Model-free predictive control method and system for permanent magnet synchronous motor based on current differential update

Abstract

The present invention discloses a method and system for model-free predictive control of a permanent magnet synchronous motor using current differential update. The method includes: estimating the inductance to obtain dq Axis estimated inductance; for any voltage vector among the eight basic voltage vectors, combined with the current difference at time k and dq The axis estimated inductance performs current differential without stagnation update to calculate the voltage vector corresponding to dq Axis current difference; the current difference value function is applied to each voltage vector corresponding to dq The shaft current is differentially calculated, and the voltage vector with the smallest result is selected as the optimal voltage vector for compensation and then used as the control signal of the inverter. The present invention aims to achieve high-performance motor control, get rid of the dependence on parameters, improve the robustness of the system, solve the problem of current differential update stagnation, ensure the accuracy of predictive control, reduce the complexity of calculation, and improve the operating performance of permanent magnet synchronous motors.

Description

Technical Field

[0001] The present invention relates to the technical field of permanent magnet synchronous motor control, and in particular to a current differential updated permanent magnet synchronous motor model-free predictive control method and system. Background Art

[0002] In recent years, permanent magnet synchronous motors have attracted attention due to their significant advantages such as high power density and high efficiency. They have been widely used in new energy vehicles, aerospace, wind power generation and other fields. In order to achieve high-performance control of permanent magnet motors, field-oriented control (FOC) and direct torque control (DTC) are two commonly used control strategies, but FOC requires the design of a proportional-integral controller and the adjustment of the controller gain. The controller gain depends on the parameters of the model, while DTC has large current and torque ripples and poor steady-state performance. Therefore, model predictive current control (MPCC) has been gradually applied in the field of permanent magnet synchronous motor control due to its advantages such as fast dynamic response, simple structure and good steady-state performance. However, traditional MPCC is a model-based control strategy that relies on accurate system models and parameters. During the operation of permanent magnet synchronous motors, the parameters are not constant due to temperature, magnetic saturation and other reasons, and deviate from the nominal values. Therefore, the performance of MPCC will inevitably deteriorate due to parameter mismatch. In order to improve the robustness of MPCC, model free perdicitive current control (model free perdicitive current control, MFPCC was proposed to overcome the dependence of MPCC on parameters. Traditional MFPCC stores the current differentials related to eight basic voltage vectors in a lookup table (LUT). Current prediction only needs to add the current current to the corresponding current differential of the LUT. When a voltage vector is used, its corresponding current difference will be updated in the lookup table. This scheme is very simple to predict current and does not require any motor parameters, only current sampling. However, MFPCC has a major problem: stagnant current difference updating (SCDU). Only the current differential caused by the applied voltage vector can be updated within one control cycle. If a voltage vector is not adopted after several consecutive control cycles, the corresponding element in the LUT will stagnate and lose prediction accuracy. The SCDU will deteriorate the control performance and even cause system instability. Summary of the invention

[0003] Technical problem to be solved by the present invention: In view of the above-mentioned problems in the prior art, a model-free predictive control method and system for a permanent magnet synchronous motor with current differential update are provided. The present invention aims to achieve high-performance control of the permanent magnet synchronous motor, get rid of the dependence on parameters, improve the robustness of the system, solve the problem of deterioration of control performance caused by stagnation of current differential update in the prior art, ensure the accuracy of predictive control, reduce computational complexity, and improve the operating performance of the permanent magnet synchronous motor.

[0004] In order to solve the above technical problems, the technical solution adopted by the present invention is:

[0005] A model-free predictive control method for a permanent magnet synchronous motor with current differential update, comprising:

[0006] S101, the inductance is estimated to obtain dq Estimated shaft inductance and ;

[0007] S102, for eight basic voltage vectors ~ Any of Voltage vector , combined with the current difference at time k as well as dq Estimated shaft inductance and Calculate the voltage vector by performing current differential update without lag Corresponding dq Shaft current differential ;

[0008] S103, using a preset current differential value function and applying it to each voltage vector Corresponding dq Shaft current differential , and select the voltage vector with the minimum result of the current differential value function calculation As the optimal voltage vector ;

[0009] S104, the optimal voltage vector The inverter nonlinearity compensation is performed, and then the voltage signal obtained after compensation is used as the inverter control signal to control the inverter of the permanent magnet synchronous motor.

