Modeling And Characteristic Improvement Of Virtual Synchronous Generator

With the penetration of renewable energy generation increases year by year,power systems with a high proportion of renewable energy generation are facing stability challenges due to the relative lack of inertia.Virtual synchronous generator(VSG)can realize friendly grid-connection of renewable energy by inheriting the inertia and damping characteristics of the traditional synchronous generator,and is a significant technical support to meet the requirements of broader application.This dissertation is devoted to the key issues in the application of VSG and conducts from three different levels of its small-signal modeling,characteristic improvement control and application extension to provide comprehensive theoretical guidance and support for the broader application of VSG.Firstly,a full-order statespace small-signal model is built,and the influence of control parameters is revealed based on the model.Secondly,aiming at the issue that lack of adjustment freedom in the dynamic response and the issue that the steady-state power coupling caused by the high resistance to inductance ratio in the middle and low voltage transmission line,methods are proposed to improve dynamic characteristic and steady-state characteristic respectively.At last,the black-box modeling of the station-level VSG with unknown parameters is studied.The main research work is as follows:(1)Focusing on the issue of the small-signal modeling and characteristic analysis of VSG,a detailed full-order state-space small-signal model that fully considers the dynamic of the inner loops and the VSG-based outer power control loop is built,which could provide a basic model for the study of characteristic improvement.Participation factor analysis is used to point out the contributions of different control links.And then the influence of control parameters on the characteristic of VSG is clarified via eigenvalue locus and singular value decomposition(SVD),which could provide guidance for the design of control parameters in engineering applications.Simulation and experimental results verify the effectiveness of the small-signal state-space model.(2)Focusing on the issue of lack of adjustment freedom in the active power dynamic response process,a method to improve the dynamic characteristic of VSG based on the d-axis current feedforward is proposed.The d-axis current is fed forward to the active power control loop through the differential link,and the dynamic performance of VSG is improved by adjusting the parameter of the differential link without affecting the droop characteristic.Meanwhile,active-and reactive-power coupling effects can be reduced effectively.Firstly,the control mechanism of improving the dynamic characteristics by adjusting the damping ratio of the system is analyzed and the method that setting the parameter of the differential link according to the desired damping ratio of system is given.Secondly,an improved state-space small-signal model considering the proposed method is established.On this basis,the participation analysis is used to point out the contributions of different states to the eigenvalues.Besides,the influences of the control parameters of the active power loop,the virtual impedance and the parameter of the differential link are analytically tuned via SVD and eigenvalue locus.The impact of the proposed current feedforward control on all eigenvalues is evaluated.Finally,the decoupling characteristics and robustness of the proposed method are analyzed.Simulation and experimental results are presented to validate the effectiveness of the proposed method to both the grid frequency change and the power reference change.(3)Focusing on the issue of steady-state power coupling error caused by the high resistance to inductance ratio in the middle and low voltage transmission line,a method to improve the steady-state characteristic of VSG based on adaptive reactive power compensation is proposed.The d-axis current is fed forward to the reactive power control loop through the compensation link and can effectively eliminate the reactive power error caused by the active power,which improves the accuracy of VSG.Firstly,it is proposed that the power angle and inner electric potential of VSG could be characterized by the output current.On this basis,the react ive power coupling error could be characterized only by the d-axis current.Secondly,the decoupling strategy is proposed and the tuning method of the introduced parameter is given in detail,and then the impact of the steady-state operating point is considered in real time to improve the applicability of the decoupling strategy.Finally,an improved state-space small-signal model considering the proposed method is established,and the influence of the proposed method on VSG is analyzed.Simulation and experimental results show that the decoupling method is effective.(4)Focusing on the issue of equivalent modeling and parameter identification of the station-level VSG,a black-box modeling method based on the dynamic response analysis under grid disturbances is proposed.The system identification method is used to build the model that can accurately reflect the external characteristic of VSG.The black-box model greatly reduces the amount of calculation,shortens the simulation time and saves the storage space,which could improve the efficiency of the station-level VSG.Firstly,the influence of black-box input and output is analyzed,and the identification models of black-box parameters are determined.The applicable conditions of the linearized model are further analyzed.Secondly,the principle and process of using the least square method to identify the black-box parameters are introduced.Finally,the black-box models are established using experimental data for the improved VSG and traditional VSG respectively,and the validity of the models is verified.At the same time,time domain simulation is conducted to explore the modeling and advantages of the simulation speed and storage space of the station-level VSG black-box model.