CN-121984358-A - Double-vector model-free prediction control method and related device for simplified matrix converter

CN121984358ACN 121984358 ACN121984358 ACN 121984358ACN-121984358-A

Abstract

The invention belongs to the technical field of power electronics, and discloses a double-vector model-free predictive control method of a simplified matrix converter and a related device, wherein the method comprises the steps of obtaining alternating-current side voltage, direct-current side voltage and direct-current side reference voltage data of the simplified matrix converter at a target moment, and calculating an inner ring reference current; the method comprises the steps of obtaining alternating current side currents of a simplified matrix converter at adjacent moments and input side currents of a switch, calculating system gain and system centralized disturbance of the simplified matrix converter at target moments, combining the obtained direct current side currents of the simplified matrix converter at the target moments, calculating alternating current side currents at the next moment corresponding to each switch state, calculating action time corresponding to an optimal vector and a suboptimal vector according to inner ring reference currents and the alternating current side currents at the next moment, and determining a wave-transmitting mode. The invention realizes the model-free predictive control of the RMC, improves the immunity of the RMC, and reduces the current ripple of the alternating current side through double-vector fixed-frequency control.

Inventors

Dang Chaoliang
ZHAI JIAHAO
SONG WEIZHANG

Assignees

西安理工大学

Dates

Publication Date: 20260505
Application Date: 20260409

Claims (10)