[0010] Optionally, in step S101, inductance estimation is performed to obtain dq Estimated shaft inductance and The function expression is:

[0011] , ,

[0012] In the above formula, To control the cycle, and They are dq The estimated inductance of the axis is the middle amount, and when the zero voltage vector is applied first and the active voltage vector is applied later:

[0013] , ,

[0014] And when the zero voltage vector is applied later and the active voltage vector is applied first, there are:

[0015] , ,

[0016] in, and The active voltage vector is applied at time k Produced dq Shaft current differential, and are k moments respectively dq Shaft current change, is the active voltage vector, and The zero voltage vector is applied at time k Produced dq Shaft current differential, and The zero voltage vector is applied at time k-1 respectively Produced dq Shaft current differential, and The k-1 moments dq Shaft current change, and is the active voltage vector The voltage on the dq axis, is the zero voltage vector.

[0017] Optionally, before step S101, the step further includes zero vector current differential reconstruction to obtain zero voltage vectors applied at time k. Produced dq Shaft current differential and : Get the voltage vector selected in the basic voltage vector for the first two cycles and , and satisfies , if the voltage vector and is the zero voltage vector, then the zero voltage vector applied at time k is calculated according to the following formula Produced dq Shaft current differential and :

[0018] or ,

[0019] In the above formula, Apply zero voltage vector at time k Produced dq Shaft current differential and , Apply zero voltage vector for k-1 time Produced dq Shaft current differential and ; If the voltage vector and is a zero voltage vector; if the voltage vector and For two different active voltage vectors, the zero voltage vector applied at time k is calculated according to the following formula: Produced dq Shaft current differential and :

[0020] ,

[0021] ,

[0022] In the above formula, and The voltage vector applied at time k is Produced dq Shaft current differential, and The voltage vector applied at time k-1 is Produced dq Shaft current differential.

[0023] Optionally, in step S102, the current difference at time k is combined as well as dq Estimated shaft inductance and Calculate the voltage vector by performing current differential update without lag Corresponding dq Shaft current differential The function expression is:

[0024] - + ,

[0025] - + ,

[0026] ,

[0027] ,

[0028] In the above formula, represents the voltage vector applied at time k, represents the unapplied voltage vector at time k, Not equal to and Respectively represent the k-th moment obtained by updating The corresponding d-axis and q-axis current differentials are: and represents the voltage vector applied at time k The resulting current difference, and They represent the dq axis current sampled at time k, and They represent the dq axis current sampled at time k-1 respectively, and They are dq Axis estimated inductance mid-axis.

[0029] Optionally, the function expression of the current difference type cost function preset in step S103 is:

[0030] ,

[0031] In the above formula, represents the current differential value function, and Represent the ideal current difference of d and q axes respectively, , , and Represent the d-axis and q-axis current reference values ​​respectively, and The voltage vector is calculated by continuously updating the current difference Corresponding dq Shaft current differential The d and q axis components in .

[0032] Optionally, in step S104, the optimal voltage vector Inverter nonlinear compensation refers to adding compensation time on the basis of dead time to compensate for voltage disturbance caused by inverter nonlinearity.

[0033] Optionally, the calculation function expression of the compensation time is:

[0034] ,

[0035] In the above formula, Indicates the compensation time, is the threshold voltage of the IGBT in the inverter, is the threshold voltage of the inverter's freewheeling diode, To control the cycle, is the DC bus voltage, is the inverter dead time delay, and It is the turn-on and turn-off delay time of IGBT in the inverter.

[0036] In addition, the present invention also provides a permanent magnet synchronous motor model-free predictive control system with current differential update, comprising a microprocessor and a memory connected to each other, wherein the microprocessor is programmed or configured to execute the permanent magnet synchronous motor model-free predictive control method with current differential update.

[0037] In addition, the present invention also provides a computer-readable storage medium, which stores a computer program / instruction, and the computer program / instruction is programmed or configured to execute the permanent magnet synchronous motor model-free predictive control method with current differential update through a processor.

