Application Research Of Hilbert-huang Transform In Power Quality Detection

Power quality problems are getting more and more attention in recent years whichaffect industrial production and the quality of daily life. The precise detection of powerquality problems is a significant guarantee of improvement on power quality.Focus ondisturbance signal, paper uses the Hilbert Huang Transform (Hilbert-Huang Transform,HHT) for power quality detection which has the obvious characteristics of thetime-frequency analysisPaper applies Hilbert-Huang Transform in power quality detection based ondetailed analysis of the detection of power quality. It applies the Hilbert-Huangtransform to several typical power quality problems: voltage rise, voltage sag andvoltage interruption, voltage transient oscillations, the harmonic and voltage sag signalwith the harmonic. The occurrence and the end moment of disturbance, duration,amplitude and components of the harmonic can be confirmed by the instantaneousfrequency and instantaneous amplitude.The accuracy and rapidity of the detection ofpower quality problems can be improved by the simulation results.The main principle of applying the extreme value point continuation method to thesolution to the endpoint effect of Hilbert-Huang Transform, is to used the signal onboth ends of the extreme value point as a continuation point to extend the length of thesignal and contain more effective information. The advantages of this method can beseen from the analysis of IMF of the several typical characteristics of HHT andcomparison of the Hilbert spectrum.The paper conducts in-depth research on the selection of sampling frequency ofempirical mode decomposition method under different sampling frequency,and analysesthe harmonic signals of the Hilbert spectrum with different sampling frequency. Theresult of the simulation analysis concludes when the sampling frequency is8times ofhighest frequency signal, the signal spectrum of change with time can be analysis byHilbert-Huang Transform correctly. The simulation analysis conclusions when the sampling frequency for8times thehighest frequency signal changes can transform through Hilbert-Huang transformcorrect analysis the signal spectrum of change with time.