CN-121984315-A - Z-source inverter control method and device based on switching system and model prediction

CN121984315ACN 121984315 ACN121984315 ACN 121984315ACN-121984315-A

Abstract

The invention discloses a Z-source inverter control method and a Z-source inverter control device based on a switching system and model prediction, which belong to the technical field of grid-connected inverter control, wherein the Z-source inverter control method comprises the steps of establishing a three-phase current switching model of a three-phase Z-source inverter, and dividing subsystems according to different switch state combinations of the three-phase Z-source inverter; the method comprises the steps of establishing a switching error model of the three-phase Z-source inverter, determining a switching rule according to a switching system theory, selecting subsystems through the switching rule, respectively inputting the selected subsystems into a pre-established model prediction cost function, selecting a next period switching state according to an output result of the model prediction cost function, and outputting a Z-source inverter control signal. The Z source inverter control method adopts refined modeling to accurately describe the dynamic characteristics of the system, combines a control strategy of model prediction and a cost function, has prospective adjusting capability, remarkably improves dynamic response speed and control precision, and reduces modulation complexity.

Inventors

HUANG JINGJING
ZHANG HAIZHEN
Qiao Jichen
YAO JINGYU
WANG XIAOYU
ZHANG YIFAN
WANG SHUO
GAO XUMING
REN ZHIGANG
XIE YURONG

Assignees

西安交通大学
华电电力科学研究院有限公司

Dates

Publication Date: 20260505
Application Date: 20260130

Claims (10)

