductance, represents the stator resistance, represents the αβ axis back electromotive force, Represents the αβ axis disturbance, which includes modeling error, α parameter deviation and inverter nonlinear error; in this example, the permanent magnet motor model is converted into the observation equation The α parameter is the inverse of the nominal value of the stator inductance, and the lumped disturbance parameter is Observed values It is observed by the super spiral sliding mode observer provided by the present invention.

[0080] Combining the above formulas (2) and (3), the lumped disturbance parameter error can be obtained: for:

[0081] ;

[0082] in, , The αβ axis components representing the actual values ​​of the lumped disturbance parameters.

[0083] From formula (3), we can get:

[0084] ;

[0085] From (5), we can see that although Contains a lot of high-frequency sliding mode noise, but because the integral operation has the effect of eliminating high-frequency signals, the observed equivalent feedback signal It does not contain high-frequency sliding mode noise and effectively suppresses the interference of high-frequency noise. Therefore, the observed lumped disturbance parameter value does not need to add an additional low-pass filter, thereby avoiding the occurrence of phase delay. In addition, the above exponential term It also makes the dynamic convergence process of the observer smoother.

[0086] The super-helical sliding mode observer provided by the present invention, which includes an equivalent feedback gain and an exponential term, is a brand-new observer, and its stability still needs to be proven.

[0087] Define the Lyapunov function V1 to satisfy:

[0088] ;

[0089] Where Δi α , Δi β is the error Δi αβ The α and β axis components.

[0090] From (5) and (6), we can get:

[0091] ;

[0092] Where, ΔF α , ΔF β is the lumped disturbance parameter error ΔF αβ The α-axis and β-axis components, for The α and β axis components.

[0093] In order to ensure the convergence of the super-helical sliding mode observer, dV1 / dt<0 must be satisfied. Therefore, it can be concluded that:

[0094] ;

[0095] Where sup{·} is a supremum function.

[0096] Ignoring the change in speed, according to formula (3), we can get:

[0097] ;

[0098] Where j is an imaginary operator.

[0099] According to (3) and (9), we can conclude that:

[0100] ;

[0101] Where, , is the error of the equivalent feedback signal, Δm α , Δm β is the error of the equivalent feedback signal The α and β axis components, is the actual value of the equivalent feedback signal.

[0102] Define the Lyapunov function V2 to satisfy:

[0103] ;

[0104] From (10) and (11), we can get:

[0105] ;

[0106] In order to ensure the convergence of the super-helical sliding mode observer, dV2 / dt<0 must be satisfied. Therefore, it can be concluded that:

[0107] ;

[0108] (8) and (13) provide a design method for sliding mode gain. When selecting sliding mode gain in practical applications, we can first select smaller values ​​of k3 and k4 and larger values ​​of k1 and k2 to ensure the convergence of the super-helical sliding mode observer, and then adjust it according to the actual observation effect.

[0109] The technical solution of the present invention can improve the equivalent feedback by reasonably designing the parameter k2 when the motor load is small. The value of , which helps to improve the observation accuracy of the lumped disturbance parameters under light load / no load, by introducing the exponential term , which can make the observer convergence process smoother, thus helping to improve the observation accuracy of lumped disturbance parameters under dynamic conditions.

[0110] In summary, when the lumped disturbance parameter observer is designed as a super-helical sliding mode observer in this embodiment, the high-frequency noise can be effectively eliminated by means of the integral form of equivalent feedback, and the introduction of the exponential term can make the dynamic convergence process smoother.

[0111] The following will illustrate with specific examples.

[0112] Example 1:

[0113] like Figure 2 As shown, an embodiment of the present invention provides a method for observing lumped disturbance parameters of a permanent magnet synchronous motor, comprising the following steps:

[0114] A super-helical sliding mode observer including equivalent feedback gain and exponential term is constructed;

[0115] Introducing the Lyapunov function to verify the stability of the super-helical sliding mode observer, and setting the gain parameters k1, k2, k3 and k4 in the super-helical sliding mode observer;

[0116] In each control cycle, the αβ axis current i of the permanent magnet synchronous motor is obtained αβ Then, the observed value of the current is identified based on the super spiral sliding mode observer. and the lumped disturbance parameter observations , for model-free current prediction of permanent magnet synchronous motors.

