Two-step model-free predictive control method and device for three-level inverter

Abstract

The invention discloses a two-step model-free predictive control method and device for a three-level inverter in the technical field of power electronics. The method comprises the following steps: constructing an over-limit equation of the three-level inverter through Clark transformation; estimating the over-confinement equation based on a neural network, and constructing a discrete time model about an approximate Euler formula; performing two-step prediction based on the time model, and constructing an evaluation function; traversing the candidate quantity of the alternating current side voltage according to the evaluation function, and taking the switching state corresponding to the candidate quantity with the minimum value of the evaluation function as a control signal of the three-level inverter at the next moment; according to the method, the neural network predictor is adopted, the method does not depend on an accurate mathematical model and system parameters, the accuracy of current tracking is effectively improved, and the robustness of the system is enhanced. The optimization range is expanded to two control periods by adopting a two-step prediction structure, and the probability of applying the same voltage vector in the continuous control period is improved, so that the switching frequency is remarkably reduced.

Description

Technical Field

[0001] The present invention relates to the field of power electronics technology, and in particular to a two-step model-free predictive control method and device for a three-level inverter. Background Art

[0002] With the rapid development of the national economy and the continuous upgrading of the industrial structure, the power system, especially the distribution network, is facing the challenges brought by the growing electricity demand and unbalanced grid design. The current distribution network has many limitations in terms of structural rationality and dispatching means, which affects its operational flexibility and reliability. In this context, the application of flexible interconnection devices has become an important technical means to optimize the topological structure of the distribution network, significantly improving the operational flexibility and control capabilities of the network. The introduction of this technology provides effective support for the closed-loop operation of the distribution network and improving the overall efficiency.

[0003] Finite control set model predictive control (FCS-MPC) has been widely used in inverter and drive control due to its simple design and implementation, no need for complex modulation process, and direct realization of optimal control input. Although the FCS-MPC method has shown obvious advantages, its performance is still constrained by many factors. First, during system operation, load parameters tend to fluctuate around their nominal values, and in some cases, there is a lack of accurate information about the actual load state. To address these problems, the model-free predictive control (MFPC) method provides a new solution. However, although the MFPC method can cope with some challenges, it does not solve the system loss problem caused by excessive switching frequency (SF). In order to balance the contradiction between switching frequency and control performance, some studies have proposed multi-step predictive control methods. However, the inherent characteristics of multi-step predictive control make the control cycle longer, resulting in a large contradiction in this method itself. Summary of the invention

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art and provide a two-step model-free predictive control method and device for a three-level inverter to solve the defects of the prior art in the background technology, thereby improving the stability and robustness of the system and reducing system losses by reducing the switching frequency.

[0005] To achieve the above object, the present invention is implemented by adopting the following technical solutions: In a first aspect, the present invention provides a two-step model-free predictive control method for a three-level inverter, comprising: Constructing the super-localized equation of the three-level inverter through Clarke transformation; The super-confined equation is estimated based on a neural network, and a discrete time model for the approximate Euler formula is constructed; Perform two-step prediction based on the time model and construct an evaluation function; The candidate quantities of the AC side voltage are traversed according to the evaluation function, and the switch state corresponding to the candidate quantity with the smallest evaluation function value is taken as the control signal of the three-level inverter at the next moment.

[0006] Optionally, the three-level inverter includes three bridge arms connected in parallel, and the bridge arm includes a switch tube connected in series between the positive and negative electrodes of the DC side bus. , switch tube , switch tube And switch tube ; The switch tubes of the three bridge arms And switch tube The common end is connected to the three phases of the AC side bus respectively.

[0007] Optionally, a voltage dividing capacitor is connected in series between the positive and negative poles of the DC busbar. And the voltage divider capacitor The bridge arm also includes a switch tube connected in series and the switch tube The common terminal of the switch tube and the switch tube The diode between the common terminals and diode , the voltage divider capacitor and the voltage divider capacitor The common terminal and the diode and the diode The common end is connected to the midpoint potential.

