The widespread application of power electronic loads has led to increasing harmonic pollution in the supply system. In order to prevent harmonics from deteriorating the power quality, detecting harmonic components for harmonic mitigations becomes a critical issue. In this paper, an effective procedure based on the radial basis function neural network is proposed to detect harmonic amplitudes of the measured signal. By comparing with several commonly used methods, it is shown that the proposed solution procedure yields more accurate results and requires less sampled data for harmonics assessment.
