High-control-precision finite control set model prediction control method and device

Abstract

The invention discloses a finite control set model prediction control method and device with high control precision. The method comprises the following steps: obtaining a state-space equation of a control object according to a system optimization target, and obtaining an increment of the optimization target from k to k + 1; according to the xj peak value of the optimization target at the moment k to the moment k + 1 and the error of the optimization target at the moment k + 1, constructing a value function of multi-target optimization of the system at the moment k; the optimal switch action position and the optimal switch state at the current k moment are solved by minimizing the constructed multi-objective optimized value function; and compensating the delay time to construct a compensation-considered value function, and obtaining an optimal switch action position and an optimal switch state at the k + 1 moment by minimizing the compensation-considered value function. According to the invention, the switching position of the switching state can be flexibly changed, the switching time is flexible, variable and online adjustable, the switching precision of the inverter is improved, and the characteristic that switching does not exceed one time in each control period is kept.

Description

Technical Field

[0001] The present invention belongs to the technical field of power electronics and electric drive, and in particular relates to a finite control set model predictive control method and device with high control accuracy. Background Art

[0002] Finite Control Set Model Predictive Control (FCS-MPC) has a wide range of applications in the fields of power electronics and power transmission. In the AC motor speed control system in the industrial field, such as the drive control of asynchronous motors and permanent magnet synchronous motors, its precise control characteristics can be used to achieve high-precision regulation of motor speed, torque, etc., and improve the dynamic performance of the system.

[0003] In the process of connecting renewable energy sources such as solar photovoltaic power generation and wind power generation to the grid, this control method is used to accurately control the switching action of the inverter, efficiently and accurately convert direct current into alternating current that meets the requirements of the grid, ensure the quality of power and the stable connection of the power generation system to the grid, and improve the utilization efficiency of renewable energy and grid compatibility.

[0004] The main advantage of traditional FCS-MPC is that it can easily achieve multi-objective optimization through a unified value function. At the same time, the controller design is simple and intuitive, and it has better system dynamic performance than the traditional proportional-integral controller. However, it has obvious disadvantages: the switch can only be switched at a fixed position, and the possible switching position is only related to the initial value, which reduces the control accuracy. In the fields of grid connection and motor control, it will reduce the output waveform quality of the power electronic converter. Summary of the invention

[0005] In view of the shortcomings of the existing finite control set model predictive control (FCS-MPC) technology, the purpose of the present invention is to provide a finite control set model predictive control method and device with high control accuracy, which can improve the problem of reduced output waveform quality of power electronic converters caused by the fixed action position of traditional FCS-MPC switches, while maintaining the advantage of high dynamic performance of traditional FCS-MPC, and the method has a wide range of applications.

[0006] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0007] In a first aspect, the present invention provides a finite control set model predictive control method with high control accuracy, the method comprising:

[0008] According to the system optimization target, the state space equation of the control object is obtained, and the state space equation is written in discrete form to obtain the increment of the optimization target from time k to k+1;

[0009] According to the optimization goal, x from time k to k+1 jThe peak value and the error at time k+1 are used to construct the value function of multi-objective optimization of the system at time k;

[0010] By minimizing the value function of the constructed multi-objective optimization, the optimal switch action position and optimal switch state at the current k moment are solved;

[0011] The delay time is compensated by constructing a value function that takes compensation into account, and the optimal switch action position and optimal switch state at time k+1 are obtained by minimizing the value function that takes compensation into account.

[0012] In one implementation, the state space equation of the control object obtained according to the system optimization target is:

[0013]

[0014] In the formula, x represents the state space, u represents the system input; A represents the state transfer matrix; B is the input matrix;

[0015]

[0016] The discrete time is t s , then the jth optimization target x j The discrete form of is:

[0017] x j (k+1)=(t s a j +1)x j (k)+t s b j u(k),j=1,2,3...,n

[0018] The change Δx of the jth optimization target from time k to time k+1 j (k) can be expressed as:

[0019] Δx j (k) = x j (k+1)-x j (k), j = 1, 2, 3, ... n

[0020] In the formula, x j (k+1) represents the value of the jth optimization objective at time k+1; x j (k) represents the value of the jth optimization objective at time k.