[0038] In addition, the present invention also provides a computer program product, including a computer program / instruction, which is programmed or configured to execute the permanent magnet synchronous motor model-free predictive control method with current differential update through a processor.

[0039] Compared with the prior art, the present invention mainly has the following advantages: the present invention solves the current differential update stagnation problem of the traditional model-free predictive control, realizes the current differential non-stagnation update, ensures the accuracy of current prediction, does not need to design controller gain and observer, has a simple structure, low computational complexity, does not need to take forced update measures at the cost of performance, adopts a new value function, does not need to predict the current at the next moment, optimizes the current differential value function, reduces the amount of calculation, and compensates for the nonlinearity of the inverter, suppresses the disturbance caused by the nonlinearity of the inverter, and makes the actual voltage close to the ideal voltage. The present invention has good robustness, eliminates the influence of parameter mismatch on the system, compensates for the nonlinearity of the inverter, has a simple control structure, realizes high-performance control of the permanent magnet synchronous motor, and improves the operating performance of the permanent magnet synchronous motor. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 Schematic diagram of the basic flow of the method of the embodiment of the present invention.

[0041] Figure 2 Schematic diagram of the connection between a permanent magnet synchronous motor and an inverter in an embodiment of the present invention.

[0042] Figure 3 8 basic voltage vector distribution diagrams in the embodiments of the present invention.

[0043] Figure 4 Schematic diagram of the timing principle of model-free predictive control in an embodiment of the present invention.

[0044] Figure 5 Schematic diagram of comparison between the ideal voltage and the actual voltage of phase A in an embodiment of the present invention.

[0045] Figure 6 A schematic diagram of the voltage of adding inverter nonlinear compensation to phase A in an embodiment of the present invention. DETAILED DESCRIPTION

[0046] The present invention includes inductance information estimation, current differential non-stop updating, current differential reconstruction, current differential value function and inverter non-linear compensation. Inductance information estimation is calculated by current sampling. Current differential non-stop updating realizes current differential update corresponding to all voltage vectors through current sampling and inductance information estimation. Current differential reconstruction reconstructs zero vector current differential based on the mathematical relationship of voltage vector. A current differential value function is proposed, and the current prediction step in traditional model predictive control is eliminated. The value function can be minimized only based on the current differential, thereby obtaining the optimal voltage vector. Inverter non-linear compensation suppresses the disturbance caused by the nonlinearity of the inverter. The present invention will be further described in detail below in conjunction with the accompanying drawings.

[0047] like Figure 1 As shown, the model-free predictive control method for a permanent magnet synchronous motor with current differential update in this embodiment includes:

[0048] S101, the inductance is estimated to obtain dq Estimated shaft inductance and ;

[0049] S102, for eight basic voltage vectors ~ Any of Voltage vector , combined with the current difference at time k as well as dq Estimated shaft inductance and Calculate the voltage vector by performing current differential update without lag Corresponding dq Shaft current differential ;

[0050] S103, using a preset current differential value function and applying it to each voltage vector Corresponding dq Shaft current differential , and select the voltage vector with the minimum result of the current differential value function calculation As the optimal voltage vector ;

[0051] S104, the optimal voltage vector The inverter nonlinearity compensation is performed, and then the voltage signal obtained after compensation is used as the inverter control signal to control the inverter of the permanent magnet synchronous motor.

[0052] like Figure 2 As shown, the permanent magnet synchronous motor (PMSM) of this embodiment adopts an internal permanent magnet synchronous motor, and a three-phase six-switch voltage source inverter is connected to the internal permanent magnet synchronous motor. is the DC bus voltage, S a1 , S a2 , S b1 , S b2 , S c1 and S c2 A, B, and C correspond to the upper and lower bridge arm switches on the three phases. To avoid short circuits, the switches of the same phase cannot be turned on or off at the same time. Therefore, there are eight switching states for the inverter switches. The distribution of the eight basic voltage vectors in space is as follows: Figure 3 As shown, the eight basic voltage vectors include two zero vectors and , six active vectors (non-zero vectors) , , , , and .

