A two-ANN approach to frequency and harmonic evaluation

In power electronics, there are often occasions where the frequency of the signal monitored varies significantly during the measurement period. This creates a huge problem for the conventional spectrum analysis using the fast Fourier transform. An obvious example is the measurement of the output of a variable voltage and variable frequency converter during the motor acceleration/deceleration stages. A solution which uses two Hopfield artificial neural networks (NN) is proposed. One NN is used to calculate the frequency of the signal while the other is used to estimate the magnitudes and phase angles of the spectra. The deceleration profile of a Danfoss variable speed drive (VSD) is used as an example to illustrate the usefulness and effectiveness of this approach.