[0008] Optionally, the switch tube of the bridge arm And switch tube A filter inductor L and a line resistor R are also connected in series between the common end and the AC side bus.

[0009] Optionally, the super-local equation of the three-level inverter is: ; ; In the formula, is the unknown equation, is the AC side current of the three-level inverter Axis component, is the AC side voltage of the three-level inverter Axis component, is the value of the filter inductor.

[0010] Optionally, estimating the super-localized equation based on a neural network and constructing a discrete time model for the approximate Euler formula includes: For the unknown equation Estimation using a neural network: ; In the formula, is the activation function, , is the output weight matrix, is the estimation error; Build the state predictor: ; In the formula, for The estimated value of for The rate of change, is a positive constant; The update rule is: ; In the formula, is a positive constant; Construct a discrete time model that approximates Euler's formula: ; ; ; ; In the formula, is the sampling time, is the sampling period.

[0011] Optionally, performing two-step prediction based on the time model and constructing an evaluation function includes: Perform a two-step forecast on the temporal model: ; ; ; ; Calculate the AC side reference voltage vector based on the two-step prediction results: ; ; An evaluation function is constructed based on the reference voltage vector: ; In the formula, , For the The AC side voltage at a sampling moment.

[0012] In a second aspect, the present invention provides a two-step model-free predictive control device for a three-level inverter, comprising: A super-localized equation construction module, configured to construct a super-localized equation of the three-level inverter through Clarke transformation; A time model building module is configured to estimate the super-localized equation based on a neural network and build a time model for the discrete approximate Euler formula; An evaluation function construction module is configured to perform two-step prediction based on the time model and construct an evaluation function; The control signal acquisition module is configured to traverse the candidate quantities of the AC side voltage according to the evaluation function, and take the switch state corresponding to the candidate quantity with the smallest evaluation function value as the control signal of the three-level inverter at the next moment.

[0013] In a third aspect, the present invention provides an electronic device, characterized in that it includes a processor and a storage medium; The storage medium is used to store instructions; The processor is used to operate according to the instructions to execute the steps according to the above method.

[0014] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that the program implements the steps of the above method when executed by a processor.

[0015] Compared with the prior art, the present invention has the following beneficial effects: The present invention provides a two-step model-free predictive control method and device for a three-level inverter, which adopts a neural network predictor, a model-free prediction method. The method does not rely on accurate mathematical models and system parameters, thereby effectively improving the accuracy of current tracking and enhancing the robustness of the system. In addition, the two-step prediction structure is used to extend the optimization range to two control cycles. This strategy increases the probability of applying the same voltage vector in consecutive control cycles, thereby helping to significantly reduce the switching frequency and reduce system losses. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a circuit diagram of a three-level inverter provided by an embodiment of the present invention; Figure 2 1 is a flow chart of a two-step model-free predictive control method for a three-level inverter provided by an embodiment of the present invention; Figure 3 is a voltage vector diagram of a three-level inverter provided by an embodiment of the present invention; Figure 4 is a waveform diagram of the output side current and midpoint potential voltage of the inverter of the control method of this embodiment under parameter mismatch provided by the embodiment of the present invention; Figure 5 is a schematic diagram of the output side current and midpoint potential voltage of an inverter of a conventional control method under parameter mismatch provided by an embodiment of the present invention; Figure 6 It is a schematic diagram of the THD value and switching frequency of the output current under different matching conditions provided by the embodiment of the present invention under the operation of the traditional control method and the control method of this embodiment. DETAILED DESCRIPTION

[0017] The present invention will be further described below in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and cannot be used to limit the protection scope of the present invention.

