Online Autotuning of a Servo Drive: Using Wavelet Fuzzy Neural Networks to Search for the Optimal Bandwidth

To achieve the optimal bandwidth at different speed commands, an online autotuning technique using two two -input, one -output wavelet fuzzy neural networks (WFNNs) for interior permanent magnet synchronous motor (IPMSM) servo drives is proposed in this article. First, the dynamic performance analysis of a field -oriented control (FOC) IPMSM servo drive system is studied. Then, two two -input, one -output, four -layer WFNNs are proposed for the online autotuning of the gains of a proportional -integral (PI) speed controller of the servo motor drive to search for the optimal bandwidth without using the information of plant parameters and the characteristics of the servo motor drive.