الفهرس 
IntroductionOnly 14 pages are availabe for public view
 |  | from | 102 |  |  |
| | | from | 102 | | |
Abstract
Propeller performance is a principal factor influencing ship energy and acoustic (vibrations and noise) optimization. This optimization aims to achieve maximum thrust and maximum efficiency. The optimized propeller performance is calculated by various CFD programs. The results of the numerical model scale are compared with that of the full scale. The applied optimization methodology based on the developed interactive genetic algorithms (IGAs) (single objective – Multi objectives) relies on the methodology of the Genetic Algorithms (IGAs) because of genetic interaction. In this algorithm, the blade is designed systematically via a genetic algorithm toward the objective’s simulations based on RANS equations. These design models were conducted to investigate the performance and efficiency of self-propulsion. The numerical simulation is conducted by the JAVA programming language and STAR-CCM+. For the establishment of an autonomous optimization procedure, the commercial optimization software CAESES is combined with the flow solver STAR CCM+. The goal function of the optimization was the open water efficiency around the designed propeller velocity. The optimization variables were the radial distribution of pitch ratio and chord length. The constraints were the propeller’s thrust coefficient (KT) and pressure coefficient (CP). The optimum radial distribution of pitch ratio and chord length was solved by a straightforward evolutionary algorithm. It is the objective to achieve a maximum efficiency ratio with constraints on the thrust coefficient (KT) and the pressure coefficient (CP). For the initial design of propeller performance which a multi-objective optimization scheme is suggested. Using an evolutionary optimization procedure, Non-dominated Sorting Genetic Algorithm II (NSGA-II) is used to approximate the set of Pareto solutions. Afterward, a decision-making strategy is applied to select the best solution for different models.