[0021] In one embodiment, the value function of the multi-objective optimization is j optimization objective functions The weighted value of multi-objective optimization value function J i The expression of (k) is:

[0022]

[0023] In the formula, λ j is the weight factor of the jth optimization objective;

[0024] The optimization objective function is:

[0025]

[0026] Where, d i (k) is the switch action point, represents the system reference value of the jth optimization objective, x j (k+d i (k)) indicates that the jth optimization goal is k+d i (k) The value at time x j (k+d i (k)) and x j (k+1) can be calculated by duty cycle d i (k) represents, the expression is:

[0027] x j (k+d i (k)) = x j (k)+Δx j (k-1)d i (k), j = 1, 2, 3, ..., n

[0028] x j (k+1)=x j (k+d i (k))+Δx j (k)(1-d i (k)), j = 1, 2, 3, ..., n

[0029] In the formula, Δx j (k-1) represents the change of the jth optimization target from time k-1 to time k; Δx j (k) represents the change of the jth optimization objective from time k to time k+1.

[0030] In one embodiment, the multi-objective optimization value function is constructed to solve the optimal switch action position and optimal switch state at the current k moment, including:

[0031] The multi-objective optimization value function is constructed to solve the optimal switch action position and optimal switch state at the current k moment, including:

[0032] Let the value function J of multi-objective optimization be i The derivative of the switch action position in (k) is 0, and the switch state V corresponding to the switching state of the previous control cycle is solved.i The optimal duty cycle d of (k) opt (k);

[0033] The expression for duty cycle is:

[0034] in,

[0035]

[0036] In the formula, y i (k) and x i (k) does not indicate a specific meaning, but only represents a symbol; represents the reference value of the jth optimization objective; x j (k-1)x j (k) represents the value of the jth optimization objective at time k-1 and time k; Δx j (k-1) represents the change of the jth optimization target from time k-1 to time k; Δx ji (k) represents the change of the jth optimization target from time k to time k+1 in the future when the i-th candidate vector is applied;

[0037] Substitute all candidate vectors into x i (k) and y i (k), obtain the voltage vector V corresponding to the i-th switch state i (k) and duty cycle d i (k);

[0038] By minimizing the value function J of multi-objective optimization i (k), to obtain the optimal duty cycle d opt (k) and the optimal switching state V opt (k).

[0039] In one implementation, a two-step prediction method is used to compensate for the time delay.

[0040] In one embodiment, at time k+1 of the system, the optimal start point d(k) considering delay compensation is introduced into the value function by the expression of the optimization target at time k+1, and the expression of the jth optimization target at time k+1 is:

[0041] x j (k+1)=x j (k)+Δx j (k)d opt (k)+Δx j (k+1)(1-d opt (k)), j = 1, 2, ..., n

[0042] In the formula, Δxj (k) represents the change of the jth optimization target from time k to time k+1;

[0043] d opt (k) represents the optimal duty cycle at time k;

[0044] Δx j (k+1) represents the change of the jth optimization target from time k+1 to time k+2.

[0045] In one embodiment, the value function considering compensation is:

[0046]

[0047] In the formula, λ j is the weight factor of the jth optimization objective; When the i-th candidate vector acts, the optimization target x j The value function of J i (k+1) represents the value function of multi-objective optimization, which is the value function of all optimization objectives when the i-th candidate vector acts. The sum.

[0048] In a second aspect, the present invention provides a finite control set model predictive control device with high control accuracy, the device comprising:

[0049] An acquisition module is used to obtain the state space equation of the control object according to the system optimization target, and write the state space equation into a discrete form to obtain the increment of the optimization target from time k to k+1;

[0050] The value function module of multi-objective optimization is used to calculate the value of x from k to k+1 according to the optimization target. j The peak value and the error at time k+1 are used to construct the value function of multi-objective optimization of the system at time k;

[0051] The calculation module 1 is used to solve the optimal switch action position and optimal switch state at the current k moment by minimizing the value function of the constructed multi-objective optimization;

[0052] The second calculation module compensates for the delay time and constructs a value function that takes compensation into account. By minimizing the value function that takes compensation into account, the optimal switch action position and optimal switch state at time k+1 are obtained.

[0053] In a third aspect, the present invention provides a computer program product, comprising a computer program / instruction, which, when executed by a processor, implements the above-mentioned finite control set model predictive control method with high control accuracy.

