This study aims to enhance equipment management in grain unloading operations at Khuzestan Ports in Iran by predicting the remaining useful life of electric motors used in grain suction systems (neuero). Utilizing LSTM models in conjunction with environmental factors, this research minimizes unexpected costs associated with equipment failures and reduces downtime in unloading and loading processes. Real-world data from Khuzestan ports demonstrates the high accuracy of the LSTM model in predicting failures. The findings support proactive maintenance strategies, thereby improving efficiency and reliability in the port and maritime industry. While challenges such as limited data, incomplete coverage of environmental factors, and reliance on deep learning models exist, this study provides a foundation for future research on optimizing maintenance and management of neuero electric motors in bulk vessels.

In this paper, the application of distributed-lumped (hybrid) modeling technique (DLMT) in the modeling of longitudinal (axial) vibration of marine shaft system is investigated. The equation of motion for the longitudinal vibration is solved in new analytical method, and modeled as a series of interconnected distributed and lumped elements. Natural frequencies of a rotor system with various elements are calculated based on the distributed lumped modeling technique (DLMT). The results obtained by this method are compared and verified with the results of other techniques, such as FEM, using ANSYS software, and the mode shapes are also presented. The method is then employed for calculating the natural frequencies of a marine propeller shaft with multiple elements such as different couplings. The results are compared and verified with the frequencies and mode shapes obtained by KissSoft software. It is shown that the presented method provides highly accurate results, while it can be simply and effectively applied to the complicated systems.