Cascade Model Predictive Control Method, System, Device and Medium for Dual Three-Phase Motors

Abstract

The present invention relates to the technical field of motor control, and provides a cascaded model predictive control method, system, device and medium for a dual-three-phase motor, including determining a target motor, collecting motor operation parameters, and establishing an initial operation equation of the target motor through the motor operation parameters; establishing a dual dq coordinate transformation matrix, obtaining an intermediate operation equation through the dual dq coordinate transformation matrix and the initial operation equation, decoupling to obtain a motor operation equation, and establishing a state space expression of the target motor through the motor operation equation; determining the operation sector of the target motor, and obtaining a time adjustment factor of the target motor therein; constructing a cost function, obtaining an input matrix set of the target motor, thereby constructing a quadratic programming equation; determining initial parameters, extracting an initial input, and using the gradient descent method to iteratively solve the quadratic programming equation to obtain a vector action time; and controlling the target motor. The present invention realizes efficient control of the target motor.

Description

Technical Field

[0001] The present invention relates to the technical field of motor control, and particularly to a cascaded model predictive control method, system, device and medium for a dual three-phase motor. Background Art

[0002] Compared with traditional three-phase motors, multiphase motors have the advantages of low-voltage high-power output, small torque ripple amplitude, strong fault tolerance, etc. The dual three-phase motor has received more attention due to its close connection with traditional three-phase motors. For the inverter in a dual three-phase motor, the three-level voltage source inverter has the advantages of low voltage stress of switching devices, small voltage change rate, small harmonic content of output voltage, etc., and is particularly suitable for medium-voltage high-power application scenarios. The three-level six-phase voltage source inverter outputs 3 voltage levels per phase, so the entire inverter has 3 6 state combinations, that is, 729 preselected voltage vectors, which is very complex for designing a space vector modulation strategy. And the redundant positive and negative voltage small vectors will cause different-direction offsets to the neutral point potential of the DC side of the three-level six-phase voltage source inverter, but reasonably distributing the action time of the positive and negative small vectors with opposite influence effects can effectively balance the neutral point potential. For the decoupling model, usually the vector space decoupling modeling method is adopted to map the variables of the motor into the α-β subspace related to electromechanical energy conversion and other subspaces unrelated to electromechanical energy conversion respectively, but it faces the problem of a large preselected vector set after decoupling; while the dq dual 6 transformation decouples the dual three-phase permanent magnet synchronous motor with isolated neutral points into two independent three-phase permanent magnet synchronous motors, which not only reduces the order of the original model, but also simplifies the number of preselected vectors from 3 3 to 2×3 Summary of the Invention

[0003] The present invention aims to solve at least one of the technical problems existing in the related art. For this reason, the present invention provides a cascaded model predictive control method, system, device and medium for a dual three-phase motor, realizing the balance of the neutral point potential of a three-phase motor and the efficient control of a dual three-phase motor.

[0004] The present invention provides a cascaded model predictive control method for a dual three-phase motor, including:

[0005] S1: Determine the target motor, collect the motor operation parameters of the target motor, and establish an initial operation equation of the target motor through the motor operation parameters;

[0006] S2: Obtain the dual three-phase electrical angles of the target motor, establish the dual dq coordinate transformation matrix of the target motor according to the dual three-phase electrical angles, obtain the intermediate operating equation through the dual dq coordinate transformation matrix and the initial operating equation, decouple through the intermediate operating equation to obtain the motor operating equation, and establish the state space expression of the target motor through the motor operating equation;

[0007] S3: Determine the operating sector of the target motor through the motor operating parameters, and obtain the time adjustment factor of the target motor in the operating sector;

[0008] S4: Construct a cost function, obtain the input matrix set of the target motor through the state space expression, and construct a quadratic programming equation through the input matrix set and the cost function;

[0009] S5: Determine the initial parameters of the quadratic programming equation, extract the initial input from the quadratic programming equation, and perform iterative solution of the quadratic programming equation using the initial parameters and the initial input with the gradient descent method to obtain the vector action time;

[0010] S6: Control the target motor through the vector action time, the state space expression, and the time adjustment factor.

[0011] According to the cascade model predictive control method for a dual three-phase motor provided by the present invention, in step S1, after collecting the motor operating parameters of the target motor, establish the initial operating equation of the target motor through the motor operating parameters:

[0012]

[0013]

[0014] Among them, is the phase voltage vector matrix, is the phase current vector matrix, is the resistance coefficient matrix, t is the operating time of the target motor in the current operating cycle, is the phase flux linkage vector matrix, is the inductance coefficient matrix, is the flux linkage coefficient matrix, is the flux linkage amplitude of the permanent magnet of the target motor.

[0015] According to the cascade model predictive control method for a dual three-phase motor provided by the present invention, step S2 specifically includes:

[0016] S21: Obtain the dual three-phase electrical angles of the target motor, construct a multi-dimensional transformation matrix, and establish a dual dq coordinate transformation matrix of the target motor through the dual three-phase electrical angles and the multi-dimensional transformation matrix;

[0017] S22: Multiply both ends of the dual dq coordinate transformation matrix and the initial operating equation to obtain the intermediate operating equation, and simplify the intermediate operating equation, thereby completing decoupling and obtaining the motor operating equation;

[0018] S23: Select state variables, perform formula transformation on the motor operating equation, and take the derivative of the state variables in the motor operating equation after formula transformation to establish the state space expression of the target motor.

