| Popis: |
Deciding on the groove shape is an important step in the design processof rotating electrical machines. Since the rotary electric machines aredesigned with the help of package programs, the groove shape is automaticallyselected. However, groove leakages fields must be taken into account whendeciding on the shape of the groove. A deep learning model that can both takeinto account the groove leakages and help the groove shape decision be made inthe fastest way and also classify the groove shapes can facilitate and speed upthe work of the designers. In this study, the convolutional neural networks(CNN) model, which is very popular among deep learning (DL) methods, wasdesigned and implemented. In order to ensure the success of the model, increaseits reliability, and ensure its generalizability, the pre-trained CNN model wasrearranged and applied in accordance with the purpose of this study with thehelp of a transfer learning technique (TL). As a result, groove shapeclassification and detection were performed with 100% accuracy with the CNNmodel, and it was proven that the CNN model could positively affect the designprocess of rotary electric machines.Keywords: Rotary Electrical Machines,CNN, TL, DL, |