[0054] In a fourth aspect, the present invention provides an electronic device comprising a processor and a memory; wherein, when the processor executes a computer program stored in the memory, the above-mentioned finite control set model predictive control method with high control accuracy is implemented.

[0055] Compared with the prior art, the present invention has the following advantages:

[0056] In the field of power electronics and power transmission, the switch opening point position of the existing FCS-MPC technology is fixed, which leads to a decrease in the quality of the converter output waveform. The high-control-precision FCS-MPC switch opening point disclosed by the present invention is not fixed, and the switch opening position is a variable. The optimal converter switch action switching position can be obtained through online optimization. The existing FCS-MPC technology does not have this feature. The present invention can effectively reduce the total harmonic distortion (THD) of the output sine wave, so that the output quality of the inverter is significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0057] The accompanying drawings are part of the present invention and are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but do not constitute an improper limitation of the present invention. Obviously, the accompanying drawings described below are only some embodiments. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without creative work.

[0058] Figure 1 A flow chart of a finite control set model predictive control method with high control accuracy provided by one embodiment of the present invention;

[0059] Figure 2 It is the space voltage vector diagram of the midpoint clamped three-level inverter;

[0060] Figure 3 The implementation principle comparison between the high control precision finite control set model predictive control method of the present invention and the traditional FCS-MPC method is provided;

[0061] Figure 4 It is an implementation principle diagram of the delay compensation of the high control precision finite control set model predictive control method of the present invention;

[0062] Figure 5 The experimental waveforms of the phase currents of the three-level grid-connected inverter using the high control accuracy finite control set model predictive control method of the present invention and the FFT analysis results thereof;

[0063] Figure 6 The experimental waveforms of phase currents of a three-level grid-connected inverter using traditional FCS-MPC and its FFT analysis results;

[0064] Figure 7The active power, reactive power and midpoint potential experimental waveforms of a three-level grid-connected inverter using the high control accuracy finite control set model predictive control method of the present invention;

[0065] Figure 8 The experimental waveforms of active power, reactive power and midpoint potential of a three-level grid-connected inverter using traditional FCS-MPC.

[0066] It should be noted that these drawings and textual descriptions are not intended to limit the conceptual scope of the present invention in any way, but are intended to illustrate the concept of the present invention for those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0067] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0068] like Figure 1 As shown, the embodiment of the present disclosure provides a high control accuracy finite control set model predictive control method, which is applied to the high control accuracy finite control set model predictive control of a three-phase neutral point clamped three-level grid-connected inverter. The method specifically includes the following steps:

[0069] Step S100: Obtain the state space equation of the controlled object according to the system optimization target, and write the state space equation into a discrete form to obtain the change of the optimization target from time k to k+1.

[0070] Furthermore, according to the system optimization goal, the state space equation of the control object is obtained, and the expression is:

[0071]

[0072] Where x represents the state space, u represents the system input, A represents the state transfer matrix, and B is the input matrix.

[0073]

[0074] The discrete time is t s , then the jth optimization target x j The discrete form of is:

[0075] x j (k+1)=(t s a j +1)x j (k)+t s bj u(k), j=1, 2, 3..., n (3)

[0076] The change Δx of the jth optimization target from time k to time k+1 j (k) can be expressed as:

[0077] Δx j (k) = x j (k+1)-x j (k), j = 1, 2, 3...n (4)

[0078] In the formula, x j (k+1) represents the value of the jth optimization objective at time k+1; x j (k) represents the value of the jth optimization objective at time k.

[0079] Step S200: According to the optimization target, x from time k to k+1 j The peak value and the error at time k+1 are used to construct the value function of multi-objective optimization of the system at time k.

[0080] The value function of multi-objective optimization is j optimization objective functions The weighted value of j is the weight factor of the jth optimization objective, the value function J i The general form of the expression of (k) can be expressed as:

[0081]

[0082] In the formula, λ j is the weight factor of the jth optimization objective;

[0083] Furthermore, we optimize the objective function It is expressed as:

[0084]

[0085] Where, d i (k) is the switch action point, represents the system reference value of the jth optimization objective, x j (k+d i (k)) indicates that the jth optimization goal is k+d i (k) The value at time x j (k+d i (k)) and x j (k+1) can be calculated by duty cycle d i (k) represents, the expression is:

[0086] x j (k+d i(k)) = x j (k)+Δx j (k-1)d i (k), j=1, 2, 3..., n (7)

[0087] x j (k+1)=x j (x+d i (k))+Δx j (k)(1-d i (k)), j=1, 2, 3..., n (8)

[0088] In the formula, Δx j (k-1) represents the change of the jth optimization target from time k-1 to time k;

[0089] Δx j (k) represents the change of the jth optimization objective from time k to time k+1.

