DEVELOPMENT OF CYBER-PHYSICAL SYSTEM BASED THREE-PHASE INDUCTION MOTOR CONDITION MONITORING SYSTEM

• Main Author: Christian, Raymond
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/76436
• Institution: Institut Teknologi Bandung

The most used source of rotational movement is three-phase induction motors. When there is a fault in the induction motor system, the production would be disturbed. Therefore induction motors need a monitoring system to detect early faults. The monitoring system can be done visually, using sensors, and utilizing Industry 4.0. The enabling technology of Industry 4.0 such as cyber-physical system can be used to improve the induction motor’s monitoring system. The concept of cyber-physical system can be used to digitize the induction motor, creating a digital twin. Digital twin of the motor is created so the motors can be easily monitored and controlled. The designed three-phase induction motor monitoring system does a motor’s condition identification using an artificial intelligence approach, which is multilayer perceptron. The classified conditions are motor working normally, idle, single phasing fault, unbalanced voltage vault, under voltage vault, and phase reversal fault. The conditions are simulated using a voltage dimmer on every induction motor input. The induction motor’s parameters used as multilayer perceptron model’s input are the RMS value of current and voltage from every phase. Based on the test result, the system cannot classify the phase reversal fault condition since the current and voltage RMS response of that condition is too similar to the normal condition. Without including phase reversal fault, the accuracy of the system classifying the three-phase induction motor’s condition is 92.64%.