1. The double-vector model-free prediction control method for the reduced matrix converter is characterized by comprising the following steps of: Obtaining alternating-current side voltage, direct-current side voltage and direct-current side reference voltage data of the simplified matrix converter at a target moment, and calculating an inner ring reference current; Acquiring alternating current side current and switching input side current of a simplified matrix converter at adjacent moments, and calculating system gain and system centralized disturbance of the simplified matrix converter at a target moment, wherein the adjacent moments comprise the target moment, the last moment and the last moment; Obtaining direct current side current of the simplified matrix converter at the target moment, calculating switch input side current corresponding to each switch state, and calculating alternating current side current at the next moment corresponding to each switch state by combining system gain and system concentrated disturbance of the simplified matrix converter at the target moment; And calculating the action time corresponding to the optimal vector and the suboptimal vector according to the inner ring reference current and the alternating current side current at the next moment, and determining a wave generation mode according to the magnitude relation between the action time of the optimal vector and the switching period and the magnitude relation between the action time of the suboptimal vector and the switching period in one switching period.
2. The method for controlling double vector model-free predictive control of a reduced matrix converter according to claim 1, wherein the method for calculating the inner loop reference current is as follows: Obtaining a voltage error according to direct-current side voltage and direct-current side reference voltage data of the reduced matrix converter at a target moment; obtaining a reference value of the inner ring active power according to the voltage error, the output value of the PID controller in the reduced matrix converter and the direct current side voltage; calculating a complex power reference value according to the reference value of the inner ring active power and the reactive power reference value, and calculating an inner ring reference current through an instantaneous power theory; The inner loop reference current is: Wherein, the Indicating the value of the active power reference, A reference value for the reactive power is indicated, Which represents the complex power reference value, Representing the units of an imaginary number, Representation of In the coordinate system of The ac side voltage at the moment in time, Representing the inner loop reference current.
3. The method of claim 1, wherein the systematic gain and systematic concentrated perturbation of the reduced matrix converter are localized in The mathematical model under the coordinate system is: Wherein, the 、 Respectively represent In the coordinate system of Time of day and time of day The ac side current gradient of the moment in time, 、、 Respectively represent In the coordinate system of Time of day, the first Time and the first The system at the moment in time concentrates the disturbance, 、、 Respectively represent In the coordinate system of Time of day, the first Time and the first The alternating side current of the moment in time, 、 Respectively represent In the coordinate system of Time of day and time of day The time of day reduced matrix converter switches the input side current, The switching period is indicated as such, Representing the system gain.
4. The method of claim 1, wherein the switch input side of the reduced matrix converter comprises six groups of rectifier switches, and the matrix model of the current at the switch input side is: Wherein, the Representing the current at the input side of the switch, The direct-current side current is represented as, 、、、、 And Respectively representing the switching states of six groups of rectifier stage switches in the reduced matrix converter.
5. The simplified matrix converter bi-vector model-less predictive control method of claim 1, wherein the alternating side current at the next instant is: Wherein, the 、 Respectively represent In the coordinate system of The real and imaginary parts of the ac side input current at the moment, 、 Respectively represent In the coordinate system of The real and imaginary parts of the ac side input current at the moment, 、 Respectively represent In the coordinate system of The system at time concentrates the real and imaginary parts of the disturbance, 、 Respectively represent In the coordinate system of The six groups of rectifier stage switch states at the moment combine the real part and the imaginary part of the corresponding switch input side current, The switching period is indicated as such, Representing the system gain.
6. The method for model-free prediction control of double vectors of a reduced matrix converter according to claim 1, wherein the method for calculating the action time corresponding to the optimal vector and the suboptimal vector is as follows: obtaining a cost function according to the inner ring reference current and the alternating current side current at the next moment; Traversing all switch states through a cost function, selecting a cost function value corresponding to an optimal vector and a suboptimal vector, and calculating the acting time corresponding to the optimal vector and the suboptimal vector; The cost function is expressed as: Wherein, the The cost function is represented as a function of the cost, 、 Respectively represent The real and imaginary parts of the reference input current in the coordinate system, 、 Respectively represent In the coordinate system of Real and imaginary parts of the ac side input current at the moment; the expression of the action time corresponding to the optimal vector and the suboptimal vector is as follows: Wherein, the 、 Representing the minimum 2 cost function values of the optimization, 、 Respectively representing the time corresponding to the optimal vector and the suboptimal vector, Representing the switching period.
7. The method of claim 1, wherein in one switching cycle, the wave-generating mode comprises: When the time corresponding to the optimal vector is more than or equal to the switching period or the time corresponding to the optimal vector is less than or equal to 0, selecting a single-vector wave transmitting mode; in the single-vector wave-transmitting mode, the optimal vector acts on the whole switching period, each switching period is divided into two equal parts, the switching state of the optimal vector is conducted in the first half switching period, and the switching state opposite to the polarity of the optimal vector is conducted in the second half switching period; In the double-vector wave-generating mode, the switching period, the optimal vector action time and the suboptimal vector action time are divided into two equal parts, in the first half of the switching period, when the switching time is within half of the optimal vector action time, the switching state of the optimal vector is conducted, when the switching time is between half of the optimal vector action time and half of the switching period, the switching state of the suboptimal vector is conducted, in the second half of the switching period, when the switching time is between half of the switching period and half of the sum of the switching period and the suboptimal vector action time, the switching state opposite to the suboptimal vector is conducted, and when the switching time is between half of the sum of the switching period and the suboptimal vector action time and the switching period, the switching state opposite to the optimal vector polarity is conducted.
8. The double-vector model-free prediction control system of the reduced matrix converter is characterized by comprising the following components: The first acquisition module is used for acquiring alternating-current side voltage, direct-current side voltage and direct-current side reference voltage data of the reduced matrix converter at the target moment and calculating an inner ring reference current; the second acquisition module is used for acquiring alternating-current side current and switching input side current of the reduced matrix converter at adjacent moments and calculating system gain and system centralized disturbance of the reduced matrix converter at the target moment, wherein the adjacent moments comprise the target moment, the last moment and the last moment; The third acquisition module is used for acquiring direct-current side current of the simplified matrix converter at the target moment, calculating switch input side current corresponding to each switch state, and calculating alternating-current side current at the next moment corresponding to each switch state by combining system gain and system concentrated disturbance of the simplified matrix converter at the target moment; The calculation module is used for calculating the action time corresponding to the optimal vector and the suboptimal vector according to the inner ring reference current and the alternating current side current at the next moment, and determining a wave generation mode according to the magnitude relation between the action time of the optimal vector and the switching period and the magnitude relation between the action time of the suboptimal vector and the switching period in one switching period.
9. A terminal device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, characterized in that the processor implements the steps of the method according to any one of claims 1-7 when the computer program is executed by the processor.
10. A computer readable storage medium storing a computer program, characterized in that the computer program when executed by a processor implements the steps of the method according to any one of claims 1 to 7.

