Research on Power System Situational Awareness under Wind Power Scale Access

With the increasing application range of new energy power generation, the uncertainty of distributed power generation leads to a decline in the safety of distribution network operation. New energy generation has become a trend, so it is urgent to solve the problem. The situational awareness of active distribution network becomes particularly important. In this paper, model driven method and machine learning method are used to conduct situational awareness of power grid. Two simulation methods are used in the same scene. Scenario 1 focuses on the randomness of distributed power sources. The model driven method is simulated by MATLAB, and the voltage margin overlimit index and branch overload severity index are used to evaluate the safety state of the system. The long short-term memory model was selected in the machine learning method, and the simulation was carried out in Python language in Spyder. RMSE, MAE, MAPE, Willmott's consistency index and NASH-Sutcliffe coefficient were used to evaluate the prediction accuracy of the model. Finally, the simulation results are summarized and analyzed.