Predicting Load Harmonics in Three Phase Systems Using Neural Networks
This document has been relocated to http://scholarsmine.mst.edu/ele_comeng_facwork/1218
There were 19 downloads as of 27 Jun 2016.
Abstract
This paper proposes a artificial neural network (ANN) based method for the problem of measuring the actual harmonic current injected into a power system network by three phase nonlinear loads without disconnecting any loads from the network. The ANN directly estimates or identifies the nonlinear admittance (or impedance) of the load by using the measured values of voltage and current waveforms. The output of this ANN is a waveform of the current that the load would have injected into the network if the load had been supplied from a sinusoidal voltage source and is therefore a direct measure of load harmonics.