A Study on Power Distribution with Minimized Losses in the Smart Micro-grids
Chao Wei · Institutional Repositories DataBase (IRDB) · 2015
With augment of global economics development in recent years, the demand of electricity is also increasing.However, when traditional centralized power plants use more fossil sources to generate electricity so as to increase production, more carbon dioxide and sulphur dioxide will be emitted into the atmosphere.These gases will aggravate global warming.Therefore, energy production and greenhouse gas put people into a dilemma.The smart micro-grids could avoid this dilemma, because it use renewable energy sources that cannot emit greenhouse gas to the atmosphere.Smart micro-grid is a new kind of power grid.Compared with traditional centralized power plants, its scale is smaller.Therefore, it could be deployed close to the users.Moreover, power loss due to power transmission is less than that caused by traditional centralized power plants.Besides of scale, by using advanced communication and measure technologies, smart micro-grids could receive demand information from the users.Based on the information, the smart micro-grids can automatically adjust their production.However, because power production and the demand of users cannot be predicted, power will be transmitted among the smart micro-grids or between the micro-grids and the macro-station.Therefore, the power distribution control algorithms are proposed to help the micro-grids to meet the demands of users and minimize power loss.In this thesis, we focus on the issue of designing algorithms to help smart micro-grids to find neighbours and exchange power with them so as to meet demands of users.Because power transmission accompanies power loss, the micro-grids want to minimize power loss.Towards such a target, we develop theoretical frameworks to analytically study smart micro-grids power distribution and power loss minimization in this thesis.At first, we consider a centralized algorithm in a 3-layer smart micro-grid without power storage device.The micro-grids will supply power to the users that linked to the micro-grids.When power load of the users is more than the supply power of micro-grid, this micro-grid will buy power from its neighbors or the macro-station.To minimize power loss, the micro-grids send their demands to the data center which is in the macrostation.Based on the demands of micro-grids and our algorithm, data center calculates the coalitions of micro-grids.In other words, the result will decide that which coalitions are the micro-grids belong to.In the same coalition, the micro-grids can exchange power with others.Then the result will be sent to the micro-grids.The micro-grids will choose partner and exchange power depending on the result.The mathematical proofs guarantee that algorithm is optimal, stable and convergent.Moreover, simulations demonstrate that our proposed algorithm makes significant performance comparing with existed method.Then, we extended the first centralized algorithm into the micro-grids with power storage devices case.A greedy coalition formulation algorithm is proposed.When the macrostation receives the demands of micro-grids, the macro-station coordinated mutual power i I am truly indebted to so many people that there is no way to acknowledge them all, or even any of them properly.Without their help and encouragement, this work would not have been done.I extend my deepest gratitude to all.At first, I would like to show my the highest esteem to my supervisor, Nei Kato for his constant supervision and support during my Ph.D. study in Tohoku University.I am grateful to him for providing me insight guidance and warm help in both my study and daily life.Moreover, he creases so nice research environment which has made my Ph.D. period one of the best period in my life.Then, I am especially thankful to Prof.