الفهرس 
AbstractOnly 14 pages are availabe for public view
 |  | from | 161 |  |  |
| | | from | 161 | | |
Abstract
In the light of the incessant pursuit of the Egyptian government for the development of coastal areas, Damietta governorate seeks to construct a new fishing harbor in Ezbet Elborg coastal city for solving the problems of fishermen. As due to the absence of a harbor for mooring, and therefore fishermen resort to mooring inside Damietta Nile branch, which leads to pollution of the River Nile from the remnants of fishing boats. In addition, providing all the services that fishermen needed in one place which will contribute to the transfer of some fishing professions to the harbor and resulted in creating job opportunities for young people. Thus, the main objective of this research is to develop an Integrated Coastal Structures Management (ICSM) framework for supporting the conceptual design management of the proposed Elborg fishing harbor. Monitoring the coastal area changes of the suggested site for the construction of harbor is the first step toward understanding and characterizing this site for efficient planning and design. For this purpose, an integrated approach based on Landsat satellite images combined with an Artificial Intelligence (AI) classifier technique called Support Vector Machines (SVMs) alongside with GIS. This approach can positively assist in evaluating the coastal changes and extraction of shoreline positions. Simultaneously, a 2D numerical simulation coastal model system (CMS) was set up and calibrated for studying sedimentation issues within the proposed harbor basin. Many different scenarios of harbor’s layout and breakwaters’ characteristics are investigated to effectively mitigate the sedimentation issues inside the harbor basin as it reduces the navigation depth and inhibits the passage of vessels, therefore costly dredging activities are necessary to maintain the required depths. Based on the numerical simulation database of the resulted sedimentation quantities, AI approaches namely Support Vector regression (SVR), Gaussian process regression (GPR), and Neural Networks (ANNs) are developed to predict the sedimentation without the need for numerical simulation models. Eventually, an optimization model by coupling Genetic Algorithms (GAs) an ANN predictive model is developed to minimize sedimentation during the design phase of harbors.