[0053] In step S101 of this embodiment, the inductance is estimated to obtain dq Estimated shaft inductance and The function expression is:

[0054] , ,

[0055] In the above formula, To control the cycle, and They are dq The estimated inductance of the axis is the middle amount, and when the zero voltage vector is applied first and the active voltage vector is applied later:

[0056] , ,

[0057] And when the zero voltage vector is applied later and the active voltage vector is applied first, there are:

[0058] , ,

[0059] in, and The active voltage vector is applied at time k Produced dq Shaft current differential, and are k moments respectively dq Shaft current change, is the active voltage vector, and The zero voltage vector is applied at time k Produced dq Shaft current differential, and The zero voltage vector is applied at time k-1 respectively Produced dq Shaft current differential, and The k-1 moments dq Shaft current change, and is the active voltage vector The voltage on the dq axis, is the zero voltage vector.

[0060] In order to increase the update frequency of the inductance and avoid the long-term non-application of the zero vector, the zero vector current difference is reconstructed to obtain the zero voltage vector applied at time k. Produced dq Shaft current differential and Specifically, before step S101, this embodiment also includes zero vector current differential reconstruction to obtain zero voltage vectors applied at time k. Produced dq Shaft current differential and : Get the voltage vector selected in the basic voltage vector for the first two cycles and , and satisfies , if the voltage vector and is the zero voltage vector, then the zero voltage vector applied at time k is calculated according to the following formula Produced dq Shaft current differential and :

[0061] or ,

[0062] In the above formula, Apply zero voltage vector at time k Produced dq Shaft current differential and , Apply zero voltage vector for k-1 time Produced dq Shaft current differential and ; If the voltage vector and is a zero voltage vector; if the voltage vector and For two different active voltage vectors, the zero voltage vector applied at time k is calculated according to the following formula: Produced dq Shaft current differential and :

[0063] ,

[0064] ,

[0065] In the above formula, and The voltage vector applied at time k is Produced dq Shaft current differential, and The voltage vector applied at time k-1 is Produced dq Shaft current differential.

[0066] In step S102 of this embodiment, the current difference at time k is combined as well as dq Estimated shaft inductance and Calculate the voltage vector by performing current differential update without lag Corresponding dq Shaft current differential The function expression is:

[0067] - + ,

[0068] - + ,

[0069] ,

[0070] ,

[0071] In the above formula, represents the voltage vector applied at time k, represents the unapplied voltage vector at time k, Not equal to and Respectively represent the k-th moment obtained by updating The corresponding d-axis and q-axis current differentials are: and represents the voltage vector applied at time k Current differential, and They represent the dq axis current sampled at time k, and They represent the dq axis current sampled at time k-1 respectively, and They are dq Axis estimated inductance mid-axis.

[0072] The function expression of the current differential cost function preset in step S103 of this embodiment is:

[0073] ,

[0074] In the above formula, represents the current differential value function, and Represent the ideal current difference of d and q axes respectively, , , and Represent the d-axis and q-axis current reference values ​​respectively, and The voltage vector is calculated by continuously updating the current difference Corresponding dq Shaft current differential The d and q axis components in .

[0075] In step S104 of this embodiment, the optimal voltage vector Performing inverter nonlinear compensation means adding compensation time on the basis of dead time to compensate for voltage disturbance caused by inverter nonlinearity. As an optional implementation, the calculation function expression of compensation time in this embodiment is:

[0076] ,

[0077] In the above formula, Indicates the compensation time, is the threshold voltage of the IGBT in the inverter, is the threshold voltage of the inverter's freewheeling diode, To control the cycle, is the DC bus voltage, is the inverter dead time delay, and It is the turn-on and turn-off delay time of IGBT in the inverter.

[0078] The current prediction equation commonly used in traditional model predictive current control MPCC is:

[0079] ,

[0080] In the above formula, is the system control period, and are the currents on the d and q axes at time k, and are the voltages of the d and q axes at time k, and is the predicted current value at time k+1, is the electrical angular velocity at time k, is the stator resistance, and They are the d-axis and q-axis inductances, is the permanent magnet flux. The traditional model predictive current control MPCC uses the value function for:

[0081] ,

[0082] In the above formula, and Represent the d-axis and q-axis current reference values ​​respectively. In traditional model-free predictive control, the current prediction value can be expressed in the form of current difference:

[0083] ,

[0084] In the above formula, and is the voltage vector applied at time k The resulting dq axis current difference is yes Obviously, the future current differential and It is not directly available, but and The current difference at the past moment can be used instead, because the current difference caused by the same voltage vector in adjacent cycles can be considered the same at a high control frequency.