[0018] Embodiment 1:

[0019] like Figure 1 As shown, the present invention provides a three-level inverter, which adopts a three-level neutral point closed (3L-NPC) inverter, including three bridge arms connected in parallel, and the bridge arm includes a switch tube connected in series between the positive and negative electrodes of the DC side bus. , switch tube , switch tube And switch tube ; The switch tubes of the three bridge arms And switch tube The common end is connected to the three phases of the AC bus. A voltage dividing capacitor is also connected in series between the positive and negative poles of the DC bus. And the voltage divider capacitor The bridge arm also includes a switch tube connected in series And switch tube The common terminal and the switch tube And switch tube The diode between the common terminals and diode , voltage divider capacitor And the voltage divider capacitor The common terminal and diode and diode The common end of the bridge arm is connected to the middle point potential. And switch tube A filter inductor L and a line resistor R are also connected in series between the common end and the AC side bus.

[0020] like Figure 2 As shown, based on the above three-level inverter, the present invention provides a two-step model-free predictive control method for a three-level inverter, comprising the following steps: Step S1: construct a super-localized equation of a three-level inverter through Clarke transformation.

[0021] The super-localized equation of the three-level inverter is:

[0022]

[0023] In the formula, is the unknown equation, is the AC side current of the three-level inverter Axis component, is the AC side voltage of the three-level inverter Axis component, is the value of the filter inductor.

[0024] Step S2: estimating the super-localized equation based on the neural network and constructing a discrete time model for the approximate Euler formula.

[0025] The construction process specifically includes: Step S2.1: For the unknown equation Estimation using a neural network: ; In the formula, is the activation function, , is the output weight matrix, is the estimation error; Step S2.2: Construct a state predictor: ; In the formula, for The estimated value of for The rate of change, is a positive constant; The update rule is: ; In the formula, is a positive constant; Step S2.3: Construct a discrete time model approximating the Euler formula: ; ; ; ; In the formula, is the sampling time, is the sampling period.

[0026] Step S3: Perform two-step prediction based on the time model and construct an evaluation function.

[0027] The construction process specifically includes: Make two-step predictions on the temporal model: ; ; ; ; Calculate the AC side reference voltage vector based on the two-step prediction results: ; ; Construct an evaluation function based on the reference voltage vector: ; In the formula, , For the The AC side voltage at a sampling moment.

[0028] Step S4, traverse the candidate quantities of the AC side voltage according to the evaluation function, and take the switch state corresponding to the candidate quantity with the smallest evaluation function value as the control signal of the three-level inverter at the next moment.

[0029] like Figure 3 As shown, the candidate quantity of the AC side voltage can be determined by the voltage vector diagram of the inverter. The switching states of the switch tubes on the three bridge arms can be directly known from the diagram, and the control signals of the corresponding switch tubes are generated through the switching states.

[0030] like Figure 4 and Figure 5 As shown, the two-step model-free predictive control method provided in this embodiment, under parameter mismatch (the inductance of the filter inductor L becomes 1 / 2 of the original), the output side current and the midpoint potential voltage of the inverter, it can be seen that the two-step model-free predictive control method provided in this embodiment effectively improves the accuracy of current tracking and enhances the robustness of the system.

[0031] like Figure 6 As shown, the two-step model-free predictive control method provided in this embodiment significantly reduces the switching frequency, and at the same time, the total harmonic distortion THD value of the output current is smaller.

[0032] Embodiment 2:

[0033] An embodiment of the present invention provides a two-step model-free predictive control device for a three-level inverter, comprising: A super-localized equation construction module, configured to construct a super-localized equation of a three-level inverter through Clarke transformation; A time model building module is configured to estimate the super-localized equation based on a neural network and build a time model for the discrete approximate Euler formula; An evaluation function building module is configured to perform two-step prediction based on the time model and build an evaluation function; The control signal acquisition module is configured to traverse candidate quantities of the AC side voltage according to the evaluation function, and take the switch state corresponding to the candidate quantity with the smallest evaluation function value as the control signal of the three-level inverter at the next moment.

[0034] Embodiment three:

[0035] Based on the two-step model-free predictive control method provided in the first embodiment, an embodiment of the present invention provides an electronic device, characterized in that it includes a processor and a storage medium; The storage medium is used to store instructions; The processor is used to operate according to the instructions to execute the steps according to the above method.

