Research On Power Quality Disturbance Detection Based On Improved EMD

Power quality involves all walks of life in the national economy and people’s safe use of electricity.High-quality power can not only improve the efficiency and service life of electrical equipment,but also reduce power consumption and production losses and improve national economic benefits.If the power quality is seriously affected,it is of great significance for the evaluation and management of power quality to discover these power quality problems in time.Therefore,it is required to accurately detect the power quality around disturbance parameters,evaluate the power quality through various disturbance parameters extracted,and implement the treatment plan.On the basis of EMD,this paper studies the detection methods of single power quality disturbance signal and composite power quality disturbance signal.Firstly,this paper introduces the effectiveness of EMD and Hilbert in detecting power quality disturbance signals.EMD method is applied to the decomposition of disturbance signals including harmonics.The results show that EMD can divide the voltage sag,rise and break and harmonic disturbance.However,there are modal confusion and false components,which reduce the detection accuracy.In this paper,MEMD is applied to the decomposition of disturbance signals with harmonic to reduce the generation of modal confusion and improve the accuracy of characteristic parameters of disturbance signals.The MEMD method has serious modal confusion when it decomposes the disturbance signal of transient pulse and transient oscillation.In order to solve this problem,this paper uses the NAMEMD to separate the fundamental frequency component from the high frequency oscillation and high frequency pulse in the transient oscillation disturbance and transient pulse signal.At the same time,when the method is applied to the disturbance signal with oscillation and short-term voltage fluctuation(voltage rise,voltage drop and voltage interruption),a single disturbance component with clear physical meaning can be extracted.After Hilbert transformation,frequency,amplitude and disturbance time information can be extracted.Then,the SAMEMD method is applied to the decomposition of disturbance signals containing harmonics,oscillations or both coexistence.Simulation results show that the method can extract both effective harmonic components and useful oscillation components,achieve the separation of various disturbance signals,and improve the endpoint distortion of EMD method.Finally,EMD,MEMD,NAMEMD and SAMEMD are used to decompose an actual collected power quality disturbance signal.The results show that SAMEMD method can extract voltage sags well,and the disturbance time and amplitude obtained by Hilbert transform also proves the effectiveness of the method.