روﺷﯽ ﻏﯿﺮﻣﺘﻤﺮﮐﺰ ﺑﺮاي ﺑﺎزﯾﺎﺑﯽ ﺷﺒﮑﻪ ﺗﻮزﯾﻊ ﻓﻌﺎل در ﺣﻀﻮر رﯾﺰﺷﺒﮑﻪ ﻫﺎي ﻣﺠﺎور ﻣﺘﺼﻞ ﺑﻪ ﻫﻢ

A Decentralized Technique for Autonomous Distribution Network Restoration in Presence of Coupling Neighboring Microgrids

ﭼﮑﯿﺪه: اﯾﻦ ﻣﻘﺎﻟﻪ ﯾﮏ روش ﻏﯿﺮﻣﺘﻤﺮﮐﺰ ﻣﺒﺘﻨﯽ ﺑﺮ ﺳﯿﺴﺘﻢ ﻫﺎي ﭼﻨﺪ ﻋﺎﻣﻠﻪ را ﺑﺮاي ﺑﺎزﯾﺎﺑﯽ ﺷﺒﮑﻪ ﺗﻮزﯾﻊ ﺑﺎ در ﻧﻈﺮ ﮔﺮﻓﺘﻦ رﯾﺰﺷﺒﮑﻪ ﻫﺎي ﻣﺠﺎور ﻣﺘﺼﻞ ﺑﻪ ﻫﻢ ﮐﻪ در آن ﻣﻨﺎﺑﻊ اﻧﺮژي ﺗﺠﺪﯾﺪﭘﺬﯾﺮ، ﺧﻮدروﻫﺎي ﺑﺮﻗﯽ، ﺳﯿﺴﺘﻢ ذﺧﯿﺮه ﺳﺎزي ﺑﺎﺗﺮي و ﺑﺎر ﻣﺪل ﺳﺎزي ﺷﺪه اﺳﺖ، ﺑﯿﺎن ﻣﯽ ﮐﻨﺪ. روﯾﮑﺮد ﺳﯿﺴﺘﻢ ﻫﺎي ﭼﻨﺪ ﻋﺎﻣﻠﻪ ﻣﯽ ﺗﻮاﻧﺪ ﻣﺴﯿﺮ ﺑﺎزﯾﺎﺑﯽ ﻣﻨﺎﺳﺐ را ﺑﻪ ازاي ﻋﺎﻣﻞﻫﺎي ﻣﻨﻄﻘﻪ اي ﻣﺨﺘﻠﻒ ﺗﻮزﯾﻊ ﺷﺪه در ﻃﻮل ﺷﺒﮑﻪ ﺑﻪ دﺳﺖ آورد و ﻫﻤﭽﻨﯿﻦ رﯾﺰﺷﺒﮑﻪ ﻫﺎي داراي اﺿﺎﻓﻪ ﺑﺎر ﻧﺎﺷﯽ از ﻋﺪم ﺗﻌﺎدل ﻣﯿﺎن ﻣﻨﺎﺑﻊ اﻧﺮژي ﺗﺠﺪﯾﺪﭘﺬﯾﺮ، ﺧﻮدروﻫﺎي ﺑﺮﻗﯽ، ﺳﯿﺴﺘﻢ ذﺧﯿﺮه ﺳﺎزي ﺑﺎﺗﺮي و ﺑﺎر را ﺑﺎ اﺳﺘﻔﺎده از اﺗﺼﺎل اﯾﻦ رﯾﺰﺷﺒﮑﻪ ﻫﺎ ﺑﻪ رﯾﺰﺷﺒﮑﻪ ﻫﺎي ﻣﺠﺎوري ﮐﻪ دﭼﺎر اﺿﺎﻓﻪ ﺑﺎر ﻧﯿﺴﺘﻨﺪ، ﻣﺪﯾﺮﯾﺖ ﮐﻨﺪ. اﯾﻦ اﺳﺘﺮاﺗﮋي، روش ﺟﺪﯾﺪي ﻣﺒﺘﻨﯽ ﺑﺮ ﻗﻮاﻧﯿﻦ ﺳﯿﺴﺘﻢ ﺧﺒﺮه ﺑﻮده ﮐﻪ ﻣﺤﺪودﯾﺖ ﻫﺎي ﻋﻤﻠﯿﺎﺗﯽ ﺳﯿﺴﺘﻢ ﻫﻤﭽﻮن ﻣﺤﺪودﯾﺖ ﻫﺎي ﺟﺮﯾﺎن ﺧﻂ و وﻟﺘﺎژ را در ﻧﻈﺮ ﻣﯽ ﮔﯿﺮد. اﯾﻦ روش ﺑﺮ روي ﺷﺒﮑﻪ ﺗﻮزﯾﻊ ﺷﻌﺎﻋﯽ 33 ﺷﯿﻨﻪ و 12/66 ﮐﯿﻠﻮوﻟﺖ اﻋﻤﺎل ﺷﺪه اﺳﺖ، و در آن ﻋﺎﻣﻞ ﻫﺎ ﺑﺮاي ﺗﺼﻤﯿﻢ ﮔﯿﺮي در ﻣﺤﯿﻂ MATLAB ﺷﺒﯿﻪ ﺳﺎزي ﺷﺪه و از ﻃﺮﯾﻖ ﻣﺤﺎﺳﺒﺎت ﭘﺨﺶ ﺑﺎر اﻋﺘﺒﺎر ﺗﺼﻤﯿﻢ ﮔﯿﺮي ﻋﺎﻣﻞ ﻫﺎ راﺳﺘﯽ آزﻣﺎﺋﯽ ﺷﺪه اﺳﺖ.

This paper expresses the decentralized strategy based on multi-agent system (MAS) for distribution network restoration with considering coupling neighboring microgrids (CNMGs) which is modelled renewable energy sources (RESs), electric vehicles (EVs), battery storage system (BSS) and load. The MAS approach can be obtained suitable restoration path with the different zone agents distributed along the network and it can be managed the overloading microgrids (MGs) due to unbalancing between of RES, BSS, EV and load using interconnecting these MGs to neighboring non-overloading MGs. This strategy is new method that is implemented in this paper, and it is based on expert system rules, and it also considers the system operational limits such as voltage and line flow limits. Finally, this technique applied to 12.66 kV and 33-bus radial distribution network, where agents and power flow calculation are simulated in MATLAB environment for decisions making and validate the correctness of the agents’ decisions, respectively.