DC Motor Friction Identification With ANFIS and LS-SVM Method / Muhammad Zaiyad Ismail ... [et al.]

• Main Authors: Ismail, Muhammad Zaiyad, Azizan, Nur Akmal, Ja’afar, Rabi’atul’adawiyah, Ayub, Muhammad Azmi, Khalid, Noor Khafifah
• Format: Article
• Language: English
• Published: Faculty of Mechanical Engineering Universiti Teknologi MARA (UiTM) 2017
• Subjects: TJ Mechanical engineering and machinery
• Online Access: http://ir.uitm.edu.my/id/eprint/39327/1/39327.pdf; http://ir.uitm.edu.my/id/eprint/39327/

Friction has been an old age problem for any motion system to accomplish its optimum performance. Friction compensation has been identified as an effective strategy to enhance the performance of a motion system. To be able to compensate the friction in motors, the friction itself needs to be identified. Through the latest development in Artificial Intelligent, it has been obvious that the major Artificial Intelligent-paradigms are able to resemble any nonlinear functions precisely and hence, being used as one approach in friction modeling and identification. In this paper, a DC motor is selected as the representative of simple motor. A real-time experiment involving a DC motor is required in getting the best velocity to friction torque relationship. By using MatLab, the friction modeling data is trained with two different methods, which are Adaptive Neuro-Fuzzy Inference System (ANFIS) and Least Squares Support Vector Machine (LS-SVM). The performance of both methods is compared and analysed.