A parameter-free finite-set model predictive control method for three-level grid-connected inverters

Abstract

The invention discloses a parameter-free finite set model predictive control method for a three-level grid-connected inverter. The system studied by the method includes a DC side voltage source, a DC side capacitor, a three-level inverter circuit, and a three-phase L filter, the equivalent resistance of the three-phase L filter and the three-phase grid. This method does not need to know the actual model parameter values, only by sampling the current flowing through the three-phase L filter and the voltage of the three-phase grid, and then used for parameter-free finite set model predictive control, improving the parameter robustness of finite set model predictive control Sexuality, at the same time does not increase the cost and is simple and effective.

Description

technical field [0001] The invention belongs to the technical field of three-level grid-connected inverter control, and more specifically relates to a parameter-free finite set model predictive control method for a three-level grid-connected inverter, which is used to improve the grid-connected current waveform of the inverter Quality, Improving the Robustness of Finite Set Model Predictive Control Under Parameter Mismatch of Three-Level Grid-connected Inverters. Background technique [0002] In recent years, finite set model predictive control has received great attention in the power electronics community. An early implementation of finite-set model predictive control was for current control and torque control of inverters, and it has been applied to various converter topologies and power electronics. Finite set model predictive control has the advantages of fast dynamic response, simple inclusion of nonlinearities and constraints, multi-objective optimization, and low sw...

AI Technical Summary

Problems solved by technology

[0008] 1. The existing model predictive control based on observer parameter identification, the addition of observers makes the inherently complex model predictive control algorithm more complicated, which is not conducive to actual implementation

[0009] 2. In the existing model predictive control based on the identification of observer parameters, the design of related parameters of the observer is complicated

Improper parameter design will lead to poor control effect, and in more serious cases, it will lead to system instability

[0010] 3. The existing model predictive control based on the identification of observer parameters needs to adjust the relevant parameters of the observer when the actual operating point of the system changes, which is quite difficult in actual implementation

Images