IJE Transactions A: Basics Vol. 25, No. 2, April 2012 - 141 OPTIMIZATION OF TREE-STRUCTURED GAS DISTRIBUTION NETWORK USING ANT COLONY OPTIMIZATION: A CASE STUDY A. Mohajeri, I. Mahdavi* Department of Industrial Engineering, Mazandaran University of Science and Technology, Babol, Iran mohajeri.amir@gmail.com, irajarash@rediffmail.com N. Mahdavi-Amiri Faculty of Mathematical Sciences, Sharif University of Technology, Tehran, Iran nezamm@sharif.edu R. Tafazzoli Mazandaran Gas Company, Sari, Iran reza.tafazzoli@gmail.com *Corresponding Author (Received: March 08, 2011 – Accepted in Revised Form: December 15, 2011) doi: 10.5829/idosi.ije.2012.25.02a.04 Abstract An Ant Colony Optimization (ACO) algorithm is proposed for optimal tree-structured natural gas distribution network. Design of pipelines, facilities, and equipment systems are necessary tasks to configure an optimal natural gas network. A mixed integer programming model is formulated to minimize the total cost in the network. The aim is to optimize pipe diameter sizes so that the location-allocation cost is minimized. Pipeline systems in natural gas network must be designed based on gas flow rate, length of pipe, gas maximum pressure drop allowance, and gas maximum velocity allowance. We use the information regarding gas flow rates and pipe diameter sizes considering the gas pressure and velocity restrictions. We apply the Minimum Spanning Tree (MST) technique to obtain a network with minimum number of arcs, spanning all the nodes with no cycle. As a main contribution here, we present and use an ant colony optimization algorithm for solving the problem. The