Zastosowanie sztucznych sieci neuronowych do redukcji stopnia iskrzenia w elektrycznych maszynach komutatorowych prądu stałego

The paper presents a possible application of the capabilities of methods and techniques of ANN artificial intelligence for diagnostics of sparking processes in electric direct current commutator machines. Hitherto applied methods of diagnostics, usually based on the visual observation of sparking, depend on the expert's knowledge and experience. The expert interprets current data obtained from observation. The drawback of this approach lies in its dependence on the expert's knowledge and experience, which makes the data difficult to use for further processing. That is why, apart from improving methods, it is necessary to find a way to objectify the sparking processes in the detection and assessment of damage as well as reduction of the level of sparking of brushes.