[0085] ,

[0086] In the above formula, and They are the past moments due to the applied voltage vector The resulting current difference.

[0087] Different from the current prediction in MPCC, the current prediction in traditional model-free predictive control is expressed as:

[0088] ,

[0089] In the above formula, and Represents the voltage vector due to the application The current prediction value at time k+1 is generated. This method does not require any motor parameters, only current sampling. This embodiment eliminates the current prediction step in traditional model-free predictive control and directly uses the current differential value function to determine the optimal voltage vector.

[0090] The basic process of model-free predictive control of the method of this embodiment is as follows: Figure 4 As shown, sampling is performed before the voltage vector is applied to avoid current spikes. After the voltage vector is applied, the corresponding current differential is updated and the optimization is performed through the value function. The above method relies entirely on current sampling and does not require motor parameters. At high sampling frequencies, since the mechanical time constant of the motor is much larger than the electrical time constant, the current differentials of adjacent cycles generated by the same voltage vector can be considered the same, but the time interval between two adjacent actions of the same voltage vector cannot be too long to maintain accuracy. In addition, the corresponding current differential is only updated when the voltage vector is applied. However, if the voltage vector is not used for several consecutive control cycles, a stagnant current differential update (SCDU) will occur, which means that and (The same situation occurs in and The error between the zero vector and the current differential update (SCDU) increases, and the stagnant current differential update (SCDU) will lead to inaccurate current prediction. Therefore, it is necessary to consider updating the current differential of the remaining voltage vector. In this embodiment, the current differential under the dq axis is divided into two parts, namely the natural attenuation caused by the zero vector and the forced response caused by the voltage vector. and , the zero vector causes the current to decay naturally and The expression is:

[0091] ,

[0092] ,

[0093] In the above formula and They are the natural attenuation of the current due to the zero vector at the d and q axes at time k. According to the above formula, the natural attenuation is only related to the current state of the motor; the current forced response caused by the applied voltage vector and , the expression of the current forced response caused by the applied voltage vector is:

[0094] ,

[0095] ,

[0096] In the above formula and The voltage vector due to the application at time k is The d and q axis currents are forced to respond, is a basic voltage vector, and yes The voltage on the d and q axes is only related to the applied voltage vector according to the above current forced response. Therefore, the dq axis current differential can be divided into two parts: one part is only related to the current state of the motor (current current, speed, electrical angle), and the other part is only related to the applied voltage vector.

[0097] Since the applied voltage vector The resulting current difference and The expression is:

[0098] = ,

[0099] = ,

[0100] The identified inductance information of this embodiment is obtained by a set of adjacent zero vector current differences and active vector current differences. The calculation of the inductance information includes:

[0101] Since the mechanical time constant is much larger than the electrical time constant, under high control frequency, the motor speed, electrical angle, and current information are approximately equal in a short period of time. Therefore, the natural decay part of the current difference in adjacent cycles can be considered the same. The active vector current difference is subtracted from the zero vector current difference to obtain the expression:

[0102] ,

[0103] ,

[0104] - = ,

[0105] - = ,

[0106] In the above formula and The voltage vector due to the application at time k is The resulting d-axis and q-axis current differential is represents the active voltage vector, and They are respectively due to the applied voltage vector at time k-1 The resulting d-axis and q-axis current differential is represents the zero voltage vector;

[0107] By using active vector current differential and zero vector current differential, the inductance information and The estimated expression is:

[0108] , ,

[0109] , ,

[0110] In the above formula and is the estimated intermediate quantity of inductance, and are the estimated values ​​of the d-axis and q-axis inductances, respectively. In addition, the above is the scenario where the zero voltage vector is applied first and the active voltage vector is applied later. If the application order is reversed, the above principle still holds true, only the positive and negative signs are changed. The specific estimation formula is:

[0111] , ,

[0112] right and Perform filtering to filter out noise and interference.

