IEEE Transactions on Power Delivery · 2015 · 26 citations · 9 references
Natural SciencesElectrical EngineeringAtmospheric ElectricityCircuit TheoryTime Series AnalysisCurve FittingLightning CurrentLightning PhysicsLightning ProtectionThunderstorm ElectricityThunderstorm ElectrificationCurrentsLightning Current ParameterizationLightning Current WaveformsLightning Induced VoltagesNoise ImmunityNonlinear RegressionNonlinear Regression ModelsWaveform Decomposition
This paper represents a fast curve-fitting technique for evaluation of either lightning impulse voltage or the current base curve. The technique based on least-square Prony analysis employs the double integration of the recorded waveform for waveform evaluation. The undesirable effect of noise exiting in Prony analysis technique is mitigated. The base curve formula is written in the complex form of the double exponential function. A decomposition technique is utilized for determination of the base curves of full lightning impulse voltages with oscillations and of lightning impulse currents. To verify the performance of the proposed method, the base curves reconstituted from the developed method were compared with the curves reconstructed from a nonlinear least-square regression prescribed in the standard. The execution time of the proposed method is much shorter than that of the conventional method because it does not need the iteration process. With noise immunity characteristics and avoidance of recorded waveform distortion, the results from the proposed technique based on the normal implementation are remarkably accurate. The results of this study indicated that the proposed method is the powerful one to apply for lightning impulse voltage and current base curve evaluation.
9