[0019] According to the cascade model predictive control method for a dual three-phase motor provided by the present invention, step S3 specifically includes:

[0020] S31: Solve the future neutral point potential of the target motor through the motor operating parameters, and determine the operating sector of the target motor;

[0021] S32: Obtain the neutral point current expression through the future neutral point potential, and obtain the time adjustment factor of the target motor through the neutral point current expression in the operating sector.

[0022] According to the cascade model predictive control method for a dual three-phase motor provided by the present invention, step S4 specifically includes:

[0023] S41: Construct a cost function, obtain the input matrix set of the target motor through the state space expression, and construct the system state equation of the operating sector through the input matrix set; the cost function of the k-th operating cycle The expression is:

[0024]

[0025] Among them, is the actual value of the dq-axis current in the (k + 1)-th operating cycle, is the theoretical value of the dq-axis current in the (k + 1)-th operating cycle, represents taking the Euclidean norm;

[0026] S42: Rewrite the cost function through the system state equation, and construct the quadratic programming equation through the rewritten cost function.

[0027] According to the cascade model predictive control method for a dual three-phase motor provided by the present invention, step S5 specifically includes:

[0028] S51: Set the linear constraint value, initial solution, initial step size, and tolerance, and use the linear constraint value, the initial solution, the initial step size, and the tolerance as the initial parameters. Extract the initial input matrix and initial input vector from the quadratic programming equation, and use the initial input matrix and the initial input vector as the initial input;

[0029] S52: Calculate the gradient using the initial input and the initial parameters, and calculate the update amount using the gradient and the projection function;

[0030] S53: Iterate on the gradient, the initial solution, the update amount, and the initial step size until the iteration end condition is reached, and obtain the vector action time from the iterated initial solution.

[0031] According to the cascade model predictive control method for a dual three-phase motor provided by the present invention, the multidimensional transformation matrix is a Clarke-Park transformation matrix.

[0032] The present invention also provides a cascade model predictive control system for a dual three-phase motor, including:

[0033] Operating equation module: used to determine the target motor, collect the motor operating parameters of the target motor, and establish the initial operating equation of the target motor through the motor operating parameters;

[0034] State space expression module: used to obtain the dual three-phase electrical angle of the target motor, establish the dual dq coordinate transformation matrix of the target motor according to the dual three-phase electrical angle, obtain the intermediate operating equation through the dual dq coordinate transformation matrix and the initial operating equation, decouple through the intermediate operating equation to obtain the motor operating equation, and establish the state space expression of the target motor through the motor operating equation;

[0035] Time adjustment factor module: used to determine the operating sector of the target motor through the motor operating parameters, and obtain the time adjustment factor of the target motor in the operating sector;

[0036] Quadratic programming equation module: used to construct a cost function, obtain the input matrix set of the target motor through the state space expression, and construct a quadratic programming equation through the input matrix set and the cost function;

[0037] Equation solving module: used to determine the initial parameters of the quadratic programming equation, extract the initial input from the quadratic programming equation, and iteratively solve the quadratic programming equation using the initial parameters and the initial input and the gradient descent method to obtain the vector action time;

[0038] Motor control module: used to control the target motor through the vector action time, the state space expression, and the time adjustment factor.

[0039] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, the steps of the cascade model predictive control method for the dual three-phase motor as described in any one of the above are implemented.

[0040] The present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of the cascade model predictive control method for the dual three-phase motor as described in any one of the above are implemented.

[0041] One or more of the above technical solutions in the embodiments of the present invention have at least one of the following technical effects:

[0042] The cascade model predictive control method, system, device, and medium for a dual three-phase motor provided by the present invention decouple the dual three-phase permanent magnet synchronous motor into two three-phase motors by using a dual dq coordinate transformation matrix, greatly reducing the control difficulty. At the same time, a time adjustment factor is introduced into the inner control loop to adjust the action time of positive and negative small vectors, effectively balancing the neutral point potential of the three-phase motor. The action time of each vector in the control period is predicted by solving a quadratic programming problem through the gradient descent method, improving the control effect on the motor.

[0043] Some additional aspects and advantages of the present invention will be given in the following description, some will become obvious from the following description, or will be understood through the practice of the present invention. Description of the Drawings

[0044] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0045] Figure 1 is a schematic flow chart of the cascade model predictive control method for the dual three-phase motor provided by the present invention.

[0046] Figure 2 is a topology diagram of the target motor of the cascade model predictive control method for the dual three-phase motor provided by the present invention.

[0047] Figure 3 is a schematic diagram of the experimental verification results of the cascade model predictive control method for the dual three-phase motor provided by the present invention.

[0048] Figure 4 It is a schematic structural diagram of the cascaded model predictive control system of the dual three-phase motor provided by the present invention.

[0049] Figure 5 It is a schematic structural diagram of the cascaded model predictive control device of the dual three-phase motor provided by the present invention.

[0050] Reference numerals:

[0051] 100, operating equation module; 200, state space expression module; 300, time adjustment factor module; 400, quadratic programming equation module; 500, equation solving module; 600, motor control module; 810, processor; 820, communication interface; 830, memory; 840, communication bus. Specific embodiments

[0052] In order to make the objectives, technical solutions and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without any creative work shall fall within the protection scope of the present invention. The following embodiments are used to illustrate the present invention, but cannot be used to limit the scope of the present invention.