[0117] It should be understood that the mathematical model of the super spiral sliding mode observer is described with reference to the above method and will not be repeated here. The lumped disturbance parameter observation method based on the super spiral sliding mode observer provided in this embodiment uses the defined lumped disturbance parameter as the observation quantity, and its equivalent feedback gain is It has an integral form and can effectively suppress high-frequency noise; in addition, compared with the traditional super-helical sliding mode observer, the additional exponential term It can make the dynamic convergence process of the observer smoother.

[0118] Example 2:

[0119] like Figure 2 as well as Figure 3 As shown, an embodiment of the present invention provides a permanent magnet synchronous motor control method, comprising the following steps:

[0120] T1. Get the αβ axis current i of the permanent magnet synchronous motor in the current control cycle αβ (t) and speed ω r (t) and voltage vector u αβ (t), the current i is filtered by the two-degree-of-freedom filter. αβ DC bias and high-frequency noise signals in (t).

[0121] In this embodiment, the αβ axis current i of the permanent magnet synchronous motor in the current control cycle is αβ (t) and speed ω r (t) can be obtained by:

[0122] Collect the phase current i of the permanent magnet synchronous motor in the current control cycle abc , and convert it to the αβ axis through the inverse Park transform to obtain the αβ axis stator current i αβ (t);

[0123] Collect the position signal θ(t) of the permanent magnet synchronous motor in the current control cycle and Calculate the speed ω r (t), For the control cycle.

[0124] It is easy to understand that the phase current signal i of the permanent magnet synchronous motor can be easily collected by using the current sensor and speed sensor. abc and position signal θ(t); voltage vector u of the current control cycleαβ (t) is determined by the previous control cycle.

[0125] T2, using the current filtered in step T1 and the super-helical sliding mode observation method to identify the current observation value and the lumped disturbance parameter observations .

[0126] T3, given d-axis current reference value i dref =0; for speed ω r and the speed reference value ω ref The difference between the two is used for PI control to obtain the q-axis current reference value i qref .

[0127] T4, set the d-axis current reference value i dref and q-axis current reference value i qref Convert to αβ axis and get αβ axis stator current reference value i αβref .

[0128] T5. For each optional voltage vector, compare it with the αβ axis stator current i αβ (t), speed ω r (t) and the observed value of the lumped disturbance parameter Substitute the model-free current prediction formula of the permanent magnet synchronous motor into the voltage vector to obtain the predicted αβ axis current i in the next control cycle. αβ (t+1), and calculate the αβ axis current reference value i αβref The degree of difference between them is used as the cost function.

[0129] Among them, the mathematical model expression of the permanent magnet synchronous motor is as follows:

[0130] (14);

[0131] Among them, u αβ ,i αβ (t), They represent the output voltage vector, current vector and observed lumped disturbance parameter vector of the inverter respectively; T s Indicates the control period.

[0132] T6. Select the voltage vector with the smallest cost function as the optimal voltage vector u αβ (t+1), and in the next control cycle, the corresponding inverter switching signal is generated and then applied to the inverter of the permanent magnet synchronous motor for drive control.

[0133] In practical applications, the three-phase windings of the permanent magnet synchronous motor are connected to the midpoints of the three-phase bridge arms of the inverter. In the inverter, the upper and lower ends of each phase bridge arm are respectively provided with a switch tube, Sa 、S b 、S c Respectively represent the drive signals of the upper switch tubes of the bridge arms connected to the A, B, and C phase windings of the permanent magnet synchronous motor. 1 represents a high level and 0 represents a low level. It is easy to understand that a high level is a level that can turn on the upper switch tube of the bridge arm and turn off the lower switch tube, while a low level is a level that can turn off the upper switch tube of the bridge arm and turn on the lower switch tube. Based on the driving conditions of the switch tubes, there are a total of 8 voltage vectors, each of which is shown in Table 1. Indicates the linear bus voltage, that is, the DC power supply voltage on the inverter input side.

[0134] Table 1 Voltage vector relationship table

[0135] .

[0136] Each voltage vector together with the αβ axis stator current i αβ (t), speed ω r (t) Lumped disturbance parameter observation value After substituting into the mathematical model of the permanent magnet synchronous motor, the αβ axis stator current i in the next control cycle can be obtained. αβ (t+1) is used to accurately screen the optimal voltage vector. In this embodiment, the cost function selected is expressed as follows:

[0137] ;

[0138] Among them, i αref and i βref Represent the α-axis component and β-axis component of the α-axis and β-axis stator current reference values ​​respectively; i α (t+1) and i β (t+1) represents the predicted current i αβ The α-axis component and the β-axis component of (t+1); g represents the cost function value.

