A virtual laboratory for neural network controlled DC motors based on a DC-DC buck converter

DC-DC converters have a wide usage as the driver circuit of direct current (DC) motors. This has necessitated sensitive speedcontrols to be made on DC motors. Classical controllers have lower performance due to the non-linear features of DC motors,such as saturation and friction. The Neural Network Controllers (NNC) are widely used in controlling poorly-defined, nonlinearand uncertain systems. NNC courses are now being offered by several universities at the bachelor0s and master’s degree levels as aresult of NNC’s successful applications in these fields. However, the training of an NNC driver circuit in a laboratory environmentis a time-consuming and expensive task. In this study, an NNC training set of the DC converter-fed Permanent Magnet DirectCurrent (PMDC) motor, which is part of the electrical machinery courses, was prepared. The set has a flexible structure and agraphical interface. Thanks to this set, it has become possible to change the PMDC motor and controller parameters, and monitorthe system’s reaction under various operational conditions in graphics. This training set can also guarantee effective learning andcomprehension of Artificial Neural Networks (ANN).