[0090] Step S300: By minimizing the constructed multi-objective optimization value function, the optimal switch action position and optimal switch state at the current k moment are solved.

[0091] In the embodiment of the present application, the constructed multi-objective optimization value function is minimized to solve the optimal switch action position and optimal switch state at the current k moment, including:

[0092] Minimize the value function J i (k) By order J i (k) About switch action position d i The derivative of (k) is 0, and the switch state V corresponding to the switching state of the previous control cycle is solved i (k)(e.g. Figure 2 As shown in Figure 2, the three-phase two-level inverter contains 8 switching states, and the three-phase three-level inverter contains 27 switching states. opt (k), can be expressed as:

[0093]

[0094] Specifically, let the optimization objective function J i The general solution form of the derivative of the switch action position in (k) is 0. Here, there needs to be a constraint that the duty cycle is [0,1], so when it is out of range, it is equal to 0 or 1; when the general solution is in the range [0,1], the expression is It is represented as the switch position calculated by the i-th candidate vector, that is, the duty cycle. The solution process is: Let That is, the partial derivative of the value function J with respect to the optimal duty cycle (switch position) is 0, and d can be obtained through the first-order derivative expression:i The general solution expression of (k).

[0095] Furthermore, the duty cycle d i (k) Expression, y calculated for the i-th candidate switch state i The general expression for (k) is:

[0096]

[0097] The x calculated for the i-th candidate switch state i The general expression of (k) is:

[0098]

[0099] Among them, y i (k) and x i (k) does not indicate a specific meaning, but only represents a symbol; represents the reference value of the jth optimization objective; x j (k-1)x j (k) represents the value of the jth optimization objective at time k-1 and time k; Δx j (k-1) represents the change of the jth optimization target from time k-1 to time k; Δx ji (k) represents the change of the jth optimization target from time k to time k+1 in the future when the i-th candidate vector is applied;

[0100] Substitute all possible candidate vectors into x i (k) and y i (k), the voltage vector V corresponding to the i-th switch state can be obtained i (k) and duty cycle (switch actuation position) d i (k), by minimizing the value function J i (k) The optimal duty cycle (switching position) d can be obtained opt (k) and the optimal switching state V opt (k).

[0101] Step S400: Compensate for the delay time and construct a value function that takes compensation into account, and obtain the optimal switch action position and optimal switch state at time k+1 by minimizing the value function that takes compensation into account.

[0102] Since the real-time controller introduces delay, a two-step prediction strategy is used to compensate for the time delay.

[0103] Furthermore, at the time k+1 of the system, the optimal starting point d(k) of delay compensation is considered to introduce the value function through the expression of the optimization target at the time k+1. The expression of the jth optimization target at the time k+1 is:j (k+1)=x j (k)+Δx j (k)d opt (k)+Δx j (k+1)(1-d opt (k)),j=1,2,..,n(12)

[0104] Optimal switching state considering delay compensation V opt (k+1) and the optimal switch position d opt (k+1) will be compensated by minimizing the value function J i (k+1) implementation.

[0105] Further, consider the compensation value function J i The general expression form of (k+1) is:

[0106]

[0107] In the formula, λ j is the weight factor of the jth optimization objective; When the i-th candidate vector acts, the optimization target x j The value function of J i (k+1) represents the value function of multi-objective optimization, which is the value function of all optimization objectives when the i-th candidate vector acts. The sum.

[0108] In a specific embodiment, the system optimization objectives include active power p, reactive power q and midpoint potential u o .

[0109] The expressions of the optimization target active power p and reactive power q are:

[0110]

[0111] Midpoint potential u o The expression is:

[0112]

[0113] The value function of multi-objective optimization is 3 optimization objective functions The weighted value of j is the weight factor of the jth optimization objective, the value function J i The expression of (k) can be expressed as:

[0114] J i (k) = λ1j 1i (k)+λ2j 2i (k)+λ3j3i (k) (16)

[0115] like Figure 3 As shown, the optimization objective function is the expression of the active power, reactive power and peak value of the midpoint potential from time k to time k+1, and the error at time k+1, which is:

[0116]

[0117] Wherein, p* represents the reference value of active power, and q* represents the reference value of reactive power;

[0118]

[0119] d i (k) is the duty cycle, i.e. the switch action point.