Description

Double-vector model-free prediction control method and related device for simplified matrix converter Technical Field The invention belongs to the technical field of power electronics, and relates to a double-vector model-free predictive control method and a related device for a simplified matrix converter. Background In the technical field of power electronics, a power converter is used as key equipment for realizing conversion and control of electric energy forms, and is widely applied to various fields such as power systems, new energy power generation, electric transmission and the like. With the continuous improvement of requirements on energy utilization efficiency, power density, electric energy quality and the like, the research and development of novel power converters become a hotspot of industrial research. The reduced matrix converter (Reduced Matrix Converter, RMC) is a new type of power converter, the RMC topology is shown in fig. 1, and it is composed of an input filter, RMC rectifier stage, high frequency transformer and controllable rectifier bridge. The RMC rectifying stage plays a core role in the whole conversion process, and can convert three-phase alternating voltage of a power grid into alternating high-frequency pulse voltage, and achieve three-phase to single-phase alternating-current conversion. The positive and negative pulse voltages of the front stage are coupled to the rear stage through a high-frequency transformer and are rectified into direct current, so that the device has the characteristics of no need of an intermediate energy storage capacitor, few conversion stages, high power density, bidirectional power flow and the like. In the control of RMC, conventional model predictive control is widely used in the field of industrial control as an advanced control strategy. However, the conventional model predictive control method for RMC is implemented by establishing an accurate mathematical model, and when the converter control parameter is not matched with the actual parameter, the control output performance is affected, thereby affecting the performance and stability of the entire RMC system. Therefore, accurate model parameters, such as inductance parameters of the reduced matrix converter, need to be used in the prediction process of the model prediction control, but when the parameters are inaccurate, the prediction value is also affected, so that the control effect of the system is affected. Meanwhile, the traditional model prediction control controls the RMC by obtaining a single switch state through traversing optimizing in each sampling period. However, the optimizing mode has randomness, and the random characteristic causes the unstable switching frequency and larger current ripple of the power grid, so that the power quality is reduced, and the application scene with higher requirements on the power quality cannot be met. Disclosure of Invention The invention aims to provide a double-vector model-free predictive control method of a simplified matrix converter and a related device, which solve the problem of low power quality caused by the influence of model parameters on the stability of an RMC system in the conventional model predictive control method. In order to achieve the purpose, the invention is realized by adopting the following technical scheme: a double-vector model-free prediction control method of a simplified matrix converter comprises the following steps: Obtaining alternating-current side voltage, direct-current side voltage and direct-current side reference voltage data of the simplified matrix converter at a target moment, and calculating an inner ring reference current; Acquiring alternating current side current and switching input side current of a simplified matrix converter at adjacent moments, and calculating system gain and system centralized disturbance of the simplified matrix converter at a target moment, wherein the adjacent moments comprise the target moment, the last moment and the last moment; Obtaining direct current side current of the simplified matrix converter at the target moment, calculating switch input side current corresponding to each switch state, and calculating alternating current side current at the next moment corresponding to each switch state by combining system gain and system concentrated disturbance of the simplified matrix converter at the target moment; And calculating the action time corresponding to the optimal vector and the suboptimal vector according to the inner ring reference current and the alternating current side current at the next moment, and determining a wave generation mode according to the magnitude relation between the action time of the optimal vector and the switching period and the magnitude relation between the action time of the suboptimal vector and the switching period in one switching period. Further, the method for calculating the inner loop reference current comprises the following step