An Embedded System for the Predictive Maintenance of Aerobic Digesters with Liquid Outputs

This paper introduces a novel embedded edge computing system for the Internet of Things, specifically designed for predictive maintenance of grinding motors in Aerobic Digesters with Liquid Outputs. Low-power sensors and Machine Learning algorithms are integrated, using vibration and sound data for real-time monitoring and proactive anomaly detection. Through the evaluation of the proposed system in real-life environments, an improved performance compared to state-of-the-art models is demonstrated, emphasizing its potential for enhancing the reliability and efficiency of resource utilization in food waste management practices.