Transient Voltage Stability Evaluation Of Wind Power Grid Connected System Based On OptGBM

When wind turbines are added to the regional power grid,it will affect the transient voltage stability of the power grid,because wind power generation is intermittent and random,and may need to absorb reactive power from the power grid.In addition,grid faults may cause wind turbines to be disconnected from the grid due to low voltage,increasing the risk of transient voltage instability.In order to quickly and accurately evaluate the transient voltage stability after the wind farm is connected to the power system,this thesis proposes a transient voltage stability evaluation method based on Optuna optimized Lightgbm(Opt GBM).Firstly,based on the analysis of the research status of wind power at home and abroad and the research on the operating characteristics of wind turbines,a multi-machine equivalent model of wind farm based on Gaussian mixture model(GMM)clustering is established.The operating parameters such as wind speed,generator angular velocity and power are selected as clustering indicators to perform multi-machine dynamic equivalence of wind farms.The GMM clustering method is used to divide the large wind farm into four unit groups,and the Bayesian information criterion(BIC)and the contour coefficient(SC)index are used to evaluate them to obtain the optimal unit grouping scheme.At the same time,the results of k-means clustering method are compared and analyzed,and the effectiveness of the proposed multi-machine equivalent modeling method is verified,which saves time for the input feature collection of subsequent chapters.Secondly,based on the existing research and following the principle of feature selection,33 original features including traditional features and wind power features are selected.Then,the XGBOOST algorithm is used to analyze the importance of these features,and the dimension of the features is reduced to 11.This helps to shorten the calculation time and provides convenience for the subsequent chapters to evaluate the transient voltage stability of wind power grid-connected systems.Finally,aiming at the problems of low robustness、slow training rate and complex parameter adjustment of the existing transient voltage stability evaluation methods of wind power grid-connected system,a transient voltage stability evaluation model based on Optuna optimized Lightgbm model(Opt GBM)is proposed.Firstly,the proposed Opt GBM model discretizes the data by using the calculation of Lightgbm’s own histogram;meanwhile,the accuracy rate,recall rate and missed judgment rate are constructed to measure the evaluation effect of transient voltage stability.Secondly,Optuna optimization is used in training to quickly determine the best parameters of the model.Lastly,simulation verification is carried out on the improved New England 39-bus system with different wind power connection ratios.