EFFECT OF NORMALIZATION METHODS ON PREDICTION THE LOAD TORQUE OF DIRECT CURRENT SHUNT MOTOR BY USING ARTIFICIAL NEURAL NETWORK
International Journal of Mechatronics and Applied Mechanics, cilt.2, sa.25, ss.131-142, 2026 (Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 2 Sayı: 25
- Basım Tarihi: 2026
- Doi Numarası: 10.17683/ijomam/issue25/v2.11
- Dergi Adı: International Journal of Mechatronics and Applied Mechanics
- Derginin Tarandığı İndeksler: Scopus
- Sayfa Sayıları: ss.131-142
- Anahtar Kelimeler: ANN, DC Shunt Motor, Load Torque, Normalization
- Açık Arşiv Koleksiyonu: AVESİS Açık Erişim Koleksiyonu
- Erciyes Üniversitesi Adresli: Evet
Özet
Direct current (DC) motors are electrical machines that are widely used in industrial applications and offer various advantages. These motors are especially preferred in different sectors such as robotic systems, factories, textile industry and machine tools due to their ability to operate in wide speed ranges, low cost and easy controllability. For DC shunt motors to operate efficiently, it is critical to accurately predict the load torque of the motor. In this study, the effect of normalization methods on the prediction of the load torque of a DC shunt motor is investigated. For this pur-pose, firstly, the DC shunt motor is modeled in MATLAB Simulink environment and a data set is created by operating the motor under different voltage (Va) and load torque (τL) values. Then, the obtained data are normalized with 5 different normalization methods. These normalization methods play an important role in making the data suitable for the artificial neural network (ANN) mod-el. The designed ANN is trained using the normalized data and the effects of different normalization techniques on ANN performance are analyzed in detail. Performance is evaluated using metrics such as error rates and learning rates, with the results presented in both tables and graphical representations.