Study On Sub-synchronous Oscillation Characteristics And Suppression Strategy Of Direct-drive Permanent Magnet Synchronous Wind Turbine Generator

In recent years,with the rapid development of renewable energy generation technology represented by wind power in China,a series of security and stability problems are also caused.One of the focuses is the Sub-Synchronous Oscillation(SSO).Different from the SSO of doubly-fed wind turbine in the past,serious SSO phenomena have also emerged in Direct-drive Permanent Magnet Synchronous Wind Turbine Generator(DDPMSG)without series compensation capacitor.Based on,this thesis studies the SSO problem of DDPMSG from the following aspects:(1)A detailed DDPMSG grid-connected mathematical model is established,including wind turbine model,shafting model,slurry pitch Angle control system model,generator model,grid-side converter(GSC)control model,machine-side converter(MSC),DC link mathematical model,etc.(2)The state matrix and eigenvalues of the system are obtained by linearizing the mathematical model of the system.Then the homotopy function is introduced to find the corresponding relation between the state variable and the eigenvalue to identify the state variable corresponding to the modal eigenvalue of the system oscillation.On this basis,the influence of corresponding parameters of state variables on the eigenvalue of system oscillation mode is further analyzed.(3)A complete DDPMSG mathematical containing Sub-synchronous Oscillation Damping Controller(SSDC)is established,on this basis,taking both eigenvalue real part of system oscillation mode and damping ratio as objective functions.The selection strategy of NSGA-? algorithm is improved in allusion to the parameter optimization problem of poor population diversity,then the improved NSGA-? algorithm is introduced into the SSDC parameters optimization design,the spatial evaluation method is used to evaluate the solution set,and the results show that the Pareto optimal solution obtained by the improved algorithm is more evenly distributed,and the spatial evaluation method is used to evaluate the solution set.Eigenvalue analysis and time-domain simulation results show that: the state variables corresponding to DDPMSG oscillation modal eigenvalues can be identified intuitively and accurately by using homotopy function instead of participation factor method,and the corresponding state variable is the intermediate variable on the GSC control link,further analysis shows that the parameters of PI controller on GSC are strongly related to the eigenvalues of oscillation modes;the improved NSGA-II algorithm has better performance and the Pareto optimal solution obtained is more evenly distributed;the SSDC with improved NSGA-II algorithm design has significant effect on SSO inhibition in DDPMSG.