Startup Behavior of Harmonic Suppression in Electrical Machines Using Iterative Learning Control and Neural Networks

Abstract:
Electrical machines generate unwanted flux and current harmonics. Harmonics can be suppressed by using various methods. In this paper, the harmonics are reduced by using iterative learning control (ILC) and neural networks (NNs). This paper focuses on the startup behavior of the control system. The ILC can compensate well for the harmonics in operation at constant speed and constant current reference values, but needs multiple rotations to learn. The NNs are trained with the data from the ILC and help to suppress the harmonics well even in transient operation and from the first rotation. The simulation model is based on flux and torque maps, depending on dq-currents and the electrical angle. The methods are also applied on the test bench and measurement results are presented.