Development of Optimization Algorithms for Shipboard Applications

This thesis explores the application of optimization algorithms to encrease the performance of ships. Several algorithms based on Mixed Integer Linear Programming (MILP) have been developed to address energy management, sizing, and scheduling problems. A Model Predictive Control (MPC) - based Energy Management System (EMS) is introduced to optimize the efficiency of hybrid shipboard power systems while ensuring secure operations. The proposed approach utilizes a MILP formulation to address process constraints, N-1 security requirements, and carbon intensity indicator limits. It also incorporates predictive modeling of ship load and speed using Recurrent Neural Networks (RNN). For power electronic system design, a MILP-based optimization algorithm is proposed to determine the optimal number and ratings of Power Electronics Building Blocks (PEBBs). This approach balances design trade-offs and ensures efficient integration within advanced distribution systems, such as the Navy Integrated Power and Energy Corridor (NiPEC). Finally, an optimal scheduling strategy is proposed for a smart inland port, integrating renewable energy resources, battery storage, and ferry allocation between ports. This strategy utilizes a state-transition model to optimize ferry routing and charging schedules while addressing population-driven transport demands. The proposed optimization algorithms are validated through simulations exploited in Matlab, Simulink, and GAMS environments, demonstrating their scalability and robustness across various maritime applications. The results highlight the effectiveness of these approaches in improving operational efficiency, establishing trade-offs, and minimizing costs, thereby supporting their practical implementation in modern maritime systems.