[0053] In the description of the embodiments of the present invention, it should be noted that the terms "first", "second", and "third" are only used for descriptive purposes and cannot be construed as indicating or implying relative importance.

[0054] In the description of the embodiments of the present invention, it should be noted that unless otherwise clearly defined and limited, the terms "connected" and "connected" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms in the embodiments of the present invention can be understood according to specific circumstances.

[0055] In the description of this specification, the descriptions referring to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the embodiments of the present invention. In this specification, the schematic expressions of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0056] The following combines Figures 1 to 5 to describe the specific implementation of the present invention:

[0057] Figure 1 is a schematic flow chart of the cascaded model predictive control method for a dual three-phase motor provided by the present invention. First, the target motor is determined, the motor operating parameters are obtained, and an initial operating equation is established; then a dual dq coordinate transformation matrix is established, an intermediate operating equation is obtained and decoupled, and a state space expression is established; subsequently, the operating sector of the target motor is determined, so as to obtain a time adjustment factor, then a cost function is constructed, and an input matrix set is obtained, thereby constructing a quadratic programming equation; then the initial parameters and initial input are determined, and the gradient descent method is used for iterative solution to obtain the vector action time; finally, the target motor is controlled by the vector action time, the state space expression, and the time adjustment factor.

[0058] The present invention provides a cascaded model predictive control method for a dual three-phase motor, including:

[0059] S1: Determine the target motor, collect the motor operating parameters of the target motor, and establish an initial operating equation of the target motor through the motor operating parameters;

[0060] Further, the purpose of this stage is to collect the operating parameters of the dual three-phase motor and establish an initial operating equation of the dual three-phase motor through the operating parameters for subsequent operations on the initial operating equation to further establish the state space expression of the dual three-phase motor. Among them, after collecting the motor operating parameters of the target motor, an initial operating equation of the target motor is established through the motor operating parameters:

[0061]

[0062]

[0063] Among them, is the phase voltage vector matrix, is the phase current vector matrix, is the resistance coefficient matrix, t is the running time of the target motor in the current running cycle, is the phase flux linkage vector matrix, is the inductance coefficient matrix, is the flux linkage coefficient matrix, is the flux linkage amplitude of the permanent magnet of the target motor.

[0064] For the above steps, the specific implementation in this embodiment is as follows:

[0065] First, determine the dual three-phase motor to be controlled and use it as the target motor. The topology diagram of the target motor is as Figure 2 shown. Then, collect the motor operation parameters of the target motor. The motor operation parameters include the current of each phase of the target motor, the resistance of the stator, the electrical angle of the rotor, the self-inductance and mutual inductance of the winding, etc. The expression of the initial operation equation of the target motor is:

[0066]

[0067]

[0068] Among them, , and is the phase voltage vector matrix, where represents the voltage of phase A, represents the voltage of phase B, represents the voltage of phase C, represents the voltage of phase U, represents the voltage of phase V, represents the voltage of phase W, represents taking the transpose of the matrix; , and is the phase current vector matrix, where represents the current of phase A, represents the current of phase B, represents the current of phase C, represents the current of phase U, represents the current of phase V, represents the current of phase W; = , and is the resistance coefficient matrix, where diag[] represents taking the diagonal matrix, represents the resistance value of the stator, and the resistance value of each stator is the same; t is the running time of the dual three-phase motor in the current running cycle, , and is the phase flux linkage vector matrix, where represents the flux linkage of phase A, The magnetic flux linkage representing phase B, The magnetic flux linkage representing phase C, The magnetic flux linkage representing phase U, The magnetic flux linkage representing phase V, The magnetic flux linkage representing phase W; , and is the inductance coefficient matrix, where, represents the self - inductance and mutual inductance inside the first set of three - phase windings of the dual three - phase motor, represents the self - inductance and mutual inductance inside the second set of three - phase windings of the dual three - phase motor, represents the mutual inductance between the first set of three - phase windings and the second set of three - phase windings of the dual three - phase motor, represents the mutual inductance between the second set of three - phase windings and the first set of three - phase windings of the dual three - phase motor; is the magnetic flux coefficient matrix,

[0069]

[0070] where, is the electrical angle between the rotor magnetic pole position and the axis of the phase A winding; is the magnetic flux amplitude of the permanent magnet of the target motor.

[0071] S2: Obtain the dual three - phase electrical angle of the target motor, establish the dual dq coordinate transformation matrix of the target motor according to the dual three - phase electrical angle, obtain the intermediate operation equation through the dual dq coordinate transformation matrix and the initial operation equation, decouple through the intermediate operation equation to obtain the motor operation equation, and establish the state - space expression of the target motor through the motor operation equation. Where,

[0072] Furthermore, the purpose of this stage is to establish the dual dq coordinate transformation matrix of the target motor, decouple through the dual dq coordinate transformation matrix to obtain the motor operation equation, and finally obtain the state - space expression of the target motor. Step S2 specifically includes:

[0073] S21: Obtain the dual three - phase electrical angle of the target motor, construct a multi - dimensional transformation matrix, and establish the dual dq coordinate transformation matrix of the target motor through the dual three - phase electrical angle and the multi - dimensional transformation matrix;

[0074] S22: Multiply both ends of the dual dq coordinate transformation matrix and the initial operation equation to obtain the intermediate operation equation, simplify the intermediate operation equation, thereby completing decoupling and obtaining the motor operation equation;

[0075] S23: Select state variables, perform formula transformation on the motor operating equation, and take the derivative of the state variables in the motor operating equation after formula transformation to establish the state space expression of the target motor. Wherein, the multi-dimensional transformation matrix is the Clarke-Park transformation matrix.