[0120] The optimal switch action position d at the current time k opt (k) and the optimal switching state V opt The solution of (k) is to minimize the value function J i (k) Implementation.

[0121] Minimize the value function J i (k) By order J i (k) About switch action position d i The derivative of (k) is 0, and the switch state V corresponding to the switching state of the previous control cycle is solved i (k)(e.g. Figure 2 As shown, the three-phase three-level inverter contains 27 switching states, i = 1, 2, 3..., 27) with the optimal duty cycle d opt (k). Duty cycle d i (k) The y calculated for the i-th candidate switch state in the expression i (k) and x i The general expression of (k) is:

[0122]

[0123] Substitute the 27 candidate vectors into x i (k) and y i (k), the voltage vector V corresponding to the i-th switch state can be obtained i (k) and switch actuation position d i (k), by minimizing the value function J i (k) The optimal switch action position d can be obtained opt (k) and the optimal switching state V opt (k).

[0124] like Figure 4 As shown in Figure 2, due to the delay introduced by the real-time controller, a two-step prediction strategy is used to compensate for the time delay. At time k+1 of the system, the expression of the optimization objective is rewritten to include the optimal start point d opt The expression of (k) is:

[0125]

[0126] Optimal switching state considering delay compensation V opt (k+1) and the optimal switch position d opt (k+1) will be obtained by minimizing the value function J i (k+1) realization, value function J i (k+1) is written as:

[0127] J i (k+1)=λ1j 1i (k+1)+λ2j 2i (k+1)+λ3j 3i (k+1) (22)

[0128] Figure 5-8 The experimental waveforms and analysis results of the three-phase three-level grid-connected inverter under the same working conditions using the high control accuracy finite control set model predictive control method of the present invention and the traditional FCS-MPC method are given. Figure 5 As shown, the phase current using the method disclosed in the present invention has a lower THD value of 3.63%; Figure 6 It can be seen that when the traditional FCS-MPC method is used, the output phase current THD is 3.81%, indicating that compared with the traditional FCS-MPC, the method of the present invention has a better control effect. Figure 7 As shown, the active power fluctuation range of the inverter using the method disclosed in the present invention is 175W, the reactive power fluctuation range is 140W, and the midpoint potential fluctuation is 6V; refer to Figure 8 As shown, when the traditional FCS-MPC method is used, the active power fluctuation range of the inverter is 190W, the reactive power fluctuation range is 150W, and the midpoint potential fluctuation is 6V, indicating that compared with the traditional FCS-MPC method, the method of the present invention has higher control accuracy.

[0129] The traditional FCS-MPC method can only switch at a fixed position, and the switching of the switch state is only related to the predicted initial value, which reduces the control accuracy. The present invention discloses a finite control set model predictive control (FCS-MPC) method with high control accuracy, which can flexibly change the switch state switching position, and the switch switching moment is flexible and variable and adjustable online, which not only improves the switching accuracy of the inverter, but also retains the characteristic of no more than one switch per control cycle. It has the advantages of FCS-MPC's simple and intuitive design and easy to achieve multi-objective optimization through a unified value function. The design of real-time optimization and adjustment of the switch position by the present invention can improve the output waveform quality of the power electronic converter, and has high practical value and broad application prospects in the typical application scenarios of power electronics and power transmission such as grid connection and motor drive.

[0130] The following is an embodiment of a finite control set model predictive control device with high control accuracy of the present invention, which can be used to execute an embodiment of a finite control set model predictive control method with high control accuracy of the present invention. For details not disclosed in the embodiment of a finite control set model predictive control device with high control accuracy of the present invention, please refer to the embodiment of a finite control set model predictive control method with high control accuracy of the present invention.