[0036] Embodiment 4:

[0037] Based on the two-step model-free predictive control method provided in Example 1, an embodiment of the present invention provides a computer-readable storage medium on which a computer program is stored, characterized in that the program implements the steps of the above method when executed by a processor.

[0038] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0039] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1A device that provides the functions specified in a block or multiple blocks.

[0040] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0041] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0042] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A two-step model-free predictive control method for a three-level inverter, characterized in that: include: Constructing the super-localized equation of the three-level inverter through Clarke transformation; The super-confined equation is estimated based on a neural network, and a discrete time model for the approximate Euler formula is constructed; Perform two-step prediction based on the time model and construct an evaluation function; The candidate quantities of the AC side voltage are traversed according to the evaluation function, and the switch state corresponding to the candidate quantity with the smallest evaluation function value is taken as the control signal of the three-level inverter at the next moment.

2. The two-step model-free predictive control method for a three-level inverter according to claim 1, characterized in that: The three-level inverter includes three bridge arms connected in parallel, each bridge arm includes a switch tube connected in series between the positive and negative electrodes of the DC bus. , switch tube , switch tube And switch tube ; The switch tubes of the three bridge arms And switch tube The common end is connected to the three phases of the AC side bus respectively.

3. The two-step model-free predictive control method for a three-level inverter according to claim 2, characterized in that: A voltage dividing capacitor is also connected in series between the positive and negative poles of the DC busbar. And the voltage divider capacitor The bridge arm also includes a switch tube connected in series and the switch tube The common terminal of the switch tube and the switch tube The diode between the common terminals and diode , the voltage divider capacitor and the voltage divider capacitor The common terminal and the diode and the diode The common end is connected to the midpoint potential.

4. The two-step model-free predictive control method for a three-level inverter according to claim 2, characterized in that: The switch tube of the bridge arm And switch tube A filter inductor L and a line resistor R are also connected in series between the common end and the AC side bus.

5. The two-step model-free predictive control method for a three-level inverter according to claim 4, characterized in that: The super-local equation of the three-level inverter is: ； ； In the formula, is the unknown equation, is the AC side current of the three-level inverter Axis component, is the AC side voltage of the three-level inverter Axis component, is the value of the filter inductor.

6. The two-step model-free predictive control method for a three-level inverter according to claim 5, characterized in that: The estimating the super-localized equation based on the neural network and constructing a discrete time model about the approximate Euler formula includes: For the unknown equation Estimation using a neural network: ； In the formula, is the activation function, , is the output weight matrix, is the estimation error; Build the state predictor: ； In the formula, for The estimated value of for The rate of change, is a positive constant; The update rule is: ； In the formula, is a positive constant; Construct a discrete time model that approximates Euler's formula: ； ； ； ； In the formula, is the sampling time, is the sampling period.

7. The two-step model-free predictive control method for a three-level inverter according to claim 6, characterized in that: The two-step prediction based on the time model and the construction of the evaluation function include: Perform a two-step forecast on the temporal model: ； ； ； ； Calculate the AC side reference voltage vector based on the two-step prediction results: ； ； An evaluation function is constructed based on the reference voltage vector: ； In the formula, , For the The AC side voltage at the sampling moment.

8. A two-step model-free predictive control device for a three-level inverter, characterized in that: include: A super-localized equation construction module, configured to construct a super-localized equation of the three-level inverter through Clarke transformation; A time model building module is configured to estimate the super-localized equation based on a neural network and build a time model for the discrete approximate Euler formula; An evaluation function construction module is configured to perform two-step prediction based on the time model and construct an evaluation function; The control signal acquisition module is configured to traverse the candidate quantities of the AC side voltage according to the evaluation function, and take the switch state corresponding to the candidate quantity with the smallest evaluation function value as the control signal of the three-level inverter at the next moment.

9. An electronic device, characterized in that: including processor and storage medium; The storage medium is used to store instructions; The processor is configured to operate according to the instructions to execute the steps of the method according to any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images