Power Quality Disturbances Classification Using Machine Learning Method With Simulink Model

Along with the complicated electric power grids,the system disturnbance issues caused by network faults,weather,or other power cable events are ubiquitous.With the advent of automation in modern industrial set-ups.the sensitivity of process to power quality events is growing.In fact,even a short event can halt processes unexpectedly and this in turn would lead to production wastage,product damage and high recovery costs.Therefore,Power Quality Disturbances(PQDs)have become an important,hitted concern because of their ability to cause mass tripping of distributed generation.The power system operation can be improved and maintained by analyzing the PQDs systematically.An attempt was firstly made to review the modeling and simulation of the PQ disturbances due to the exploitation of various types of loads.The PQDs are created by using parametric equations as well as electrical power distribution system models in MATLAB/Simulink environment.The PQDs of voltage magnitude variation such as sag,swell and interruption,as well as the frequency variation like harmonics are created by applying different types of faults and heavy load in the power distribution model.Similarly to realtime signals,the results of PQD waveforms,obtained by both techniques,are suitable for then checking the performance of the new automatic classification algorithms.Nowadays,numerous algorithms have been developed for the classification of unstructured data.While in this paper,classification of PQDs are performed based on Coiivolution Neural Networks(CNN).Recurrent Neural Network(RNN)and Long Short-Term Memory(LSTM).as well as a proposed hybrid architecture.The third method gave a high accuracy,and the proposed work is believed to serve the needs of the future smart grid applications,which are fast and automatic analysis of the electricity grid and taking automatic countermeasures against potential PQ events.