Study On Key Techniques Of Stability And Coordinated Control In DC Microgrid

With the extensive application of DC equipments,such as photovoltaic power generation,battery energy storage and variable-frequency power supply,as well as the development of power electronics technology,DC microgrid technology has become one of the research hotspots.As the basic equipment of DC microgrid,converter significantly improves the performance and efficiency of DC microgrid,however,brings some difficulties on stability and control of the converter,converter to converter,DC microgrid system and so on.Moreover,considering the economy and reliability of DC microgrid,it is necessary to coordinate the operation among converters,which makes analysis and control of DC microgrid more difficult.Hence,the stability and coordinated control of DC microgrid is focused on.Especially,several key technologies,such as accurately modeling of the DC-DC converter and high-frequency interaction,stability laws of DC microgrid system,adaptive droop control,secondary shifting control and current control of AC/DC grid interface converter,are further discussed.Firstly,a matrix-based small signal model of PWM comparator with the unified carrier is proposed by using double sampling equivalence of disturbance signal.The original matrix-based small signal modeling method is extended to integrate the models of PWM comparator with common triangular carrier and saw-tooth carrier.Therefore,the matrix-based small signal model of closed-loop control and the matrix-based small signal model of output impedance are established for Buck converter used by triangular carrier.In order to further expand the application scope of the modeling method,the matrix-based small signal model of the PWM comparator with carrier phase shifting and the matrix-based small signal model of its closed-loop control are proposed.To demonstrate the practicality of the model,the phenomenon of beat frequency oscillation,caused by the interaction of switching ripples between parallel converters in DC microgrid,is discussed.The matrix-based small signal model of output impedance is used to analyze and obtain the generating conditions and avoiding measures of beat frequency oscillation.The models provide a practical method for accurately modeling of DC-DC converters with different topologies,high-frequency interaction in DC microgrid.Secondly,in order to obtain the stability law of DC microgrid system,the equivalent model of DC microgrid is established by employing the equivalent power modeling method.Through analyzing the eigenvalue distribution of the Jacobian matrix,the static characteristics and the bifurcation,the key indicators and critical conditions for the stability and the bifurcation of DC microgrid system are obtained.Based on this,the quantitative relationship between the stability of DC microgrid system and the parameters,such as the coefficient of droop control,line impedance and bus capacitance,is identified.Moreover,the stability judgment flow of DC mircogrid is designed.This provides an effective basis for parameter design and system stability prediction of DC microgrid.Thirdly,releasing the contradiction between DC voltage quality and power distribution characteristics,and reducing line loss,an adaptive droop control strategy based on dual factors is proposed,which considers DC voltage quality and line loss.The factors include power impact factor and line impedance impact factor.Compared with the traditional droop control and the adaptive droop control based on a single factor,the proposed strategy can effectively reduce DC voltage fluctuation by adaptively changing the slope of the droop curve,and has the ability to approach the optimal power allocation point,thus reducing the transmission loss of DC microgrid.Forthly,a multi-segment droop control of DC microgrid based on multi-objective evolutionary optimization is proposed,so as to keep the range of droop control and the controllable range after the secondary shifting for droop control.Considering the constraints on stability of DC microgrid,and so as to reduce the impact of line impedance and capacity matching,the Non-Dominated Sorting Genetic Algorithm with an elite strategy is employed to find the optimal segment points.Then keeping the beginning and the end of the droop curve unchanged,and the segmented droop curves are fitted.The optimal droop curves can ensure that the range of droop control remains unchanged after moving the rated work points of the converters,and the controllable range of DC microgrid is effectively guaranteed.Finally,the AC/DC grid interface converter has high operating power and low switching frequency,which causes the serious coupling of current control in dq axis.The complex vector modeling method is introduced to build the control models of the L-type and LCL-type interface converter,respectively.And then the relationship between switching frequency and current coupling is proved.Furthermore,a multiple decoupling current control method is proposed to realize the controls of two types of interface converters decoupling.And the resonance suppression without sensors on the capacitive branch is realized by combining the notch damping.Through the analysis of system stability,dynamic performance and parameter sensitivity,it is proved that the method can realize current control decoupling,enhance the stability and robustness of the control systems,and improve dynamic performance.The dissertation has a total of 104 figures,9 tables,175 references.