Determination of ship main engine and establishment of optimum organic rankine cycle waste heat recovery system with artificial intelligence approaches

In the presented thesis, the determination of the ship main engine and the establishment of the optimum ORC based waste heat recovery system using artificial intelligence approaches were examined. First, models for predicting the main engine power for containers, tankers and bulk carriers were developed using artificial neural networks (ANN). Then, major criteria in the selection of the ship's main engine were determined and evaluated with the fuzzy AHP method. Finally, the optimum ORC based waste heat recovery system was investigated for a container ship. An optimization study was carried out using a multi-objective gray wolf algorithm for the optimum waste heat recovery system. As a result of the analysis, it was observed that the ANN models developed to predict the main engine power in ships exhibit quite superior performance. As a result of the evaluation of the major criteria in the selection of the main engine performed with Fuzzy AHP, the three most important criteria were determined as fuel consumption, easy operation of the main engine and maintenance cost. In the last part, pareto solutions were obtained as a result of optimization studies using different working fluids and different ORC configuration in the ORC application for the container ship. Using the pareto solutions, the final working fluid was determined by the unique comprehensive evaluation strategy developed by taking into account the thermodynamic, economic, environmental impact and hazard level of the working fluid. As a result of the analysis, it was determined that the most suitable working fluid for both basic ORC and RORC is R245fa. Then, analyzes were performed for design and off-design operating conditions using the R245fa working fluid. Finally, the basic ORC and RORC systems were compared and it was determined that the RORC system was superior.