[0131] In one embodiment, a finite control set model predictive control device with high control accuracy is proposed, the device comprising:

[0132] An acquisition module is used to obtain the state space equation of the control object according to the system optimization target, and write the state space equation into a discrete form to obtain the increment of the optimization target from time k to k+1;

[0133] The value function module of multi-objective optimization is used to calculate the value of x from k to k+1 according to the optimization target. j The peak value and the error at time k+1 are used to construct the value function of multi-objective optimization of the system at time k;

[0134] The calculation module 1 is used to solve the optimal switch action position and optimal switch state at the current k moment by minimizing the value function of the constructed multi-objective optimization;

[0135] The second calculation module compensates for the delay time and constructs a value function that takes compensation into account. By minimizing the value function that takes compensation into account, the optimal switch action position and optimal switch state at time k+1 are obtained.

[0136] It should be noted that the finite control set model predictive control device with high control accuracy provided in the above embodiment only uses the division of the above functional modules as an example when executing a finite control set model predictive control method with high control accuracy. In actual applications, the above functional distribution can be completed by different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. In addition, the finite control set model predictive control device with high control accuracy provided in the above embodiment and the embodiment of a finite control set model predictive control method with high control accuracy belong to the same concept. The implementation process thereof is detailed in the embodiment of a finite control set model predictive control method with high control accuracy, which will not be repeated here.

[0137] In one embodiment, a computer program product is provided. When the computer program / instruction is executed by a processor, one or more processors are caused to perform the following steps: obtain a state space equation of a control object according to a system optimization target, and write the state space equation into a discrete form to obtain an increment of the optimization target from time k to k+1; and calculate an increment of the optimization target from time k to k+1 according to the x of the optimization target from time k to k+1. j The peak value and the error at the k+1 moment are used to construct the value function of the multi-objective optimization of the system at the k moment; the optimal switch action position and the optimal switch state at the current k moment are solved by minimizing the constructed multi-objective optimization value function; the delay time is compensated to construct a value function that takes compensation into account, and the optimal switch action position and the optimal switch state at the k+1 moment are obtained by minimizing the value function that takes compensation into account.

[0138] In one embodiment, an electronic device is provided, wherein the computer device includes a processor and a memory; wherein the processor implements the following steps when executing a computer program stored in the memory: obtaining a state space equation of a control object according to a system optimization target, and writing the state space equation in a discrete form, obtaining an increment of the optimization target from time k to k+1; and calculating an increment of the optimization target from time k to time k+1 according to the x of the optimization target from time k to time k+1. j The peak value and the error at the k+1 moment are used to construct the value function of the multi-objective optimization of the system at the k moment; the optimal switch action position and the optimal switch state at the current k moment are solved by minimizing the constructed multi-objective optimization value function; the delay time is compensated to construct a value function that takes compensation into account, and the optimal switch action position and the optimal switch state at the k+1 moment are obtained by minimizing the value function that takes compensation into account.

[0139] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the aforementioned storage medium can be a non-volatile storage medium such as a disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

[0140] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0141] The above-mentioned embodiments only express several implementation methods of the present invention, and the description thereof is relatively specific and detailed, but it cannot be understood as limiting the scope of the patent of the present invention. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, which all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.

Claims

1. A finite control set model predictive control method with high control accuracy, characterized in that: The method comprises: The state space equation of the control object is obtained according to the system optimization target, and the state space equation is written in discrete form to obtain the change of the optimization target from time k to k+1; According to the optimization goal, x from time k to k+1 j The peak value and the error at time k+1 are used to construct the value function of multi-objective optimization of the system at time k; By minimizing the value function of the constructed multi-objective optimization, the optimal switch action position and optimal switch state at the current k moment are solved; The delay time is compensated by constructing a value function that takes compensation into account, and the optimal switch action position and optimal switch state at time k+1 are obtained by minimizing the value function that takes compensation into account.

2. The finite control set model predictive control method with high control accuracy according to claim 1 is characterized in that: The general form of the state space equation of the control object obtained according to the system optimization target is: In the formula, x represents the state space, u represents the system input, A represents the state transfer matrix, and B is the input matrix; The discrete time is t s , then the jth optimization target x j The discrete form of is: x j (k+1)=(t s a j +1)x j (k)+t s b j u(k),j=1,2,3...,n The change Δx of the jth optimization target from time k to time k+1 j (k) can be expressed as: Δx j (k)=x j (k+1)-x j (k),j=1,2,3...n In the formula, x j (k+1) represents the value of the jth optimization objective at time k+1; x j (k) represents the value of the jth optimization objective at time k.

