Research on Harmonic Detection Method Based on Wavelet Neural Network with a New Structure

Combining wavelet transform and artificial neural network, a wavelet neural network for harmonic detection is proposed. By setting the wavelet function as the excitation transfer function of neural network, a new structure of the neural network is designed to meet the requirements of harmonic detection, which can obtain the parameters such as fundamental frequency, amplitude and phase at one time. Finally, through the simulation analysis, the experimental results show that the method is effective and accurate.