A Parallel Structure Spectrum Optimization Method and System Based on Model Predictive Control

Abstract

A parallel structure spectrum optimization method and system based on model predictive control of the present invention adopts a parallel structure of a reference value tracking module and a spectrum shaping module to realize spectrum control, wherein reference value tracking is based on model predictive control to realize reference value tracking of control variables Spectrum shaping is to perform time compensation calculation and prediction calculation on the filtered control variable, predict the frequency component of the control variable in the passband frequency range of the filter, and add it to the cost function. When the error of the switch state is the same or In a similar situation, if a certain switching state can generate fewer frequency components in the passband of the filter, so that the cost function is minimized, then this switching state will be preferentially selected to achieve spectrum shaping control. The invention aims at the filtering characteristics of the high-pass filter and the band-pass filter, adopts reference value tracking and spectrum shaping in parallel, and can realize spectrum shaping for different frequency ranges of control variables, that is, error tracking and spectrum shaping parallel control.

Description

technical field [0001] The invention belongs to the technical field of power electronic control, relates to the control of asynchronous motors, in particular to a parallel structure spectrum optimization method and system based on model predictive control. Background technique [0002] Model predictive control is an effective control method, which can evaluate the future behavior of variables through the cost function and select the state that minimizes the cost function as the future control action of the system. Since the power converter has only a limited number of switching states, when using the model predictive control method for control, only the available switching states contained in the limited control set need to be predicted, and the amount of computation required for system prediction can be greatly reduced. This control strategy is called Finite Set Model Predictive Control (FCS-MPC), which only predicts the future behavior of the system under the possible swit...

AI Technical Summary

Problems solved by technology

[0005] This spectrum control method has good spectrum control performance for band-stop filter-based spectrum shaping, but poor controller performance when band-pass filters, high-frequency filters, or low-frequency filters are used

In the case of some of the above filters, the frequency components of the control variable within the passband of the filter can be suppressed to a certain extent, but the ability of the control variable to track changes in the reference value is significantly deteriorated

In addition, when the above-mentioned filters are used, the phenomenon of system instability will widely appear, that is, the ability of the control variable to track the change of the reference value is severely deteriorated, and it is impossible to simultaneously achieve spectrum control without affecting the stability of the system.

Images