Abstract
This study aims to develop a digital twin system for electric motors using MATLAB Simulink, Arduino Mega, ESP32 Wi-Fi controller, and AWS cloud integration. A combination of software simulations and hardware components was used for real-time data acquisition, normalization, and analysis. Key performance parameters, including motor speed, shaft angle, current, and voltage, were monitored. The system architecture enabled seamless communication between physical components and the virtual model, allowing dynamic updates and synchronization.
Performance was evaluated using statistical metrics such as Mean Absolute Deviation (MAD) and Mean Squared Error (MSE). The digital twin closely matched the physical motor’s behavior, achieving low error rates (e.g., MAPE values below 8%). Experimental findings demonstrated the system’s ability to reliably capture and replicate motor behavior in real time. Despite challenges such as sensor noise and data latency, the system successfully predicted potential failures by identifying anomalies and deviations from expected operating conditions.
The developed digital twin provides an effective tool for predictive maintenance, reducing downtime and enhancing operational efficiency. Future improvements will focus on integrating machine learning, advancing edge computing capabilities, and enhancing data security.