Research On Key Parameter Identification Of Microgrid Virtual Synchronous Generator Control

As the world energy crisis and environmental protection problems are getting more and more intense,renewable energy power generation technology has attracted attention.Due to the continuous development of clean energy generation,its penetration rate in the power grid is increasing,leading to a state of low inertia and damping in the power grid,while the volatility and randomness and other characteristics of new energy generation itself make the stability of the power grid greatly reduced and frequent frequency fluctuations occur.For this reason,researchers propose to simulate the characteristics of synchronous generators through grid-connected inverters to make new energy generation have the same frequency and voltage regulation characteristics as synchronous generators,and this technology is Virtual Synchronous Generator(VSG).This paper studies the influence and identification of key parameters of virtual synchronous generator.The main research contents are as follows:The principle of virtual synchronous generator technology is described and mathematical models are established to design its active-frequency and reactive-voltage control links separately.By analyzing different VSG modeling methods,a model that can uniformly represent the fixed structure VSG is proposed,which covers the characteristics of different VSG modeling methods and provides the basis for VSG key parameter identification.A small-signal state-space model of the unified VSG model is established,and the influence of system stability and key parameters on the regulation performance of the system is analyzed.The key parameters of the VSG are identified when the grid frequency is in the ideal state.The basic principle of parameter identification is introduced,and a small perturbation linearized transfer function between the VSG output power and the command power is established by applying a step perturbation to the VSG command power.The dynamic changes of the system are stimulated by simulating the grid-connected loading experiments,and the key VSG parameters are identified on a time-domain basis using the multi-innovation stochastic gradient method,and the effectiveness and feasibility of the proposed method are verified by simulation.The key parameters of the VSG are identified when the grid frequency is in fluctuation.The linearized transfer function of the VSG output power versus grid frequency is established by simulating the frequency fluctuations of the actual grid and considering the power filters of the phase-locked and active loops required to collect the grid frequency.By analyzing the effects of parameters,the system is divided into a reduced-order model and a full-order model,and the equivalent inertia and damping parameters of VSG output are identified by using the multi-innovation stochastic gradient method and combining the inverter off-grid load perturbation and grid frequency perturbation.At the same time,the step response invariant method is proposed to discretize the VSG model without differences and then use the particle swarm intelligent optimization algorithm to directly identify the key parameters of VSG,and finally the effectiveness of the proposed method is verified by simulation.