Research On Power Quality Disturbance Detection, Identification And Control Strategy Of AC-DC Hybrid Microgrid With Energy Storage

AC/DC hybrid microgrid HMG)follow the principle of decentralized layout and flexible operation,effectively absorb clean energy such as photovoltaic and wind power,reduce or replace the use of fossil energy,but the capacity of micro grid itself is small,especially in island operation,the stability is weak,the output of micro power supply is strong randomness due to environmental factors,the start and stop of fans and load switching will affect its voltage,power electronic equipment,etc In this paper,the detection and identification of power quality disturbance(PQD)on the AC side and the control strategy on the DC side of hybrid microgrid are studied.Based on MATLAB / Simulink,the photovoltaic grid connection model is established,and the mechanism of voltage fluctuation caused by micro sources,such as photovoltaic,is analyzed in detail.The dead time of photovoltaic inverter,the influence of signal modulation on its output characteristics and the power quality problem of DC side caused by AC side of hybrid micro grid under non ideal working conditions,such as background harmonic and asymmetric fault,are considered The harmonic transfer relationship between microgrid and distribution network and the influence of load change on Microgrid voltage are analyzed.Compared with the existing PQD analysis methods,HHT is used as the theoretical basis of microgrid PQD detection.The EMD algorithm and EEMD algorithm are studied and analyzed.In view of the defects of the two methods in the decomposition process,such as end-point effect and modal aliasing,this paper uses two end-point processing methods for comparison.Based on the EEMD method,a non-stationary component determination method based on multi-scale permutation entropy(MPE)is proposed to avoid over decomposition,and the standard deviation of signal extreme point is used to self The results show that the improved method has comprehensive improvement,improves the detection accuracy of all kinds of PQD and has a certain anti noise interference performance.Finally,the improved method is applied to the simulation model to show its applicability.In this paper,the improved HHT method and parameter optimized SVM arecombined to form PQD classifier.The disturbance signal features are extracted by improved EEMD and HT,and the penalty function C and RBF kernel parameters are optimized by the adaptive grid search method(GS)and the gravitational search method(GSA).By comparing the recognition accuracy of 12 PQDS under the two optimization methods,gsa-svm is similar to other heuristic algorithms However,the improved gs-svm can flexibly adjust the optimization interval and step size to improve the performance of PQD event classification.Finally,aiming at the problem of voltage fluctuation caused by load switching and output variation of micro power supply,hybrid energy storage system is established System,Hess)determines its internal sub unit control method,and studies and analyzes the coordinated control strategy of Hess and bidirectional AC / DC converter;quantitatively analyzes the operation characteristics of HMG in grid connected and island mode,and adjusts the droop characteristics and control method of bidirectional converter in two modes,and determines that Hess based on peer-to-peer control participates in coordinated control with high priority,combined with lithium battery In MATLAB 2016 a,the AC / DC HMG model is established,and the simulation results show that the coordinated control in this paper has a good power suppression effect.