Self-Tuning for each PMSM Controller using Big Data based ANN

Methods based on linear analysis have been studied for stable control of permanent magnet synchronous motors; however, they are difficult to apply in the operating regions and under control conditions that cannot be linearized. In such instances, trial and error tuning is required to obtain the desired characteristics. In this study, we investigate a method of learning for an artificial neural network using a large amount of adjusted PMSM parameter data and derive the control parameters to stably drive the PMSM.