1. The Z source inverter control method based on the switching system and model prediction is characterized by comprising the following steps of: establishing a three-phase current switching model of the three-phase Z-source inverter, and dividing subsystems according to different switch state combinations of the three-phase Z-source inverter; Establishing a switching error model of the three-phase Z-source inverter, determining a switching rule according to a switching system theory, and selecting a subsystem through the switching rule; And respectively inputting the selected subsystems into a pre-established model prediction cost function, selecting the next period switch state according to the output result of the model prediction cost function, and outputting a Z source inverter control signal.
2. The Z source inverter control method based on switching system and model prediction according to claim 1, wherein the three-phase current switching model of the three-phase Z source inverter is established based on kirchhoff voltage current law, and the expression is as follows: In the formula, The method comprises the steps of taking a system state vector, taking a state matrix as a A, taking b σ as a switching affine term, taking i a (t),i b (t),i c (t) as a three-phase current in an ABC coordinate system, taking u a (t),u b (t),u c (t) as a three-phase voltage in the ABC coordinate system, taking R as a three-phase Z-source inverter circuit equivalent resistance value, taking L as a three-phase Z-source inverter circuit equivalent inductance value, taking V PN as a three-phase Z-source inverter direct current side voltage value, taking F j (j=a, b and c) as a discrete switching function, wherein S j =1 is that an upper switching tube of a j-phase bridge arm is conducted and a lower switching tube is disconnected, taking S j =0 is that an upper switching tube of the j-phase bridge arm is disconnected and a lower switching tube is conducted, and taking S Z =1 is that the three-phase Z-source inverter is in a direct zero state, and taking S Z =0 as a non-direct state.
3. The Z source inverter control method based on switching system and model prediction according to claim 2, wherein the step of performing subsystem division according to different switch state combinations of the three-phase Z source inverter is divided into 9 subsystems, which are respectively as follows: Subsystem S u1 , at this time S a =0,S b =0,S c =0,S Z =0, and subsystem S u1 has F a =0,F b =0,F c =0; Subsystem S u2 , at this time S a =0,S b =0,S c =1,S Z =0, in subsystem S u2 F a =-1/3,F b =-1/3,F c =2/3; Subsystem S u3 , at this time S a =0,S b =1,S c =0,S Z =0, F a =-1/3,F b =2/3,F c = -1/3 in subsystem S u3 ; Subsystem S u4 , at this time S a =0,S b =1,S c =1,S Z =0, and in subsystem S u4 , F a =-2/3,F b =1/3,F c =1/3; Subsystem S u5 , at this time S a =1,S b =0,S c =0,S Z =0, F a =2/3,F b =-1/3,F c = -1/3 in subsystem S u5 ; Subsystem S u6 , at this time S a =1,S b =0,S c =1,S Z =0, and in subsystem S u6 , F a =1/3,F b =-2/3,F c =1/3; Subsystem S u7 , at this time S a =1,S b =1,S c =0,S Z =0, in subsystem S u7 F a =1/3,F b =1/3,F c = -2/3; Subsystem S u8 , at this time S a =1,S b =1,S c =1,S Z =0, and subsystem S u8 has F a =0,F b =0,F c =0; Subsystem S u9 , at which point S a ,S b ,S c is arbitrary, S Z =1, and subsystem S u9 has F a =0,F b =0,F c =0.
4. The method for controlling a Z-source inverter based on a switching system and model prediction according to claim 3, wherein a switching error model expression of the three-phase Z-source inverter is as follows: In the formula, , Is the expected value of the system state vector, Is a system state vector.
5. The method for Z-source inverter control based on switching system and model prediction as claimed in claim 4, wherein Lyapunov function is defined as The switching rule expression determined according to the switching system theory is as follows: substituting the Lyapunov function and the switching error model expression of the three-phase Z-source inverter to obtain the following components: In the formula, The method comprises the steps of representing errors of a state vector and an expected vector, wherein A is a state matrix, b n is a switching affine term, S ui represents an ith subsystem, the value range of i is 1-9, and when the subsystems are selected through a switching rule, the total of 9 subsystems of S u1 -S u9 are sequentially substituted into the switching rule to be calculated, and the subsystem with the minimum switching rule value is selected as an output control signal.
6. The Z-source inverter control method based on switching system and model prediction according to claim 4, wherein the model prediction value function is established as follows: constructing a DC side capacitor of a Z source inverter the voltage model and the inductance current model are as follows: Wherein V c represents a voltage value on a capacitor in a Z source network, i L represents a current value on an inductor in the Z source network, C Z represents a capacitance value in the Z source network, L Z represents an inductance value in the Z source network, i load =i a S a +i b S b +i c S c , i a ,i b ,i c represents three-phase currents in an ABC coordinate system respectively, S a ,S b ,S c represents a switching state of a three-phase bridge arm, S Z =1 represents that the three-phase Z source inverter is in a straight-through zero state, S Z =0 represents that the three-phase Z source inverter is in a non-straight-through state, and V 0 represents a direct-current side input voltage; the capacitance voltage value and the inductance current value on the Z source network in the K+1th control period of the three-phase Z source inverter are calculated according to the following steps: wherein V c (K+1) represents the capacitance voltage value on the Z source network in the K+1th control period, i L (K+1) represents the inductance current value on the Z source network in the K+1th control period, V c (K) represents the capacitance voltage value on the Z source network in the K control period, i L (K) represents the inductance current value on the Z source network in the K control period, and T s represents the time of one control period; the model predictive cost function J for capacitance voltage and inductance current on the Z source network is considered as follows: Where V cr is the desired value of the capacitance voltage on the Z-source network and i Lr is the desired value of the inductance current on the Z-source network.
7. The method for controlling a Z source inverter based on switching system and model prediction according to claim 6, wherein the selected subsystems are respectively input into a pre-established model prediction cost function, a next period switch state is selected according to an output result of the model prediction cost function, and a Z source inverter control signal is output, wherein subsystem S u1 ,S u8 ,S u9 is respectively substituted into a model prediction cost function J expression considering capacitance voltage and inductance current on a Z source network, and the obtained results are J 1 ,J 8 ,J 9 ; S Z in the subsystem S u1 ,S u8 is 0, so J 1 =J 8 ; If J 1 >J 9 , then the next cycle subsystem is selected as S u9 ; If J 1 ≤J 9 , then the next cycle subsystem is selected as S u1 or S u8 ; Selecting a subsystem according to the minimum switch change, selecting S u1 for the Kth+1th cycle when the subsystem of the Kth control cycle is S u1 ,S u2 ,S u3 ,S u5 , and selecting S u8 for the Kth+1th cycle when the subsystem of the Kth control cycle is S u4 ,S u6 ,S u7 ,S u8 ,,S u9 ; and outputting the obtained switch state corresponding to the subsystem in the K+1th control period as a control signal.
8. A Z source inverter control system based on a switching system and model prediction, comprising: The subsystem dividing module is used for establishing a three-phase current switching model of the three-phase Z-source inverter and dividing the subsystem according to different switch state combinations of the three-phase Z-source inverter; the subsystem selection module is used for establishing a switching error model of the three-phase Z-source inverter, determining a switching rule according to a switching system theory and selecting a subsystem through the switching rule; the control signal output module is used for respectively inputting the selected subsystems into a pre-established model prediction cost function, selecting the next periodic switching state according to the output result of the model prediction cost function, and outputting a Z source inverter control signal.
9. An electronic device, comprising: A memory storing at least one instruction, and A processor executing instructions stored in the memory to implement the switching system and model prediction based Z source inverter control method of any one of claims 1 to 7.
10. A computer readable storage medium having stored therein at least one instruction for execution by a processor in an electronic device to implement the switching system and model prediction based Z source inverter control method of any one of claims 1 to 7.