[0113] The current differential is updated without stalling by and And the current current differential realizes all unapplied voltage vector current differential updates. When the current differential is updated without stagnation, combined with the current differential at time k as well as dq Estimated shaft inductance and Calculate the voltage vector by performing current differential update without lag Corresponding dq Shaft current differential The function expression is:

[0114] - + ,

[0115] - + ,

[0116] ,

[0117] ,

[0118] In the above formula, represents the voltage vector applied at time k, represents the unapplied voltage vector at time k, Not equal to and Respectively represent the k-th moment obtained by updating The corresponding d-axis and q-axis current differentials are: and represents the voltage vector applied at time k Current differential, and They represent the dq axis current sampled at time k, and They represent the dq axis current sampled at time k-1 respectively, and They are dq Axis estimated inductance mid-axis.

[0119] Repeating the steps in each cycle realizes the non-stop update of the current differential, solving the stagnation problem of the current differential update. At the same time, if the zero vector and active vector are applied in adjacent cycles, the steps are performed. and to update.

[0120] The current differential reconstruction of this embodiment reconstructs the zero vector current differential to avoid the long-term non-application of the zero vector. and Update stagnation phenomenon, current differential reconstruction includes:

[0121] Through the mathematical relationship of the voltage vector, there are three groups of current differentials corresponding to the voltage vectors that can realize the zero vector current differential reconstruction. The three groups are: Group 1: and ; Group 1: and ; Group 3: and ;

[0122] Define a set of voltage vector sequences , including the voltage vectors applied in the first two cycles, the zero vector current differential reconstruction follows the following three conditions:

[0123] Condition 1: are two different active vectors;

[0124] Condition 2: ;

[0125] Condition 3: The expression of the current zero vector current difference is:

[0126] ,

[0127] ,

[0128] The above formula reconstructs the zero vector current difference through the voltage vector corresponding to the current that meets conditions 1 and 2, so the result can be used to update the inductance information.

[0129] This embodiment proposes a current differential value function, which eliminates the current prediction step in traditional predictive control and can determine the optimal voltage vector based on the current current differential. , reducing the amount of calculation, the proposed current differential value function expression is:

[0130] ,

[0131] In the above formula, represents the current differential value function, and Represent the ideal current difference of d and q axes respectively, , , and Represent the d-axis and q-axis current reference values ​​respectively, and The voltage vector is calculated by continuously updating the current difference Corresponding dq Shaft current differential The d and q axis components in .

[0132] Taking into account the nonlinearity of the inverter, the actual dq axis mathematical model of the permanent magnet synchronous motor can be expressed as:

[0133] ,

[0134] In the above formula, the dead zone voltage is and the threshold voltage of the IGBT ( ) and the threshold voltage of the freewheeling diode ( )、On / Off Delay Time(( and ) and dead time delay ( )related, and It is related to the current polarity and electrical angle and is expressed as:

[0135] ,

[0136] In the above formula, , ,and is the three-phase current of A, B, and C, is a symbolic function, and its function expression is:

[0137] ,

[0138] Inverter nonlinearity will cause additional errors in current prediction. Considering the nonlinearity of the inverter, it can be rewritten as:

[0139] ,

[0140] ,

[0141] In the above formula and It is the current differential response caused by the nonlinearity of the inverter.

[0142] At high control frequencies, the current and electrical angle of adjacent cycles can be considered constant, so and It can also be considered to remain unchanged. The offset caused by the inverter nonlinearity will be offset by subtraction. The final estimation result has nothing to do with the inverter nonlinearity. At the same time, the recorded current difference and This includes the impact of the inverter nonlinearity. Figure 5 As shown in the figure, taking phase A as an example, it shows the ideal voltage and the actual voltage due to factors such as dead time, where V a is the phase A voltage, V dc DC bus voltage, Sa1 is the A phase upper bridge arm drive signal, Sa2 is the A phase lower bridge arm drive signal, ia is the a-phase current, t D is the total error time. The effective conduction time of phase A The actual value is:

[0143] ,

[0144] Above is the ideal conduction time of phase A, Figure 5 The example is , is the control cycle. The same analysis is performed on phase B and phase C to obtain:

[0145] ,

[0146] Actual phase voltage of phases A, B and C , , It is expressed as:

