Intelligent Engine Mismatch Detection in Detroit Diesel Alusion DW.6WE Using Artificial Neural Networks

Ryan Yudha Adhitya (Politeknik Perkapalan Negeri Surabaya)

International Journal of Marine Engineering Innovation and Research
Vol. 11 No. 2 (2026): In Progress

The maritime industry plays a strategic role in supporting international trade, particularly in the distribution of goods between countries. The main engine, as the primary component that powers a ship, is crucial for its smooth operation. Therefore, monitoring engine condition is crucial. Indicators such as main bearing temperature, exhaust gas temperature, and engine rotation speed are used to assess engine health. Preventive maintenance approaches that rely on fixed schedules are unable to identify damage early. Therefore, Artificial Neural Network (ANN) technology is used to analyze sensor data in real time. Data is sent to the PLC, processed by the ANN, and classified into Normal, Alert, or Danger conditions. This information is displayed visually and accessible to the ship's crew via the node-red platform with the Modbus TCP/IP protocol. Results show that the system has a classification accuracy of 94.11%, is capable of activating automatic alarms, and sends data in real time via the MQTT protocol. This demonstrates the effectiveness of ANN in an intelligent and adaptive ship engine monitoring system.