A Hybrid Meta-heuristic Algorithm for Optimizing Vertical Alignment of Highways

نوع: Type: thesis

مقطع: Segment: masters

عنوان: Title: A Hybrid Meta-heuristic Algorithm for Optimizing Vertical Alignment of Highways

ارائه دهنده: Provider: Amirhossein chaharmahali

اساتید راهنما: Supervisors: Dr Mohsen Babaei

اساتید مشاور: Advisory Professors:

اساتید ممتحن یا داور: Examining professors or referees: Dr Seyyed Mahdi Hosseyniyan and Dr Javad Taheri Nezhad

زمان و تاریخ ارائه: Time and date of presentation: 2022

مکان ارائه: Place of presentation: http://vc.basu.ac.ir/eng-thesis02/

چکیده: Abstract: Abstract: Limited energy resources and environmental issues arising from its high consumption require methods to determine project vertical alignment that, while ensuring driver safety and comfort and reducing road construction costs, reduce vehicle fuel consumption. Although these costs may not be important from the point of view of road builders, the costs paid by users are also part of the costs that are consumed from the country's resources and therefore should be considered in the construction of roads. In this research, a combinatorial optimization model is proposed to determine the vertical alignment of roads, in which the optimal levels at the points of longitudinal slope change are determined in a way that the sum of construction cost and fuel cost are minimized. Moreover, this study makes use of a linear mathematical programming model method the cost of earthworks has been replaced by the linear programming modeling method instead of the conventional Bruckner method, to perform more accurately in calculating the actual costs of soil movement. To solve this problem, a bi-level framework is proposed based on the SA, PSO, GA, GOA, ACO metaheuristic algorithm and SA-GA, SA-PSO, SA-GOA hybrid algorithms, so that at the upper level, the optimal levels at longitudinal slope change points are determined based on the lowest total construction cost and fuel cost, and at the lower level the construction cost is minimized through a linear mathematical programming model. Results show that: a) during a project life cycle, the cost of fuel consumption will be close to the cost of construction, and should not be ignored, b) the speed and accuracy of the proposed algorithm is very appropriate compared to the brute force method and can be used in real-world projects. On average, SA algorithm after 400 iterations, PSO algorithm after 140 iterations, GA algorithm after 40 iterations, GOA algorithm after 60 iterations, ACO algorithm after 50 iterations, SA-GA hybrid algorithm after 40 iterations, hybrid algorithm SA-PSO after 40 iterations and SA-GOA hybrid algorithm after 50 iterations have converged. The average of the results obtained in 5 consecutive executions shows the superiority and better performance of the SA-GA algorithm compared to other algorithms. The SA-GA hybrid algorithm uses the cost function to find the optimal solution and convergence almost fourfold more than other algorithms, and this increases the execution time of the program. It is important to note that the hybrid algorithm finds the optimal answer in 40 iterations, and the average execution time of the program in this number of iterations is 250 seconds, which is a very short time

فایل: ّFile: Download فایل