[0139] In summary, this embodiment addresses the problems of poor parameter robustness and susceptibility of control performance to changes in flux parameters in the application of traditional model predictive current control to permanent magnet synchronous motors. The observed lumped disturbance parameters are fed back into the model-free predictive current control algorithm in real time, greatly improving the parameter robustness of the control system.

[0140] Example 3:

[0141] A model-free current prediction control system for a permanent magnet synchronous motor comprises at least a feedback module, a lumped disturbance parameter observation module, a current prediction module, an optimal vector selection module and a control module.

[0142] Among them, the feedback module is used to obtain the αβ axis stator current i of the permanent magnet synchronous motor in the current control cycle αβ (t) and speed ω r (t), and the voltage vector u αβ (t); The sampling current filter module has an input end for receiving the αβ axis current i of the permanent magnet synchronous motor αβ , and is used to filter out the DC bias and high-frequency noise in the current sampling signal according to the designed two-degree-of-freedom filter. The lumped disturbance parameter observation module, whose first input terminal is used to receive the voltage vector u of the permanent magnet synchronous motor drive inverter αβ The second input terminal is used to receive the αβ axis current i of the permanent magnet synchronous motor. αβ , and then calculate the observed value of the current according to the designed super spiral sliding mode observer and the lumped disturbance parameter observations ; Current prediction module, used to use the αβ axis current i αβ , voltage vector u αβ and the lumped disturbance parameter observations Substitute the model-free current prediction model of the permanent magnet synchronous motor to predict the αβ axis current prediction value in the next control cycle; the optimal vector selection module is used to select the voltage vector with the smallest cost function as the optimal voltage vector u αβ (t+1); The control module is used to generate an inverter switching signal based on the optimal voltage vector and apply the inverter switching signal to the inverter for controlling the permanent magnet synchronous motor.

[0143] In this embodiment, for each voltage vector, the cost function is obtained by: comparing the voltage vector with the primary current i αβ (t), and the observed value of the lumped disturbance parameter Substitute the model-free current prediction formula of the permanent magnet linear synchronous motor into the voltage vector to obtain the αβ axis primary current i in the next control cycle corresponding to the voltage vector. αβ (t+1), and calculate the primary current reference value i of the αβ axis αβref The degree of difference between them is used as the cost function.

[0144] In some other embodiments, the control system further comprises:

[0145] D-axis current control module, sets the d-axis current reference value i dref =0;

[0146] The q-axis current control module is used to control the speed v and the speed reference value v ref The difference between the two is used for PI control to obtain the q-axis current reference value i qref ;

[0147] The coordinate transformation module is used to transform the d-axis current reference value idref and q-axis current reference value i qref Convert to αβ axis and get the αβ axis primary current reference value i αβref .

[0148] It should be understood that the functions and implementation processes of the above-mentioned d-axis current control module, q-axis current control module and coordinate transformation module are all existing technologies.

[0149] It should also be understood that the specific implementation process of each module please refer to the above method content, the present invention will not go into details here, and the division of the above functional modules is only for example illustration. In some embodiments, some functional modules can be merged, and some functional modules can be split. Each functional module can be implemented in software or hardware or a combination of software and hardware. Among them, software and hardware equipment include but are not limited to general-purpose computer equipment, programmable gate arrays, digital signal processors, microprocessors and their corresponding programming or burning software.

[0150] Example 4:

[0151] A permanent magnet synchronous motor system includes a permanent magnet synchronous motor, an inverter, and a control system. The control system is connected to the permanent magnet synchronous motor and the inverter, respectively. The midpoints of the three-phase bridge arms of the inverter are connected to the three-phase windings of the permanent magnet synchronous motor.

[0152] Example 5:

[0153] This embodiment provides a readable storage medium storing a control program, wherein the control program is called by a processor to implement:

[0154] Get the αβ axis current i of the permanent magnet synchronous motor in the current control cycle αβ and speed ω r and the voltage vector u αβ ;

[0155] Based on the αβ axis current i αβ and voltage vector u αβ , a super-helical sliding mode observer including an equivalent feedback gain and an exponential term is introduced, and the observed value of the current is obtained using the super-helical sliding mode observer and the lumped disturbance parameter observations ;

[0156] Using the αβ axis current i αβ , voltage vector u αβ and the lumped disturbance parameter observations The model-free current prediction model of the permanent magnet synchronous motor is substituted into the predicted αβ axis current value in the next control cycle, which is used for optimal voltage vector screening. Then, the inverter switching signal is generated based on the optimal voltage vector and applied to the inverter of the permanent magnet synchronous motor to realize drive control.

