Font Size: a A A

Multi-objective Optimization Of Three Level APF Based On Model Predictive Control

Posted on:2020-05-04Degree:MasterType:Thesis
Country:ChinaCandidate:B J ShiFull Text:PDF
GTID:2392330572981511Subject:Engineering
Abstract/Summary:PDF Full Text Request
With the development of power electronics technology,a large number of non-linear components are applied in power system,which leads to the distortion of voltage and current in power system.Active power filter is widely used in harmonic suppression because of its flexible compensation and obvious effect.Therefore,this paper focuses on the harmonic detection and control methods of Three-level Active Power filter.Firstly,this paper introduces the principle of harmonic current detection of three-level APF,analyses the P-Q method,ip-iq method and adaptive LMS filtering algorithm based on instantaneous reactive power theory,and compares their advantages and disadvantages.Aiming at the problem of slow real-time detection and low extraction accuracy of conventional filtering algorithm,an improved adaptive LMS algorithm is adopted.The traditional ip-iq method,adaptive LMS algorithm and the improved LMS adaptive algorithm are simulated by MATLAB.The simulation and experiment results show that the improved LMS algorithm has better real-time performance,higher extraction accuracy and adaptive adjustment ability.Secondly,an improved multi-objective optimization model predictive control strategy is proposed.A three-level APF discrete mathematical model based on alpha-beta coordinate system is established.The point potential balance control problem is analyzed.The sector is decomposed to reduce the number of rolling optimization times,and the control delay of the strategy is compensated.Current tracking accuracy,switching frequency and neutral point potential are set as objective functions,and hierarchical optimization is carried out.The effects of sampling frequencyand weight coefficient on the objective function are analyzed to achieve multi-objective optimal control.Finally,the dynamic response performance is verified by load fluctuation.The simulation results show that the multi-objective optimization strategy of the improved model predictive current control has outstanding ability of current compensation and dynamic response.Finally,a three-level APF experimental platform based on DSP is designed,and the related hardware and software circuits are designed to verify the experiment.The experimental results verify the effectiveness of the improved harmonic current detection method and the three level inverter control strategy.
Keywords/Search Tags:three-level APF, improved LMS algorithm, model predictive control, multi-objective optimization
PDF Full Text Request
Related items