[0076] For the above steps, the specific implementation in this embodiment is as follows:

[0077] First, obtain the dual three-phase electrical angle of the dual three-phase motor from the motor operating parameters. Since the stator winding of the dual three-phase motor consists of two sets of three-phase symmetrical windings, it can be regarded as a combination of two three-phase subsystems. Traditional three-phase motor coordinate transformation is applied to each subsystem respectively. Therefore, two Clarke-Park transformation matrices are constructed as the multi-dimensional transformation matrices. Here, the dual three-phase electrical angle refers to the electrical angle between the two three-phase subsystems, which is 30° in this embodiment. Then, the dual dq coordinate transformation matrix T of the target motor can be established through the dual three-phase electrical angle and the multi-dimensional transformation matrix. 61 :

[0078]

[0079] Wherein, P1 is the first Clarke-Park transformation matrix, P2 is the second Clarke-Park transformation matrix, and

[0080] ;

[0081]

[0082] Next, multiply both ends of the dual dq coordinate transformation matrix and the initial operating equation to obtain the intermediate operating equation:

[0083]

[0084]

[0085] Wherein, is the inverse matrix of the dual dq coordinate transformation matrix, and d represents differentiation.

[0086] Simplify and solve the intermediate operating equation to decouple the initial operating equation to the dq plane and obtain the motor operating equation:

[0087]

[0088]

[0089] Wherein, is the component of the decoupled voltage of phase A, phase B, and phase C on the d axis, is the component of the decoupled voltage of phase A, phase B, and phase C on the q axis. is the component of the decoupled voltage of the U-phase, V-phase, and W-phase on the d-axis, is the component of the decoupled voltage of the U-phase, V-phase, and W-phase on the q-axis, is the component of the decoupled current of the A-phase, B-phase, and C-phase on the d-axis, is the component of the decoupled current of the A-phase, B-phase, and C-phase on the q-axis, is the component of the decoupled current of the U-phase, V-phase, and W-phase on the d-axis, is the component of the decoupled current of the U-phase, V-phase, and W-phase on the q-axis, is the electrical angular velocity of the rotor, is the component of the decoupled magnetic flux of the A-phase, B-phase, and C-phase on the d-axis, is the component of the decoupled magnetic flux of the A-phase, B-phase, and C-phase on the q-axis, is the component of the decoupled magnetic flux of the U-phase, V-phase, and W-phase on the d-axis, is the component of the decoupled magnetic flux of the U-phase, V-phase, and W-phase on the q-axis, is the derivative of, is the derivative of, is the derivative of, is the derivative of, L d is the component of the inductance on the d-axis, L q is the component of the inductance on the q-axis, L dd is the self-inductance of the d-axis inductance, L qq is the self-inductance of the q-axis inductance.

[0090] Next, select the state variables. In this embodiment, the selected state variables are , perform formula transformation on the motor operation equation, and take the derivative of the state variables in the motor operation equation after formula transformation to establish the state space expression of the target motor:

[0091]

[0092]

[0093]

[0094]

[0095] Among them, is the derivative of, is the derivative of, is the derivative of , is the derivative of , is the mutual inductance between the q-axis inductance and the d-axis inductance, L dq is the mutual inductance between the d-axis inductance and the q-axis inductance, is the current input matrix, is the voltage input matrix, is the flux linkage input matrix.

[0096] S3: Determine the operating sector of the target motor based on the motor operating parameters, and obtain the time adjustment factor of the target motor in the operating sector;

[0097] Furthermore, the purpose of this stage is to determine the operating sector of the target motor and calculate the time adjustment factor of the target motor in the operating sector. Among them, step S3 specifically includes:

[0098] S31: Solve the future neutral point potential of the target motor through the motor operating parameters, and determine the operating sector of the target motor;

[0099] S32: Obtain the neutral point current expression through the future neutral point potential, and obtain the time adjustment factor of the target motor in the operating sector through the neutral point current expression.

[0100] For the above steps, the specific implementation in this embodiment is as follows:

[0101] First, since it is desired that the future neutral point potential is 0, the future neutral point potential of the target motor in the next operating cycle can be solved through Kirchhoff's law and the motor operating parameters U np :

[0102]

[0103] Among them, is the voltage of the first capacitor on the DC side of the target motor in the next operating cycle, that is, the (k + 1)-th operating cycle, is the voltage of the second capacitor on the DC side of the target motor in the next operating cycle, is the voltage of the second capacitor of the target motor in the current operating cycle, is the voltage of the first capacitor of the target motor in the current operating cycle, that is, the k-th operating cycle, is the time length of an operating cycle, is the current when the second capacitor of the target motor is charging and discharging in the current operating cycle, is the current when the first capacitor of the target motor is charging and discharging in the current operating cycle, are the capacitance values of the first capacitor and the second capacitor, is the neutral point potential of the target motor in the current operating cycle, that is, the measured value of the neutral point potential, is the neutral point current, that is, the measured value of the current at the neutral point of the target motor.

