Research On Data-driven Static Stability Situation Awareness Method

With the rapid development of the energy internet,the scale,complexity and uncertainty of the power grid are increasing day by day,and the traditional analysis methods based on physical modeling and simulation are inevitably limited to the next-generation power system.Wide-area measurement data,a large amount of simulation data and big data technology,and artificial intelligence methods have laid the foundation for data and methods for analyzing the stability of power systems from a data-driven perspective.Data-driven is an effective supplement to physical modeling methods.Therefore,the research on static state-stable situational awareness of power systems combining data-driven and physical modeling methods can provide more valuable technical solutions for power systems.At present,situational awareness is the core link for real-time monitoring of large power grids.Based on the wide-area operation status information of the power grid,it obtains,understands,and displays stable information that can cause changes in the state of the power grid,and predicts future development trends.Lay the foundation for online safety analysis and intelligent active control.Based on the analysis method of situational awareness,this paper makes a preliminary exploration and research on the stability assessment,state prediction and visual display of the power system in static scenarios.The thesis first studies how to integrate situational awareness and static stability of power system,and establishes a framework for analyzing the static stability of power grid from the perspective of data driving.In order to realize the mapping relationship between power grid status data and power system weak links,this paper proposes a weak link identification method based on clustering of complex voltage trajectory features,constructs a set from the node voltage amplitude and phase angle data,and maps the actual power grid to The complex network model,and the mathematical model is compared with the physical characteristics of the power grid.Using the Euclidean distance as the clustering partition criterion,the curvature radius of the complex voltage trajectory is clustered to realize the identification of the weak links of the power grid.The simulation verifies the correctness of the proposed method.As the traditional static voltage stability analysis method is difficult to meet the accuracy and speed requirements,this paper studies the problem of sta tic voltage stability evaluation of data-driven power system.First,the simulation tool is used to capture the system state data along different time sections along the time trajectory,and the characteristic attributes with high correlation with the syst em voltage stability are selected,and the static load active power margin is used to label the data.Then,a random forest classifier integrating multiple decision trees is used to convert the original state label data into stable information and behavior models,and the static voltage stable behavior is perceived and learned from the data.Finally,by querying the classification results of the random forest,the judgment of static voltage stability is realized.Simulation results show that the model is fe asible and effective,and is suitable for static voltage stability assessment.Aiming at the problem of power system state prediction,this paper studies the prediction of complex power grid state from the perspective of data driving.First,the prediction method of Gaussian process regression is introduced.In order to consider the actual power grid operation status,a continuous power flow data matrix integrated with multiple operation modes is constructed.By introducing random numbers in the time direction,the continuous power flow under different operation modes is disrupted.,Makes the distribution of status data more random,and thus simulates the real situation of the daily operation of the power grid.Simulation shows that compared with other intelligent algorithms,the prediction results of this paper have the best results.In view of the problem of visual development and design of power system static and stable situational awareness,this paper uses the data visualization design technology under the environment of Baidu Sugar to develop a visual graphical interface for static and stable situational awareness,which mainly includes the overall safety indicator module,key nodes and lines of the power grid Stability information module,grid stability curve module,static voltage situation assessment module,weak link identification module.