| تعداد نشریات | 286 |
| تعداد نشریات سازمان | 84 |
| تعداد شماره‌ها | 2,856 |
| تعداد مقالات | 32,385 |
| تعداد مشاهده مقاله | 41,499,101 |
| تعداد دریافت فایل اصل مقاله | 18,697,590 |
کاربرد و مقایسه روش های بهینه سازی تک هدفه (GA) و دو هدفه (NSGA-II) در طراحی شبکه های تحت فشار | ||
| تحقیقات مهندسی سازه های آبیاری و زهکشی | ||
| دوره 22، شماره 84، آبان 1400، صفحه 1-18 اصل مقاله (827.17 K) | ||
| نوع مقاله: مقاله پژوهشی | ||
| شناسه دیجیتال (DOI): 10.22092/idser.2021.355057.1477 | ||
| نویسندگان | ||
| آتنا حاضری1؛ رسول قبادیان* 2؛ محمد مهدی حیدری3 | ||
| 1دانشجوی دکتری؛ گروه مهندسی آب، دانشگاه رازی، کرمانشاه، ایران | ||
| 2دانشیار گروه مهندسی آب، دانشگاه رازی، کرمانشاه، ایران | ||
| 3استادیار گروه مهندسی آب، دانشگاه رازی، کرمانشاه، ایران | ||
| چکیده | ||
| طراحی بهینه شبکه‌های آبیاری و آبرسانی تحت فشار با هدف کاهش هزینه‌ها همواره مورد توجه می-باشد. در این تحقیق طراحی بهینه شبکه با روش‌های بهینه‌سازی تک هدفه الگوریتم ژنتیک باینری و دوهدفه NSGA-II انجام شد. تابع هدف دوم درNSAGA-II بصورت کمینه نمودن مجموع کمبود فشار در کل سیستم تعریف گردید. چنانچه مقدار این تابع به صفر یا نزدیک آن سوق داده شود نتایج روش تک هدفهGA و دو هدفه NSGA-II قابل مقایسه می‌باشد. به منظور طراحی بهینه شبکه، کدهای کامپیوتری برای بهینه سازی تک هدفه، بهینه سازی دو هدفه و تحلیل هیدرولیکی شبکه به روش ماتریسی شیب در محیط برنامه نویسیVB تهیه و به همدیگر جفت شدند. پس از صحت‌سنجی مدل، یک شبکه مطرح دوحلقه‌ای متشکل از7 لوله و 8 گره که از یک مخزن تغذیه می‌کند با هر دو روش بهینه‌سازی طراحی شد. نتایج نشان‌داد در هر دو روش هزینه لوله‌گذاری تقریبا یکسان و با اختلاف کمتر از یک درصد برآورد شد در حالیکه هزینه محاسبات کامپیوتری در روش NSGA-II در حدود یک پنجاهم روش الگوریتم ژنتیک برآورد گردید. با توجه به اینکه هزینه محاسبات در روش NSGA-II بسیار کمتر از روش GA بدست آمد، استفاده از این روش برای حل مسئله تک هدفه طراحی شبکه تحت فشار نیز توصیه می شود به شرط اینکه در این روش تابع هدف دوم به گونه ای تعریف شود که در صورت رعایت همه قیود مقدار آن به صفر نزدیک شود. برای این منظور تابع هدف مجموع کمبود فشار مناسب تشخیص داده شد. | ||
| کلیدواژه‌ها | ||
| طراحی شبکه تحت فشار؛ الگوریتم ژنتیک؛ الگوریتم ژنتیک مرتب سازی غیرمغلوب | ||
| عنوان مقاله [English] | ||
| Application and comparison of single-objective (GA) and multi-objective (NSGA-II) optimization methods in optimal design of pressurized network | ||
| نویسندگان [English] | ||
| Atena Hazeri1؛ RASOOL Ghobadian2؛ Mohammad Mehdi Heidari3 | ||
| 1department of water engineering, Razi University | ||
| 2Associated professor, Razi univercity, Kermanshah, Iran. | ||
| 3Department of water engineering, Razi University | ||
| چکیده [English] | ||
| Background and Objectives: Proper design of technical and hydraulic parameters plays an essential role in the success of a pressurized irrigation or urban water distribution project and its economy. Therefore, engineers should be able to select the best solution in different stages in terms of design, construction, maintenance and operation according to the existing limitations and make the necessary decisions.The ultimate objective of such decisions is to minimize costs or maximize benefits by considering limitations.The objectives defined for each system may be different but it is certain that in today's engineering world, one-sided objectives are never defined.Today, meta-exploration optimization methods for the optimal design of irrigation and water supply networks have been considered.It is not possible to compare one-objective and two-objective methods in appearance. But in the two-objective method, one of the objectives is defined in such a way that it eventually goes to zero this comparison is possible. Materials and Methods: Hence in the present study, the optimal design of a pressurized network with one-objective binary genetic algorithm and two- objective NSGAII has been done.Genetic algorithm is a method that evaluates different designs through trial and error with analogy criteria and maintains the best designs and eventually achieves the proper design. Multi-objective optimization is a sub-branch of the MCDM multi-criteria decision-making set that takes place among an unlimited set of possible solutions. In such cases unlike single-objective optimization problems, due to the existence of several conflicting goals, a set of answers is obtained instead of just one answer. In order to compare the two methods in terms of accuracy of results and speed of calculations the second objective function in NSGA-II was defined as the sum of the pressure deficiencies in the network. Observance of minimum pressure constraints in the network causes the value of this objective function to reach zero and the results of the two methods are comparable. In order to analyze the network and obtain the pipe flow and pressure in the system nodes, the matrix shape of the gradient method was used. Computer code was developed for single-objective (GA) and multi-objective (NSGAII) optimization methods in VB programming environment. Also, the simulation code according to the matrix shape of the gradient method was prepared in this programming environment. Finally, All the codes were linked to each other. Result: In order to validate the NSGA-II developed cod, its ability to solve several constrained and none- constrained multi-objective mathematical problems was proposed. The results showed that there