[0104] Here, according to the topology diagram of the target motor, it can be known that the neutral point is located between the first capacitor and the second capacitor, and there are no other components between the first capacitor and the second capacitor. Therefore, the neutral point potential is 1 / 2 of the voltage difference between the first capacitor and the second capacitor, and the current flowing into the neutral point is defined as positive.

[0105] Next, determine the operating sector of the target motor. For the operation of a dual three-phase motor, there are six large sectors in total for the three phases, and there are six small sectors in each large sector. When the load of the target motor is certain, as the motor rotor rotates, it will operate in a certain number of determined small sectors in sequence. These small sectors are the operating sectors of the target motor.

[0106] At this time, let the future neutral point potential be 0, and the neutral point current expression can be obtained:

[0107]

[0108] In the operating sector, different vectors exist in different sectors. There are three vectors in each sector, and each vector has a corresponding action time. At this time, taking the first small sector of the first large sector of the ABC phase as an example, there are redundant small vectors ONN and POO, small vector OON, and zero vector OOO. Let the total action time of the redundant small vectors ONN and POO be T1, the action time of the small vector OON be T2, and the action time of the zero vector OOO be T3. According to the different phase currents participating in the action in different vectors, there are:

[0109]

[0110] And there is:

[0111]

[0112] Among them, the influence of each vector on the neutral point current is different, and the redundant small vector ONN is , and that of POO is , the small vector OON is , and the zero vector OOO is .

[0113] Among them, is the time adjustment factor of this sector. According to the above equation relationship and replacing with , the time adjustment factor of this sector can be calculated as:

[0114]

[0115] For other sectors in the operating sector, the corresponding time adjustment factors are also solved using the same method according to the differences in the sectors, so as to obtain the time adjustment factors of the target motor. Since there are two redundant small vectors ONN and POO, and T1 is the total action time of the redundant small vectors ONN and POO, a time adjustment factor is required to control the respective action times of the two redundant small vectors. During the process of controlling the operation of the target motor, the corresponding time adjustment factor can be selected according to the sector passed through when the target motor operates.

[0116] S4: Construct a cost function, obtain the input matrix set of the target motor through the state space expression, and construct a quadratic programming equation through the input matrix set and the cost function;

[0117] Furthermore, the purpose of this stage is to construct a cost function and obtain the input matrix set of the target motor, so as to construct a quadratic programming equation and thus solve the quadratic programming equation. Among them, step S4 specifically includes:

[0118] S41: Construct a cost function, obtain the input matrix set of the target motor through the state space expression, and construct the system state equation of the operating sector through the input matrix set; The cost function of the k-th operating cycle The expression of

[0119]

[0120] Among them, is the actual value of the dq-axis current in the (k + 1)-th operating cycle, is the theoretical value of the dq-axis current in the (k + 1)-th operating cycle, represents taking the Euclidean norm

[0121] S42: Rewrite the cost function through the system state equation, and construct the quadratic programming equation through the rewritten cost function.

[0122] For the above steps, the specific implementation in this embodiment is as follows:

[0123] First, construct a cost function :

[0124]

[0125] Among them, is the actual value of the dq-axis current in the (k + 1)-th operating cycle, that is, the next operating cycle, is the theoretical value of the dq-axis current in the (k + 1)-th operating cycle, denotes taking the Euclidean norm.

[0126] Next, the input matrix set of the target motor is obtained through the voltage input matrix, current input matrix, and flux linkage input matrix of the state space expression. The input matrix set includes the sector voltage input matrix , the sector current input matrix and the sector flux linkage input matrix . Taking the three-phase motor model in the first small sector of the first large sector of the A-phase, B-phase, and C-phase in the operating sector as an example, the system state equation of the operating sector can be written as:

[0127]

[0128]

[0129]

[0130]

[0131] where is the current value on the d-axis in the (k + 1)-th operating cycle, is the current value on the q-axis in the (k + 1)-th operating cycle, is the actual value of the current on the d-axis in the k-th operating cycle, i.e., the current operating cycle, is the actual value of the current on the q-axis in the k-th operating cycle, is the actual value of the voltage on the d-axis in the k-th operating cycle, is the actual value of the voltage on the q-axis in the k-th operating cycle.

[0132] Next, the cost function is rewritten through the system state equation as:

[0133]

[0134] where T park is the Park transformation matrix, T SVM is the vector amplitude matrix, is the dq-axis current value in the k-th operating cycle, is the matrix of the action time of each vector in the sector to be solved.

[0135] Construct the first auxiliary variable r, and and the second auxiliary variable M, and , then:

[0136]

[0137] where is the transpose of r, is the transpose of M, is the transpose of. Here, since r is a 2×1 matrix, M is a 2×3 matrix, and t is a 3×1 matrix, so is a 1×1 matrix, that is, a scalar, is also a 1×1 matrix, i.e., a scalar. Therefore, the two scalars can be combined to obtain . At this time, the solution of the cost function can be transformed into a quadratic programming problem, and a quadratic programming equation is constructed:

[0138]

[0139] where H is the initial input matrix, and , f is the initial input vector, and , min() represents the minimum value within the parentheses, b represents the linear constraint value of the quadratic programming equation, includes T1, T2 and T3.

