| The increasing penetration of dispersed wind power(DWP)poses a challenge to the safety and stability of the distribution network(DN).The full dynamic simulation(FDS)on DN containing DWP is an effective means.However,the complexity of wind turbine and the large quantity of turbines in DWP cluster make it a high-order equation,which causes troubles to simulation.It is necessary to study DWP models and techniques of FDS which achieve a balance of simulation accuracy and speed.Focusing on the above problems,the major work in this dissertation is summerized as follows:(1)The single machine model of DWP and its control strategies was proposed,which could be applied to the FDS.Analysing the basic structure of the doubly-fed induction generator(DFIG)system,reduction to the detailed model was applied and dynamic state(DS)models as well as dynamic and quasi-steady state(QSS)models were established.For the automatic voltage control(AVC)using the reactive power control capability of DWP,a AVC strategy is presented,which could minimize the reactive power and thus maximize the output when voltage is within limits.For the active and reactive power control logic during voltage drop,a active priority control strategy was given,which could raise the voltage more significant.Finally,the effectiveness of the models and strategies was verified by simulation.(2)The multi-machine equivalent modeling method for DWP clusters oriented to FDS was studied.By analysing the multi-loop control of DWP,the grouping indicators were selected.For the wind speed fluctuation and the voltage drop simulation scenes,the hierarchical clustering method was used for clustering,and the method of determining the number of clusters was given.The steps of the equivalent parameters of the wind turbine,collector lines and loads were given.Finally,the effectiveness of the equivalence modeling method was verified by simulating a distribution network system with high-density DWP.(3)The FDS algorithm of DN with wind power was studied.The stiffness of the DN equations were analyzed,which should be solved by using an algorithm with high numerical stability and reducing the stiffness of the equation.The method of differential and algebraic equations alternately solving based on implicit trapezoidal integral method was given,which takes into account the numerical stability and the expansibility of the program.A strategy that DS model and QSS model adaptive switch in simulation was proposed,which calculate the switching criterion according to system state,accelerating simulation further.Finally,the above models and methods were programmed on the self-developed simulation system DGRSS,showing that the proposed methods achieved faster simulation speed while maintaining simulation accuracy. |