Monitoring Pemeliharaan Prediktif Agitator Mixer pada Water Treatment Berbasis Data (IoT)

Abstract

Clean water is a vital necessity for human life and industry, so the clarity of the water treatment system (water treatment) is a crucial factor in maintaining the continuity of clean air supply. One of the important component in this system is the agitator mixer, which functions to mix coagulant and flocculant chemicals so that the dirt particle inspection process runs optimally. Damage to the agitator such as bearing wear, blade alignment, or electric motor disruption can cause a decrease in air quality and increase maintenance costs. This research developed an Internet of Things (IoT)-based predictive maintenance monitoring system to detect the working condition of the agitator mixer in real-time through vibration, temperature, and rotational speed (RPM) sensors. The obtained data was analyzed using the Isolation Forest algorithm to detect anomalies and ANFIS to predict maintenance times. The test results showed a MAPE value of 0.518% and a correlation coefficient of 0.9997, indicating high accuracy between sensor data and actual conditions. This system is able to provide early warning of potential damage, so that maintenance can be carried out in a planned manner without stopping the water treatment process. The implementation of this system improves operational efficiency, extends equipment life, and supports the digital transformation towards a smart and sustainable water treatment industry.