[0147] ,

[0148] Therefore, the actual A, B, and C phase neutral point voltages , , for:

[0149] ,

[0150] The ideal neutral point voltage of phases A, B, and C is , , It is expressed as:

[0151] ,

[0152] The error between the actual value and the ideal value , , for:

[0153] ,

[0154] like Figure 6 The figure shows the compensated A-phase voltage schematic diagram, adding the dead time Used to compensate for the nonlinearity of the inverter, When it is 0, it means that the voltage disturbance caused by the nonlinearity of the inverter is suppressed, which can be calculated as for:

[0155] ,

[0156] The above steps realize nonlinear compensation for the inverter, make the actual voltage solve the ideal voltage, and optimize the motor control performance.

[0157] In summary, the current differential updated permanent magnet synchronous motor model-free predictive control method of this embodiment includes inductance information estimation, current differential non-stagnation update, current differential reconstruction, current differential value function and inverter nonlinear compensation. The inductance information estimation is calculated by current sampling. The current differential non-stagnation update realizes the current differential update corresponding to all voltage vectors through current sampling and inductance information estimation. The current differential reconstruction reconstructs the zero vector current differential based on the mathematical relationship of the voltage vector, proposes a current differential value function, cancels the current prediction step in the traditional model predictive control, and only needs to minimize the value function according to the current differential to obtain the optimal voltage vector. The inverter nonlinear compensation suppresses the disturbance caused by the inverter nonlinearity. The model-free predictive control method is suitable for surface-mounted and built-in permanent magnet synchronous motors. The current differential updated permanent magnet synchronous motor model-free predictive control method of this embodiment can realize high-performance control of permanent magnet synchronous motors, get rid of the dependence on parameters, improve the robustness of the system, solve the stagnation problem of current differential update, ensure the accuracy of predictive control, reduce the computational complexity, and improve the operation performance of permanent magnet synchronous motors.

[0158] In addition, this embodiment also provides a permanent magnet synchronous motor model-free predictive control system with current differential update, including a microprocessor and a memory connected to each other, and the microprocessor is programmed or configured to execute the permanent magnet synchronous motor model-free predictive control method with current differential update.

[0159] In addition, this embodiment also provides a computer-readable storage medium, which stores a computer program / instruction, and the computer program / instruction is programmed or configured to execute the permanent magnet synchronous motor model-free predictive control method with current differential update through a processor.

[0160] In addition, this embodiment also provides a computer program product, including a computer program / instruction, which is programmed or configured to execute the current differential updated permanent magnet synchronous motor model-free predictive control method through a processor.

[0161] Those skilled in the art should understand that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application can take the form of a computer program product implemented on one or more computer-readable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the process Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be stored in a computer-readable memory that can guide a computer or other programmable data processing device to work in a specific way, so that the instructions stored in the computer-readable memory produce a product including an instruction device, which implements the functions specified in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0162] The above is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments. All technical solutions under the concept of the present invention belong to the protection scope of the present invention. It should be pointed out that for ordinary technicians in this technical field, some improvements and modifications without departing from the principle of the present invention should also be regarded as the protection scope of the present invention.

Claims

1. A model-free predictive control method for a permanent magnet synchronous motor with current differential update, characterized in that: include: S101, the inductance is estimated to obtain dq Estimated shaft inductance and ; S102, for eight basic voltage vectors ~ Any of Voltage vector , combined with the current difference at time k as well as dq Estimated shaft inductance and Calculate the voltage vector by performing current differential update without lag Corresponding dq Shaft current differential ; S103, using a preset current differential value function and applying it to each voltage vector Corresponding dq Shaft current differential , and select the voltage vector with the minimum result of the current differential value function calculation As the optimal voltage vector ; S104, the optimal voltage vector Performing inverter nonlinear compensation, and then using the compensated voltage signal as the inverter control signal to control the inverter of the permanent magnet synchronous motor; In step S102, the current difference at time k is combined as well as dq Estimated shaft inductance and Calculate the voltage vector by performing current differential update without lag Corresponding dq Shaft current differential The function expression is: - + ， - + , ， , In the above formula, represents the voltage vector applied at time k, represents the unapplied voltage vector at time k, Not equal to , and Respectively As the voltage vector The corresponding dq Shaft current differential The d and q axis current components, and Respectively As the voltage vector The corresponding dq Shaft current differential The d and q axis current components, and They represent the dq axis current sampled at time k, and They represent the dq axis current sampled at time k-1 respectively, and They are dq The estimated inductance of the axis is , , To control the cycle.