3. The finite control set model predictive control method with high control accuracy according to claim 2 is characterized in that: The value function of the multi-objective optimization is j optimization objective functions The weighted value of multi-objective optimization value function J i The expression of (k) is: In the formula, λ j is the weight factor of the jth optimization objective; Optimizing the objective function for: Where, d i (k) is the switch action point, represents the system reference value of the jth optimization objective, x j (k+d i (k)) indicates that the jth optimization goal is k+d i (k) The value at time x j (k+d i (k)) and x j (k+1) can be calculated by duty cycle d i (k) represents, the expression is: x j (k+d i (k))=x j (k)+Δx j (k-1)d i (k),j=1,2,3...,n x j (k+1)=x j (k+d i (k))+Δx j (k)(1-d i (k)),j=1,2,3...,n In the formula, Δx j (k-1) represents the change of the jth optimization target from time k-1 to time k; Δx j (k) represents the change of the jth optimization objective from time k to time k+1.

4. The finite control set model predictive control method with high control accuracy according to claim 1 is characterized in that: The multi-objective optimization value function is constructed to solve the optimal switch action position and optimal switch state at the current k moment, including: Let the value function J of multi-objective optimization be i The derivative of the switch action position in (k) is 0, and the optimal duty cycle d corresponding to the switch state Vi(k) of the previous control cycle is solved. opt (k); The expression for duty cycle is: in, In the formula, y i (k) and x i (k) does not indicate a specific meaning, but only represents a symbol; represents the reference value of the jth optimization objective; x j (k-1)x j (k) represents the value of the jth optimization objective at time k-1 and time k; Δx j (k-1) represents the change of the jth optimization target from time k-1 to time k; Δx ji (k) represents the change of the jth optimization target from time k to time k+1 in the future when the i-th candidate vector is applied; Substitute all candidate vectors into x i (k) and y i (k), obtain the voltage vector V corresponding to the i-th switch state i (k) and duty cycle d i (k); By minimizing the value function J of multi-objective optimization i (k), to obtain the optimal duty cycle d opt (k) and the optimal switching state V opt (k).

5. The finite control set model predictive control method with high control accuracy according to claim 1 is characterized in that: A two-step prediction method is used to compensate for the time delay.

6. The finite control set model predictive control method with high control accuracy according to claim 5 is characterized in that: At time k+1 in the system, the optimal start point d considering delay compensation is opt (k) The value function is introduced by the expression of the optimization target at time k+1. The expression of the jth optimization target at time k+1 is: x j (k+1)=x j (k)+Δx j (k)d opt (k)+Δx j (k+1)(1-d opt (k)),j=1,2,...,n In the formula, Δx j (k) represents the change of the jth optimization target from time k to time k+1, d opt (k) represents the optimal duty cycle at time k; Δx j (k+1) represents the change of the jth optimization objective from time k+1 to time k+2.

7. The finite control set model predictive control method with high control accuracy according to claim 1 is characterized in that: The value function considering compensation is: In the formula, λ j is the weight factor of the jth optimization objective; When the i-th candidate vector acts, the optimization target x j The value function of J i (k+1) represents the value function of multi-objective optimization, which is the value function of all optimization objectives when the i-th candidate vector acts. The sum.

8. A finite control set model predictive control device with high control accuracy, characterized in that: The device comprises: An acquisition module is used to obtain the state space equation of the control object according to the system optimization target, and write the state space equation into a discrete form to obtain the increment of the optimization target from time k to k+1; The value function module of multi-objective optimization is used to calculate the value of x from k to k+1 according to the optimization target. j The peak value and the error at time k+1 are used to construct the value function of multi-objective optimization of the system at time k; The calculation module 1 is used to solve the optimal switch action position and optimal switch state at the current k moment by minimizing the value function of the constructed multi-objective optimization; The second calculation module compensates for the delay time and constructs a value function that takes compensation into account. By minimizing the value function that takes compensation into account, the optimal switch action position and optimal switch state at time k+1 are obtained.

9. A computer program product comprising a computer program / instructions, characterized in that: When the computer program / instructions are executed by a processor, the finite control set model predictive control method with high control accuracy as described in any one of claims 1 to 7 is implemented.

10. An electronic device, characterized in that: It comprises a processor and a memory; wherein, when the processor executes the computer program stored in the memory, it implements the finite control set model predictive control method with high control accuracy as described in any one of claims 1 to 7.

Images