[0157] For the specific implementation process of each step, please refer to the description of the above method.

[0158] The readable storage medium is a computer-readable storage medium, which can be an internal storage unit of the software and hardware device described in any of the aforementioned embodiments, such as a hard disk or memory of a controller. The readable storage medium can also be an external storage device of the controller, such as a plug-in hard disk equipped on the controller, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. Furthermore, the readable storage medium can also include both an internal storage unit of the controller and an external storage device. The readable storage medium is used to store the computer program and other programs and data required by the controller. The readable storage medium can also be used to temporarily store data that has been output or is to be output.

[0159] Based on this understanding, the technical solution of the present invention, or the portion that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes instructions for causing a computer device (such as a personal computer, server, or network device) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned readable storage medium includes various media capable of storing program code, such as a USB flash drive, a mobile hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0160] It will be easily understood by those skilled in the art that the above are merely preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A permanent magnet synchronous motor control method, characterized in that: include: S1. Get the αβ axis current i of the permanent magnet synchronous motor in the current control cycle αβ and speed ω r and the voltage vector u αβ ; S2, based on αβ axis current i αβ and voltage vector u αβ , a super-helical sliding mode observer including an equivalent feedback gain and an exponential term is introduced, and the observed value of the current is obtained using the super-helical sliding mode observer and the lumped disturbance parameter observations ; S3, using αβ axis current i αβ , voltage vector u αβ and the lumped disturbance parameter observations The model-free current prediction model of the permanent magnet synchronous motor is substituted into the predicted αβ axis current value in the next control cycle, which is used to screen the optimal voltage vector. Then, the inverter switching signal is generated based on the optimal voltage vector and applied to the inverter of the permanent magnet synchronous motor for drive control. The mathematical model of the super spiral sliding mode observer including the equivalent feedback gain and the exponential term is: ； Where α is a non-physical parameter, namely the voltage vector u αβ The coefficient of represents the observed value of the equivalent feedback parameter of the αβ axis, , The αβ axis current observation value represented by Relative to the αβ axis current sampling value i αβ The error; sgn() is the sign function; k1, k3 and k4 are gain parameters, t is time; Among them, the model-free current prediction formula of permanent magnet synchronous motor is as follows: ； in, , , Respectively represent the output voltage vector of the inverter, the control cycle t The corresponding current vector and the observed lumped disturbance parameter vector; T s Indicates the control period; The cost function for optimal voltage vector screening is expressed as follows: ； Among them, i αref and i βref Represent the α-axis component and β-axis component of the α-axis and β-axis stator current reference values ​​respectively; i α (t+1) and i β (t+1) represents the α-axis component and the β-axis component of the predicted current in the next control cycle t+1 respectively; g represents the cost function.

2. The permanent magnet synchronous motor control method according to claim 1, characterized in that: Observed values ​​of equivalent feedback parameters of the αβ axis satisfy: ； Where k2 is the gain parameter.

3. The permanent magnet synchronous motor control method according to claim 2, characterized in that: Gain coefficients k1 and k2 satisfy: ； ； Where Δi α , Δi β is the error Δi αβ The α-axis and β-axis components; for α and β axis components; Δm α , Δm β is the error Δm of the equivalent feedback signal αβ The α and β axis components, , is the actual value of the equivalent feedback signal, sup{·} is the supremum function; , m is the equivalent feedback parameter αβ The α-axis component and the β-axis component, ω r is the rotation speed.

4. The permanent magnet synchronous motor control method according to claim 1, wherein: Step S1 obtains the αβ axis current i of the permanent magnet synchronous motor in the current control cycle αβ After that, a two-degree-of-freedom filter is introduced to the current i αβ Perform pretreatment; The two-degree-of-freedom filter includes a first low-pass filter and a second low-pass filter connected in parallel for subtraction, and the output signals of the first low-pass filter and the second low-pass filter are subtracted to obtain the output signal of the two-degree-of-freedom filter. The transfer function is expressed as: ； Where, G, G LPF1 and G LPF2 Represent the transfer functions of the two-degree-of-freedom filter, the first low-pass filter, and the second low-pass filter respectively; ω c1 and ω c2 They represent the cutoff frequencies of the first and second low-pass filters respectively; s represents the differential operator.