2. The current differential updated permanent magnet synchronous motor model-free predictive control method according to claim 1, characterized in that: In step S101, the inductance is estimated to obtain dq Estimated shaft inductance and The function expression is: ， ， In the above formula, To control the cycle, and They are dq The estimated inductance of the axis is the middle amount, and when the zero voltage vector is applied first and the active voltage vector is applied later: ， ， And when the zero voltage vector is applied later and the active voltage vector is applied first, there are: ， ， in, and The active voltage vector is applied at time k Produced dq Shaft current differential, and are k moments respectively dq Shaft current change, is the active voltage vector, and The zero voltage vector is applied at time k Produced dq Shaft current differential, and The zero voltage vector is applied at time k-1 respectively Produced dq Shaft current differential, and The k-1 moments dq Shaft current change, and is the active voltage vector The voltage on the dq axis, is the zero voltage vector.

3. The current differential updated permanent magnet synchronous motor model-free predictive control method according to claim 2, characterized in that: Before step S101, the process also includes zero vector current differential reconstruction to obtain zero voltage vectors applied at time k. Produced dq Shaft current differential and : Get the voltage vector selected in the basic voltage vector for the first two cycles and , and satisfies , if the voltage vector and is the zero voltage vector, then the zero voltage vector applied at time k is calculated according to the following formula Produced dq Shaft current differential and : or , In the above formula, Apply zero voltage vector at time k Produced dq Axis current differential, including d and q axis current components and , Apply zero voltage vector for k-1 time Produced dq Axis current differential, including d and q axis current components and ; If the voltage vector and For two different active voltage vectors, the zero voltage vector applied at time k is calculated according to the following formula: Produced dq Shaft current differential and : ， ， In the above formula, and The voltage vector applied at time k is Produced dq Shaft current differential, and The voltage vector applied at time k-1 is Produced dq Shaft current differential.

4. The method for model-free predictive control of a permanent magnet synchronous motor using current differential update according to claim 1, characterized in that: The function expression of the current differential cost function preset in step S103 is: ， In the above formula, represents the current differential value function, and Represent the ideal current difference of d and q axes respectively, , , and Represent the d-axis and q-axis current reference values ​​respectively, and The voltage vector is calculated by continuously updating the current difference Corresponding dq Shaft current differential The d and q axis components in .

5. The current differential updated permanent magnet synchronous motor model-free predictive control method according to claim 1, characterized in that: In step S104, the optimal voltage vector Inverter nonlinear compensation refers to adding compensation time on the basis of dead time to compensate for voltage disturbance caused by inverter nonlinearity.

6. The method for model-free predictive control of a permanent magnet synchronous motor using current differential update according to claim 5, characterized in that: The calculation function expression of the compensation time is: ， In the above formula, Indicates the compensation time, is the threshold voltage of the IGBT in the inverter, is the threshold voltage of the inverter's freewheeling diode, To control the cycle, is the DC bus voltage, is the inverter dead time delay, and It is the turn-on and turn-off delay time of IGBT in the inverter.

7. A permanent magnet synchronous motor model-free predictive control system with current differential update, comprising a microprocessor and a memory connected to each other, characterized in that: The microprocessor is programmed or configured to execute the permanent magnet synchronous motor model-free predictive control method with current differential update as described in any one of claims 1 to 6.

8. A computer-readable storage medium having a computer program / instruction stored therein, characterized in that: The computer program / instruction is programmed or configured to execute the permanent magnet synchronous motor model-free predictive control method with current differential update as claimed in any one of claims 1 to 6 through a processor.

9. A computer program product comprising a computer program / instructions, characterized in that The computer program / instruction is programmed or configured to execute the permanent magnet synchronous motor model-free predictive control method with current differential update as claimed in any one of claims 1 to 6 through a processor.

Images