proposed method is applied to a real life situation. Our obtained results are compared to the ones obtained by an exact method. The results show that ACO is an effective approach for gas distribution network optimization. A case study in Mazandaran Gas Company in Iran is conducted to illustrate the validity and effectiveness of the proposed approach. Keywords Natural-gas network; Minimum Spanning Tree (MST); Location-allocation; Ant Colony Optimization (ACO) ﭼﮑ ﯿ ﺪه در اﯾﻦ ﻣﻘﺎﻟﻪ، از اﻟﮕﻮرﯾﺘﻢ ﺑﻬﯿﻨﻪ ﺳﺎزي ﻣﻮرﭼﮕﺎن ﺑﺮاي ﺑﻬﯿﻨﻪ ﺳﺎزي ﺷﺒﮑﻪ ﮔﺎزرﺳﺎﻧﯽ آﻧﺘﻨﯽ اﺳﺘﻔﺎده ﻣﯽ ﺷﻮد. ﻃﺮاﺣﯽ ﻟﻮﻟﻪ ﻫﺎ، ﺗﺴﻬﯿ ﻼت و ﺗﺠﻬﯿﺰات ﺟﺰ وﻇﺎﯾﻒ ﺿﺮوري ﺑﺮاي ﻃﺮاﺣﯽ ﯾﮏ ﺷﺒﮑﻪ ﮔﺎزرﺳﺎﻧﯽ ﺑﻬﯿﻨﻪ ﻣﺤﺴﻮب ﻣﯽ ﺷﻮﻧﺪ. ﯾﮏ ﻣﺪل ﺑﺮﻧﺎﻣﻪ رﯾﺰي ﻋﺪد ﺻﺤﯿﺢ آﻣﯿﺨﺘﻪ ﺑﻪ ﻣﻨﻈﻮر ﻣﯽ ﻧﯿﻤﻢ ﺳﺎزي ﮐﻠﯿﻪ ﻫﺰﯾﻨﻪ ﻫﺎ در ﺷﺒﮑﻪ اراﺋﻪ ﻣﯽ ﺷﻮد. ﻫﺪف اﯾﻦ ﻣﺪل، ﺑﻬﯿﻨﻪ ﺳﺎزي اﻧﺪازه ﻗﻄﺮ ﻟﻮﻟﻪ ﻫﺎ ﻫﻤﺮاه ﺑﺎ ﻣﯽ ﻧﯿﻤﻢ ﺳﺎزي ﻫﺰﯾﻨﻪ ﻣﮑﺎن ﯾﺎﺑﯽ- ﺗﺨﺼﯿﺺ ﻣﯽ ﺑﺎﺷﺪ. ﺳﯿﺴﺘﻢ ﻟﻮﻟﻪ ﮔﺬاري در ﯾﮏ ﺷﺒﮑﻪ ﮔﺎزرﺳﺎﻧﯽ ﻣﯽ ﺑﺎﯾﺴﺖ ﻣﻨﻄﺒﻖ ﺑﺮ ﻧﺮخ ﺟﺮﯾﺎن ﮔﺎز، ﻃﻮل ﮔﺎز ﺑﺎﺷﺪ ﻟﻮﻟﻪ، ﻣﺎﮐﺰﯾﻤﻢ اﻓﺖ ﻓﺸﺎر ﻣﺠﺎز ﮔﺎز و ﻣﺎﮐﺰﯾﻤﻢ ﺳﺮﻋﺖ ﻣﺠﺎز. در اﯾﻦ ﻣﻘﺎﻟﻪ، از اﻃﻼﻋﺎﺗﯽ ﻣﺮﺗﺒﻂ ﺑﺎ ﻧﺮخ ﺟﺮﯾﺎن ﮔﺎز و اﻧﺪازه ﻗﻄﺮ ﻟﻮﻟﻪ ﻫﻤﺮاه ﺑﺎ در ﻧﻈﺮ ﮔﺮﻓﺘﻦ ﻣﺤﺪودﯾﺖ ﻫﺎي ﺳﺮﻋﺖ و ﻓﺸﺎر ﮔﺎز اﺳﺘﻔﺎده ﻣﯽ ﺷﻮد. ﻫﻤﭽﻨﯿﻦ از ﺗﮑﻨﯿﮏ ﻣﯽ ﻧﯿﻤﻢ درﺧﺖ ﭘﻮﺷﺸﯽ ﺑﺮاي ﺑﺪﺳﺖ آوردن ﺷﺒﮑﻪ اي ﺑﺎ ﮐﻤﺘﺮﯾﻦ ﯾﺎل و ﭘﻮﺷﺶ ﺗﻤﺎﻣﯽ ﮔﺮه ﻫﺎ ﺑﺪون ﺷﮑﻞ ﮔﯿﺮي ﻣﺴﯿﺮ ﺑﺴﺘﻪ اي ﺑﻬﺮه ﮔﺮﻓﺘﻪ ﻣﯽ ﺷﻮد. ﺳﻬﻢ ﻋﻤﺪه اﯾﻦ ﻣﻘﺎﻟﻪ ﺑﻬﯿﻨﻪ در اﺳﺘﻔﺎده از اﻟﮕﻮرﯾﺘﻢ ﺳﺎزي ﻣﻮرﭼﮕﺎن ﺑﺮاي ﺣﻞ اﯾﻦ ﮔﻮﻧﻪ ﻣﺴﺎﯾﻞ اﺳﺖ. ﻧﺘﺎﯾﺞ ﺑﺪﺳﺖ آﻣﺪه از اﯾﻦ اﻟﮕﻮرﯾﺘﻢ ﺑﺎ ﻧﺘﺎﯾﺞ ﺑﺪﺳﺖ آﻣﺪه از روش ﺣﻞ دﻗﯿﻖ ﻣﻘﺎﯾﺴﻪ ﻣﯽ ﺷﻮﻧﺪ. ﻧﺘﺎﯾﺞ ﻧﺸﺎن ﻣﯽ دﻫﻨﺪ ﮐﻪ اﻟﮕﻮرﯾﺘﻢ ﭘﯿﺸﻨﻬﺎدي ﺑﻬﯿﻨﻪ ﺳﺎزي ﻣﻮرﭼﮕﺎن، روﯾﮑﺮدي ﻣﻮﺛﺮدر ﺑﻬﯿﻨﻪ ﺳﺎزي ﺷﺒﮑﻪ ﮔﺎزرﺳﺎﻧﯽ ﻣﯽ ﺑﺎﺷﺪ. ﻣﻄﺎﻟﻌﻪ ﻣﻮردي اﻧﺠﺎم ﺷﺪه در ﺷﺮﮐﺖ ﮔﺎز اﺳﺘﺎن ﻣﺎزﻧﺪران، ﺻﺤﺖ روﯾﮑﺮد ﭘﯿﺸﻨﻬﺎدي را ﺗﺎﯾﯿﺪ ﻣﯽ ﮐﻨﺪ.