[0140] S5: Determine the initial parameters of the quadratic programming equation, extract the initial input from the quadratic programming equation, and use the initial parameters and the initial input to iteratively solve the quadratic programming equation using the gradient descent method to obtain the vector action time;

[0141] Furthermore, the purpose of this stage is to iteratively solve the quadratic programming equation using the gradient descent method to obtain the action time of each redundant vector for control. Specifically, step S5 specifically includes:

[0142] S51: Set the linear constraint value, initial solution, initial step size, and tolerance, and use the linear constraint value, the initial solution, the initial step size, and the tolerance as the initial parameters. Extract the initial input matrix and the initial input vector from the quadratic programming equation, and use the initial input matrix and the initial input vector as the initial input;

[0143] S52: Calculate the gradient through the initial input and the initial parameters, and calculate the update amount through the gradient and the projection function;

[0144] S53: Iterate the gradient, the initial solution, the update amount, and the initial step size until the iteration end condition is reached, and obtain the vector action time through the iterated initial solution.

[0145] For the above steps, the specific implementation in this embodiment is as follows:

[0146] In this embodiment, first set the linear constraint value b of the quadratic programming equation, and b = , since the purpose of solving the quadratic programming equation is to obtain the action time of each vector in the sector to be solved, three values are included in the initial solution x0, and x0 = {1 / 3, 1 / 3, 1 / 3}, the initial step size is alpha_0, and alpha_0 = 0.1, the tolerance is tol, and the initial solution, the initial step size, and the tolerance are used as initial parameters. The initial input matrix H and the initial input vector f are extracted from the quadratic programming equation, and the initial input matrix and the initial input vector are used as the initial input.

[0147] Then calculate the gradient g through the initial input and the initial parameters c :

[0148]

[0149] Among them, represents matrix multiplication. And calculate the update amount z through the gradient and the projection function c :

[0150]

[0151] Among them, proj() represents the projection function that projects the content in the parentheses onto the linear plane.

[0152] Subsequently, calculate the iterative initial solution x d :

[0153]

[0154] Calculate the iterative update amount z d and the iterative gradient g d :

[0155]

[0156] Calculate the gradient difference y c and calculate the iterative initial step size alpha_1 through the gradient difference:

[0157]

[0158] Among them, z c ' is the transpose of z c .

[0159] Subsequently, let alpha_0 = alpha_1, x0 = x d , g c = g dAnd return to step S51 to recalculate for iteration until the update amount is less than the tolerance or the number of iterations reaches the iteration number threshold set according to experience, then complete the iteration of the initial solution. Take the iterated initial solution as the vector action time. The three values included in the initial solution respectively correspond to T1, T2, and T3 in the quadratic programming equation in sequence.

[0160] For other sectors in the operating sector, use the same method to solve their corresponding initial solutions according to the different phases of the sectors, so as to obtain the initial solution of the target motor. During the process of controlling the operation of the target motor, select the corresponding initial solution according to the sector passed by the target motor during operation.

[0161] S6: Control the target motor through the vector action time, the state space expression, and the time adjustment factor.

[0162] Furthermore, in this stage, according to the sectors passed by the target motor at different times, use the vector action time and the time adjustment factor to control each vector of the target motor in the operating sector, and combine the state space expression to characterize the working state of the target motor, thus completing the control of the target motor.

[0163] In order to verify the effectiveness of the present invention, on a dual-three-phase motor drive experimental platform, an experimental verification was carried out on the cascaded model predictive control method for the dual-three-phase motor proposed by the present invention. The experimental results are as Figure 3 shown. In terms of phase current, taking the current of phase A as an example, as Figure 3 shown in (a), the unit of the horizontal axis is seconds. Input the experimental data shown in Figure 3 (a) into the MATLB tool for analysis, and the total harmonic distortion value of the method provided by the present invention can be obtained as 3.25%. In terms of d the d-axis current and q-axis current, as Figure 3 shown in (b), the unit of the horizontal axis is seconds. Input the experimental data shown in Figure 3 (b) into the MATLB tool for analysis. In the method provided by the present invention, d the current on the d-axis fluctuates within the range of the reference value of 0 A, and the peak-to-peak ripple is 0.57 A. q The current on the q-axis also fluctuates within the reference value range, and its peak-to-peak ripple is 0.41 A. As Figure 3 shown in (c), the unit of the horizontal axis is seconds. Input the experimental data shown in Figure 3 (c) into the MATLB tool for analysis. In terms of neutral point potential control, through the method provided by the present invention, the neutral point potential is basically stabilized at 0 V, and the peak-to-peak ripple is 0.95 V. Figure 3The error in the experiment is mainly due to a certain delay between the control command for the target motor and the actual operating condition of the motor, resulting in fluctuations in the operating state of the target motor.

[0164] The cascade model predictive control device for a dual three-phase motor provided by the present invention will be described below. The cascade model predictive control device for a dual three-phase motor described below can be mutually referred to in correspondence with the cascade model predictive control method for a dual three-phase motor described above.