Description

Z-source inverter control method and device based on switching system and model prediction Technical Field The invention belongs to the technical field of grid-connected inverter control, and particularly relates to a Z-source inverter control method and device based on a switching system and model prediction. Background The traditional inverter adopting a two-stage boosting and inverting structure has improved performance and is limited by the inherent circuit topology and control method for a long time. The limitation is mainly characterized in that the output voltage peak value is strictly limited by the voltage level of the direct current side, ideal voltage output is difficult to realize in an application scene with lower input voltage, and the bridge arm straight-through risk exists, so that the system is protected by setting dead time, and the measure not only increases the control complexity, but also causes output waveform distortion and system efficiency reduction. The technical bottlenecks severely limit the adaptability and development potential of the traditional two-stage inverter in application scenes of photovoltaic power generation, electric automobile driving and other wide voltage ranges. Aiming at the defects, the Z-source inverter is in a direct-current zero state of being conducted with an upper switching tube and a lower switching tube in the same bridge arm simultaneously by introducing a Z-source impedance network structure, so that the Z-source inverter can finish boosting and inversion of a direct-current side in the same topology. This feature brings technical advantages thereto. In summary, the Z-source inverter provides an effective single-stage conversion scheme for solving the limitations of the conventional two-stage structure, and the technology is expected to play an important role in improving the overall cost effectiveness of the system reliability. However, the special operating mechanism of the Z source inverter also presents new technical challenges. In modeling aspect, the state space average method commonly adopted in the existing research ignores key factors such as nonlinear characteristics, parasitic parameter effects, switching transient processes and the like of a power device based on an idealized assumption, so that the built model has insufficient description precision under the conditions that the switching frequency is more than 10kHZ and the nonlinear load is provided. In terms of control strategies, due to inherent nonlinearity and multimode switching characteristics of a system, the traditional linearization control method is difficult to accurately coordinate and stably control the system. Therefore, when the input voltage suddenly changes or the load changes in a large range, the system often has the problems of slow dynamic response, large output overshoot, continuous oscillation and the like. In addition, the existing SPWM, SVPWM and other methods need to carry out complex modulation design when realizing the direct zero state, and the limitation severely restricts the full play of the performance of the Z-source inverter and the further improvement of the engineering application effect. Disclosure of Invention The invention aims to solve the problems in the prior art, and provides a Z-source inverter control method and device based on a switching system and model prediction, which accurately describe the dynamic characteristics of a system by adopting refined modeling, improve the control precision and reduce the modulation complexity. In order to achieve the above purpose, the present invention has the following technical scheme: in a first aspect, a Z-source inverter control method based on a switching system and model prediction is provided, including: establishing a three-phase current switching model of the three-phase Z-source inverter, and dividing subsystems according to different switch state combinations of the three-phase Z-source inverter; Establishing a switching error model of the three-phase Z-source inverter, determining a switching rule according to a switching system theory, and selecting a subsystem through the switching rule; And respectively inputting the selected subsystems into a pre-established model prediction cost function, selecting the next period switch state according to the output result of the model prediction cost function, and outputting a Z source inverter control signal. As a preferable scheme, the three-phase current switching model of the three-phase Z source inverter is built based on kirchhoff voltage current law, and the expression is as follows: In the formula, The method comprises the steps of taking a system state vector, taking a state matrix as a A, taking b σ as a switching affine term, taking i a(t),ib(t),ic (t) as a three-phase current in an ABC coordinate system, taking u a(t),ub(t),uc (t) as a three-phase voltage in the ABC coordinate system, taking R as a three-phase Z-source inverter circuit equi