5. The permanent magnet synchronous motor control method according to claim 4, characterized in that: Cutoff frequency ω c1 and ω c2 The setting is 1.5ω r and 0.5ω r ,ω r is the rotation speed.

6. A control system based on the permanent magnet synchronous motor control method according to any one of claims 1 to 5, characterized in that: include: Feedback module, used to obtain the αβ axis current i of the permanent magnet synchronous motor in the current control cycle αβ and speed ω r and the voltage vector u αβ ; Lumped disturbance parameter observation module, based on αβ axis current i αβ and voltage vector u αβ , a super-helical sliding mode observer including an equivalent feedback gain and an exponential term is introduced, and the observed value of the current is obtained using the super-helical sliding mode observer and the lumped disturbance parameter observations ; Current prediction module, used to use the αβ axis current i αβ , voltage vector u αβ and the lumped disturbance parameter observations Substitute the model-free current prediction model of the permanent magnet synchronous motor to predict the αβ axis current prediction value in the next control cycle; Optimal vector selection module, used for optimal voltage vector screening; The control module is used to generate an inverter switching signal based on the optimal voltage vector and apply it to the inverter of the permanent magnet synchronous motor to achieve drive control.

7. A permanent magnet synchronous motor system, comprising: Permanent magnet synchronous motor; The inverter has its three-phase bridge arm midpoints connected to the three-phase windings of the permanent magnet synchronous motor respectively; and the control system according to claim 6; The control system is connected to the permanent magnet synchronous motor and the inverter respectively.

8. A lumped disturbance parameter observation method applicable to a permanent magnet synchronous motor, characterized by: The following steps are involved: A super-helical sliding mode observer including equivalent feedback gain and exponential term is constructed; Introducing the Lyapunov function to verify the stability of the super-helical sliding mode observer, and setting the gain parameters k1, k2, k3 and k4 in the super-helical sliding mode observer; In each control cycle, the steps S1-S2 in the permanent magnet synchronous motor control method according to claim 1 are implemented as follows: obtaining the αβ axis current i of the permanent magnet synchronous motor; αβ Then, the observed value of the current is identified based on the super spiral sliding mode observer. and the lumped disturbance parameter observations , which is then used for model-free current prediction of the permanent magnet synchronous motor in step S3.

9. A readable storage medium, characterized in that: A control program is stored, which is called by the processor to implement: Get the αβ axis current i of the permanent magnet synchronous motor in the current control cycle αβ and speed ω r and the voltage vector u αβ ; Based on the αβ axis current i αβ and voltage vector u αβ , a super-helical sliding mode observer including an equivalent feedback gain and an exponential term is introduced, and the observed value of the current is obtained using the super-helical sliding mode observer and the lumped disturbance parameter observations ; Using the αβ axis current i αβ , voltage vector u αβ and the lumped disturbance parameter observations Substitute the model-free current prediction model of the permanent magnet synchronous motor into the predicted αβ axis current value in the next control cycle, which is used to screen the optimal voltage vector. Then, based on the optimal voltage vector, the inverter switching signal is generated and applied to the inverter of the permanent magnet synchronous motor to realize drive control. The mathematical model of the super-spiral sliding mode observer including the equivalent feedback gain and the exponential term is: ； Where α is a non-physical parameter, namely the voltage vector u αβ The coefficient of represents the observed value of the equivalent feedback parameter of the αβ axis, , The αβ axis current observation value represented by Relative to the αβ axis current sampling value i αβ The error; sgn() is the sign function; k1, k3 and k4 are gain parameters, t is time; Among them, the model-free current prediction formula of permanent magnet synchronous motor is as follows: ； in, , , Respectively represent the output voltage vector of the inverter, the control cycle t The corresponding current vector and the observed lumped disturbance parameter vector; T s Indicates the control cycle; The cost function for optimal voltage vector screening is expressed as follows: ； Among them, i αref and i βref Represent the α-axis component and β-axis component of the α-axis and β-axis stator current reference values ​​respectively; i α (t+1) and i β (t+1) represents the α-axis component and the β-axis component of the predicted current in the next control cycle t+1 respectively; g represents the cost function.

Images