[0165] Figure 4 The structural schematic diagram of the cascade model predictive control system for a dual three-phase motor is exemplified, as Figure 4 shown, for implementing the cascade model predictive control method for a dual three-phase motor as described above, including:

[0166] Operating equation module 100: for determining a target motor, collecting motor operating parameters of the target motor, and establishing an initial operating equation of the target motor through the motor operating parameters;

[0167] State space expression module 200: for obtaining the dual three-phase electrical angles of the target motor, establishing a dual dq coordinate transformation matrix of the target motor according to the dual three-phase electrical angles, obtaining an intermediate operating equation through the dual dq coordinate transformation matrix and the initial operating equation, decoupling through the intermediate operating equation to obtain a motor operating equation, and establishing a state space expression of the target motor through the motor operating equation;

[0168] Time adjustment factor module 300: for determining the operating sector of the target motor through the motor operating parameters, and obtaining the time adjustment factor of the target motor in the operating sector;

[0169] Quadratic programming equation module 400: for constructing a cost function, obtaining an input matrix set of the target motor through the state space expression, and constructing a quadratic programming equation through the input matrix set and the cost function;

[0170] Equation solving module 500: for determining the initial parameters of the quadratic programming equation, extracting an initial input from the quadratic programming equation, and iteratively solving the quadratic programming equation using the gradient descent method through the initial parameters and the initial input to obtain a vector action time;

[0171] Motor control module 600: for controlling the target motor through the vector action time, the state space expression, and the time adjustment factor.

[0172] On the other hand, Figure 5 The entity structural schematic diagram of an electronic device is exemplified, as Figure 5As shown, the electronic device may include: a processor 810, a communication interface 820, a memory 830, and a communication bus 840. Among them, the processor 810, the communication interface 820, and the memory 830 complete communication with each other through the communication bus 840. The processor 810 can call the logical instructions in the memory 830 to execute the cascaded model predictive control method for a dual-three-phase motor. The method includes:

[0173] S1: Determine the target motor, collect the motor operation parameters of the target motor, and establish an initial operation equation of the target motor through the motor operation parameters;

[0174] S2: Obtain the dual-three-phase electrical angle of the target motor, establish a dual dq coordinate transformation matrix of the target motor according to the dual-three-phase electrical angle, obtain an intermediate operation equation through the dual dq coordinate transformation matrix and the initial operation equation, decouple through the intermediate operation equation to obtain a motor operation equation, and establish a state space expression of the target motor through the motor operation equation;

[0175] S3: Determine the operation sector of the target motor through the motor operation parameters, and obtain a time adjustment factor of the target motor in the operation sector;

[0176] S4: Construct a cost function, obtain an input matrix set of the target motor through the state space expression, and construct a quadratic programming equation through the input matrix set and the cost function;

[0177] S5: Determine the initial parameters of the quadratic programming equation, extract an initial input from the quadratic programming equation, and perform iterative solution of the quadratic programming equation using the initial parameters and the initial input with the gradient descent method to obtain a vector action time;

[0178] S6: Control the target motor through the vector action time, the state space expression, and the time adjustment factor.

[0179] In addition, when the logical instructions in the above-mentioned memory 830 can be implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a 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 several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.

[0180] In another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is used to implement the cascade model predictive control method for a dual-three-phase motor provided by the above-mentioned various methods. The method includes:

[0181] S1: Determine the target motor, collect the motor operation parameters of the target motor, and establish an initial operation equation of the target motor through the motor operation parameters;

[0182] S2: Obtain the dual-three-phase electrical angles of the target motor, establish a dual-dq coordinate transformation matrix of the target motor according to the dual-three-phase electrical angles, obtain an intermediate operation equation through the dual-dq coordinate transformation matrix and the initial operation equation, decouple through the intermediate operation equation to obtain a motor operation equation, and establish a state space expression of the target motor through the motor operation equation;

[0183] S3: Determine the operation sector of the target motor through the motor operation parameters, and obtain a time adjustment factor of the target motor in the operation sector;

[0184] S4: Construct a cost function, obtain an input matrix set of the target motor through the state space expression, and construct a quadratic programming equation through the input matrix set and the cost function;

[0185] S5: Determine the initial parameters of the quadratic programming equation, extract an initial input from the quadratic programming equation, and iteratively solve the quadratic programming equation using the gradient descent method through the initial parameters and the initial input to obtain a vector action time;

[0186] S6: Control the target motor based on the vector action time, the state space expression, and the time adjustment factor.

[0187] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art can understand and implement it without creative effort.

[0188] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solutions or the part that contributes to the prior art can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0189] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. Cascade model predictive control method for a dual three-phase motor, characterized in that Including: S1: Determine the target motor, collect the motor operation parameters of the target motor, and establish an initial operation equation of the target motor based on the motor operation parameters; S2: Obtain the double three-phase electrical angles of the target motor, establish a double dq coordinate transformation matrix of the target motor according to the double three-phase electrical angles, obtain an intermediate operation equation through the double dq coordinate transformation matrix and the initial operation equation, decouple through the intermediate operation equation to obtain a motor operation equation, and establish a state space expression of the target motor through the motor operation equation; S3: Determine the operation sector of the target motor through the motor operation parameters, and obtain the time adjustment factor of the target motor in the operation sector; wherein, let the future neutral point potential be 0 to obtain a neutral point current expression. At this time, for the first small sector of the first large sector of the ABC phase, there are redundant small vectors ONN and POO, small vector OON, and zero vector OOO. Suppose the total action time of the redundant small vectors ONN and POO is T1, the action time of the small vector OON is T2, and the action time of the zero vector OOO is T3. According to the different phase currents participating in the action in different vectors, there are: And there is: Among them, the influence of each vector on the neutral point current is different, and the redundant small vector ONN is , that of POO is , the small vector OON is , the zero vector OOO is , is the capacitance value of the first capacitor and the second capacitor, is the measured value of the neutral point potential, is the time length of one operating cycle, is the neutral point current, represents the current of phase A, represents the current of phase B, represents the current of phase C, is the time adjustment factor of this sector. According to the equation relationship and replacing with , the time adjustment factor of this sector is calculated as: For other sectors in the operation sector, the corresponding time adjustment factors are also solved using the same method according to the different sectors; S6: Construct a cost function, obtain an input matrix set of the target motor through the state space expression, and construct a quadratic programming equation through the input matrix set and the cost function; S7: Determine the initial parameters of the quadratic programming equation, extract an initial input from the quadratic programming equation, and perform iterative solution of the quadratic programming equation using the initial parameters and the initial input with the gradient descent method to obtain the vector action time; S8: Control the target motor through the vector action time, the state space expression, and the time adjustment factor.

