Yasmen Amr Hamed Elsaid Khalil,

A Modified Genetic Algorithm-Heuristic Programming Model For Optimal Design Of Sewer Networks / نموذج معدل بخوارزمية جينية مع برمجة مباشرة لتحقيق التصميم الأمثل لشبكات الصرف الصحي / by Yasmen Amr Hamed Elsaid Khalil ; Under the Supervision of Prof. Ashraf Hassan Mohib Ghanem, Dr. Yehya Emad Imam - 106 pages : illustrations ; 30 cm. + CD.

Thesis (M.Sc.)-Cairo University, 2023.

Bibliography: pages 50-52.

Among the Sustainable Development Goals (SDGs) established by the United
Nations, SDG 6 requires water and sanitation for all. However, construction of
wastewater collection systems requires considerable investments. These investments
can be reduced by using automated optimization tools for identifying the optimal design
alternative that achieves minimal system cost while satisfying design constraints. Much
of the related research over the past decade has focused on developing optimization
design tools that are more robust and efficient. Therefore, there is a need to develop an
efficient tool for obtaining the optimum design of the sewage systems with reasonable
computational effort.
In this study, a recent hybrid optimization model that was introduced by Hassan
et al. (2018) and relies on Genetic Algorithm (GA) and Heuristic Programming (HP),
was modified to improve efficiency by reducing runtime and the number of
computations. Similar to the original GA-HP model, the Modified GA-HP (MGAHP)
model identifies optimal pipe diameters and slopes by minimizing the total cost of
gravity sewers and manholes depths for a selected system layout.
The MGAHP model greatly reduces the extensive computational calculations,
checks, and associated penalty costs that are applied in the original GA-HP model by
initially limiting the set of feasible diameters for individual sewer lines. The efficiency
of the proposed model was tested using four case studies ranging in size from 12 to 40
sewers. In the first case study with 12 pipes, the modified model required on average
~17% less function evaluations for the Genetic Algorithm and reduced the overall
runtime by 98%. In the second case study with 20 pipes, the modified model required
10% less GA function evaluations and reduced the overall runtime by 98.2%. In the
third case study with 40 pipes, the modified model required 52% less GA function
evaluations and reduced the overall runtime by 99%. The fourth example was solved
using Bentley SewerGems, GA-HP model and MGAHP model. In the fourth case study
with 20 pipes, the modified model required 14% less GA function evaluations and
reduced the overall runtime by 97%. The cost of the SewerGems design is 17.7 %
higher than the design achieved by GA-HP model and MGAHP model. The efficiency
of the proposed model enables its reliable application for optimal design of large sewer
networks and for examining an extensive set of possible layouts يهدف هذا البحث الى تقديم نموذج معدل بخوارزمية جينية مع برمجة مباشرة لتحقيق التصميم الامثل لشبكات الصرف الصحي، هذا النموذج هو نموذج تم تطويره من نموذج تم تقديمه في Hassan et al. (2018). تم تطوير النموذج الأصلي لتحسين الكفاءة عن طريق تقليل نطاق البحث وذلك بتحديد مجموعة من الاقطار لكل ماسورة في البداية و استبعاد بعض الأقطار التي لا تحقق القيم التصميمية المقبولة من سرعات و نسب امتلاء. يهدف النموذج المقدم الى الوصول الى التصميم الأمثل بكفاءة عالية و يقلل النموذج المقدم بشكل كبير من الحسابات و بالتالي فهو أسرع من النموذج الأصلي.




Text in English and abstract in Arabic & English.


Irrigation and Hydraulics

Optimal design Sewer Network Heuristic Programming Wastewater Management Optimization

627