is a very good agreement between the results of the present model and the results presented by previous researchers. In order to validate the genetic algorithm model, the model was used to solve the linear and nonlinear constrained optimizations problems that have analytical solutions. it has been shown that the results obtained from the model are exactly equal to the results of analytical solutions. After verifying the prepared codes from a programming point of view, a proposed two-loop network consisting of 7 pipes and 8 nodes whit one earth reservoir was designed with both GA and NSGA-II algorithms. The result showed, estimated cost of implementing the studied network by tow method was the same and with a difference of less than 1%, while the cost of calculations in NSGA-II method was estimated to be about 2% of the genetic algorithm method. That is, the time to reach the optimal answer in NSGA-II method is 50 times faster than GA method. Conclusion: Given that the cost of calculations in the NSGA-II method is much lower than the GA method, the use of this method to optimal design of water presuurized network is recommended, Provided that in this method the second objective function is defined in such a way that if all the constraints are observed, its value will be close to zero. For this purpose, the objective function of the sum of pressure deficiencies Was deemed appropriate. | ||
| کلیدواژه‌ها [English] | ||
| Pressurized network design, genetic algorithm(GA), non-dominant sorting genetic algorithm(NSGA-II) | ||
| مراجع | ||
|
Aggrawal, P., Naved, F., & Haider, M. (2015). Genetic Algorithm Based on Sorting Techniques. Journal of Computer Sciences and Applications, 3(2), 40-45. Alperovits, E., & Shamir, U. (1977). Design of optimal water distribution systems. Water resources research, 13(6), 885-900. Bakhshi Ani, A., Ale Ebrahim, H., & Azarhoosh, M. J. (2015). Simulation and multi-objective optimization of a trickle-bed reactor for diesel hydrotreating by a heterogeneous model using non-dominated sorting genetic algorithm II. Energy & Fuels, 29(5), 3041-3051. Deb, K., Pratap, A., Agarwal, S., & Meyarivan, T. (2002). A fast and elitist multiobjective genetic algorithm: NSGA-II. IEEE transactions on evolutionary computation, 6(2), 182-197. Ekinci, Ö., & Konak, H. (2009). An optimization strategy for water distribution networks. Water resources management, 23(1), 169-185. Elferchichi, A., Gharsallah, O., Nouiri, I., Lebdi, F., & Lamaddalena, N. (2009). The genetic algorithm approach for identifying the optimal operation of a multi-reservoirs on-demand irrigation system. Biosystems Engineering, 102(3), 334-344. Garroussi, Z., Ellaia, R., Talbi, E.-G., & Lucas, J.-Y. (2020). A hybrid non-dominated sorting genetic algorithm for a multi-objective demand-side management problem in a smart building. International Journal of Electrical & Computer Engineering (2088-8708), 10(1). Herrera, M., Meniconi, S., Alvisi, S., & Izquierdo, J. (2018). Advanced hydroinformatic techniques for the simulation and analysis of water supply and distribution systems. In: Multidisciplinary Digital Publishing Institute. Jung, B. S., Muleta, M., & Boulos, P. F. (2009). Multi-objective design of transient network models. World Environmental and Water Resources Congress 2009: Great Rivers. Mansouri, R., & Torabi, H. (2015). Application of differential evolution (DE) algorithm for optimizing water distribution networks (case study: Ismail Abad pressurized irrigation network). Journal of Water and Soil, 25(4.2), 81-95. Menke, R., Abraham, E., Parpas, P., & Stoianov, I. (2015). Approximation of system components for pump scheduling optimisation. Procedia Engineering, 119, 1059-1068. Moghaddas, S. M. J., & Samani, H. (2016). Application of central force optimization method to design transient protection devices for water transmission pipelines. Modern Applied Science, 11(3), 76. Moinaldini, E., Mohamad Reza Pour, O., & Zeynali, M. (2018). Application of Imperialist Competitive Algorithm in Optimizations of Pipe Diameters for Urban Water Network (Case study: Shahrdari town, Kerman). Water and Soil Science, 28(2), 29-41. Razi, F. F., Eshlaghy, A. T., Nazemi, J., Alborzi, M., & Poorebrahimi, A. (2015). A hybrid grey-based fuzzy C-means and multiple objective genetic algorithms for project portfolio selection. International Journal of Industrial and Systems Engineering, 21(2), 154-179. Reca, J., Martínez, J., & López, R. (2017). A hybrid water distribution networks design optimization method based on a search space reduction approach and a genetic algorithm. Water, 9(11), 845. Sousa, J., Muranho, J., Marques, A. S., & Gomes, R. (2014). WaterNetGen helps C-town. Procedia Engineering, 89, 103-110. Swamee, P. K., & Sharma, A. K. (2008). Design of water supply pipe networks. John Wiley & Sons. Todini, E., & Pilati, S. (1988). A gradient algorithm for the analysis of pipe networks. In Computer applications in water supply: vol. 1---systems analysis and simulation (pp. 1-20). Torabi, H., Goleij, H., Mirshahi, D., & Shahi Nejad, B. (2018). Irrigation networks Optimization using expanded particle swarm algorithm and linear programming. Irrigation and Water Engineering, 8(2), 187-199. | ||
|
آمار تعداد مشاهده مقاله: 1,483 تعداد دریافت فایل اصل مقاله: 763 |
||