Research On Voltage Sag Source Identification Of Microgrid Based On Improved Harris Hawk Optimization Algorithm

Nowadays,with the rapid development of high-tech industries,the power system is heavily invested in a variety of sensitive electronic devices,which have high requirements for voltage quality,so voltage sag problem can bring huge losses to the power system.In order to manage the voltage sag problem,the first task is to detect and identify the voltage sag source.In this paper,the dynamic tracking Kalman filter is mainly used to extract the eigenvalues with voltage variation rate,and then input to the improved support vector machine for the identification of voltage sag sources,which mainly includes:Firstly,this paper analyzes the formation causes and voltage sag characteristics of six types of voltage sag sources: single-phase grounded short circuit,two-phase grounded short circuit,two-phase inter-phase short circuit,three-phase short circuit,transformer commissioning and induction motor starting,uses traditional Kalman filter and dynamic tracking Kalman filter to get the residual curve of voltage after denoising,the RMS curve of voltage and the RMS curve of voltage change rate,and extracts the effective voltage sag characteristic values according to these characteristic curves,which provides a basis for the identification of voltage sag sources.Secondly,the support vector machine is selected as the key method for classification in the identification and classification stage of voltage sag sources,where the selection of penalty factor and kernel parameters is very important.Therefore,in order to obtain the appropriate parameters,the improved Harris Hawk optimization algorithm is used to find the optimum.To address the shortcomings of the initial algorithm with low search rate and easy to fall into local optimum,three improvement strategies are proposed: firstly,introducing Sine chaos to replace the random initialization of the population;secondly,introducing energy periodically decreasing regulation method to control the local and global search ability of the algorithm;thirdly,introducing simulated annealing algorithm for the iterative prey position to make the prey exchange information with other Harris hawk individuals to jump out of the local optimality.To demonstrate the superiority of the improved algorithm,six benchmark test functions are selected for testing and comparison.Finally,a microgrid voltage sag simulation system is built in MATLAB/Simulink to collect a large amount of voltage sag data,and then the feature values are extracted and input to different types of classifiers for the classification and identification of voltage sag sources.The simulation experiments show that the proposed identification method with voltage variation rate can identify six types of voltage sag sources more effectively and meet the requirements of voltage sag source identification compared with the method without voltage variation rate.Since there are many steps of detection,denoising,feature extraction and classification in the process of voltage sag source identification,a microgrid voltage sag analysis platform is built,and the above steps are made into a graphical user interface to improve human-computer interaction.Finally,the effectiveness of the proposed method is verified by collecting real measurement data in the microgrid real measurement platform.