2. The cascade model predictive control method for a dual-three-phase motor according to claim 1, characterized in that, In step S1, after collecting the motor operation parameters of the target motor, establish the initial operation equation of the target motor through the motor operation parameters: Among them, is the phase voltage vector matrix, is the phase current vector matrix, is the resistance coefficient matrix, t is the running time of the target motor in the current operating cycle, is the phase flux linkage vector matrix, is the inductance coefficient matrix, is the flux linkage coefficient matrix, is the flux linkage amplitude of the permanent magnet of the target motor.

3. The cascade model predictive control method for the dual three-phase motor according to claim 1, characterized in that, Step S2 specifically includes: S21: Obtain the double three-phase electrical angles of the target motor, construct a multi-dimensional transformation matrix, and establish a double dq coordinate transformation matrix of the target motor through the double three-phase electrical angles and the multi-dimensional transformation matrix; S22: Multiply both ends of the double dq coordinate transformation matrix and the initial operation equation to obtain the intermediate operation equation, simplify the intermediate operation equation, so as to complete decoupling and obtain the motor operation equation; S23: Select state variables, perform formula transformation on the motor operation equation, and take the derivative of the state variables in the motor operation equation after formula transformation to establish a state space expression of the target motor.

4. The cascade model predictive control method for the dual-three-phase motor according to claim 1, characterized in that, Step S3 specifically includes: S31: Solve the future neutral point potential of the target motor through the motor operation parameters, and determine the operation sector of the target motor; S32: Obtain the neutral point current expression from the future neutral point potential, and obtain the time adjustment factor of the target motor in the operating sector from the neutral point current expression.

5. The cascade model predictive control method for a dual-three-phase motor according to claim 1, wherein Step S4 specifically includes: S41: Construct a cost function, obtain the input matrix set of the target motor through the state space expression, and construct the system state equation of the operating sector through the input matrix set; the cost function in the k-th operating cycle is expressed as: Among them, is the actual value of the dq-axis current in the (k + 1)-th operating cycle, is the theoretical value of the dq-axis current in the (k + 1)-th operating cycle, represents taking the Euclidean norm; S42: Rewrite the cost function according to the system state equation, and construct the quadratic programming equation from the rewritten cost function.

6. The cascade model predictive control method for a dual-three-phase motor according to claim 1, wherein Step S5 specifically includes: S51: Set the linear constraint value, initial solution, initial step size, and tolerance, and use the linear constraint value, the initial solution, the initial step size, and the tolerance as the initial parameters. Extract the initial input matrix and initial input vector from the quadratic programming equation, and use the initial input matrix and the initial input vector as the initial input; S52: Calculate the gradient from the initial input and the initial parameters, and calculate the update amount from the gradient and the projection function; S53: Iterate the gradient, the initial solution, the update amount, and the initial step size until the iteration end condition is reached, and obtain the vector action time from the iterated initial solution.

7. The cascade model predictive control method for a dual-three-phase motor according to claim 3, characterized in that The multidimensional transformation matrix is the Clarke-Park transformation matrix.

8. Cascaded model predictive control system for a dual three-phase motor, for performing the cascaded model predictive control method of a dual three-phase motor according to any one of claims 1 to 7, characterized in that, It includes: Operating equation module: Used to determine the target motor, collect the motor operating parameters of the target motor, and establish the initial operating equation of the target motor from the motor operating parameters; State space expression module: Used to obtain the double three-phase electrical angles of the target motor, establish the double dq coordinate transformation matrix of the target motor according to the double three-phase electrical angles, obtain the intermediate operating equation from the double dq coordinate transformation matrix and the initial operating equation, decouple through the intermediate operating equation to obtain the motor operating equation, and establish the state space expression of the target motor from the motor operating equation; Time adjustment factor module: Used to determine the operating sector of the target motor from the motor operating parameters, and obtain the time adjustment factor of the target motor in the operating sector; Quadratic programming equation module: Used to construct the cost function, obtain the input matrix set of the target motor from the state space expression, and construct the quadratic programming equation from the input matrix set and the cost function; Equation solving module: Used to determine the initial parameters of the quadratic programming equation, extract the initial input from the quadratic programming equation, and iteratively solve the quadratic programming equation using the gradient descent method from the initial parameters and the initial input to obtain the vector action time; Motor control module: Used to control the target motor from the vector action time, the state space expression, and the time adjustment factor.

9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein, When the processor executes the computer program, it implements the steps of the cascade model predictive control method for the double three-phase motor according to any one of claims 1 to 7.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the cascade model predictive control method for the double three-phase motor according to any one of claims 1 to 7.

Images