Research Of Distribution Network Reconfiguration Faced Smart Grid

With global warming, the rapid improvement of energy pressure and awareness ofecological civilization, as well as the reliability of power supply and power quality arebecoming more stringent in future digital society, Smart Grid will be the inevitable trend ofthe world’s electricity. In smart grid distribution side, the main manifestation is that a largenumber of distributed generations (DGs) are accessing grid. With the proportion of distributedgeneration in smart grid gradually increased, reasonable DG output power has played apositive role in reducing distribution network losses and improving power quality andreliability of power supply. At the same time, in view of the special nature of wind power,photovoltaic and other distributed generation, the impacts of DG Accessing on reconstruction,running, controlling and all aspects of the distribution network can not be ignored. Therefore,studying on distribution network reconstruction technology, the influence of power loss andpower quality to the distribution network after DG accessing is very useful. It can promote thedevelopment of distributed power supply technology and guarantee the secure and efficientoperation of the future distribution system.This paper first introduces the concept of smart grid and the relevant researches in thisfield at home and abroad. Meanwhile, the author makes a brief introduction of the status ofdistributed generation technology at home and abroad, as well as the effects of distributedgeneration on distribution network, and then analyzes the objective function of thedistribution network reconfiguration and explores the research status of differential evolutionalgorithm and their development trends, which paves the way for the model establishment andproblem solving.Then, considering the weak capability of the forward/backward sweep method disposingPV nodes, an improved algorithm based on forward/backward sweep to calculate power flowis proposed in this paper, which is available to calculating distribution power flow with DG.Several schemes are designed for the different modes of DG, and are tested repeatedly in69-node distribution system. The test results verify that the proposed algorithm is feasible andeffective. At the same time, in order to reflect the influence of voltage and network lossproduced by distributed generation accessing grid, IEEE33node test system is employed tocalculate and analyze. The results show that if the distributed generations are rationallyallocated, the loss of network distribution system will reduce and the distribution generationcan improve the voltage level. So the distributed generation will have a good prospect in the application of smart grid distribution network construction.Finally, this paper presents the Differential Evolution (DE) Algorithm for solving theoptimal distribution system reconfiguration problem in order to realize power lossminimization. According to the characteristics of distribution network, an integer encodingmethod based on independent loop is proposed for reducing the dimension of decisionvariable.In addition, according to the invalid solution problems existing in the evolution, acriterion based on algebraic connectivity of graph theory is presented for completelyremoving the invalid solution, further narrowing solution space. This method is suitable forcomputing the complicated and practical distribution network. DGs are integrated toreconfiguration network to reduce the network loss and improve the quality of voltage.Finally, simulative results of the IEEE33-bus testing system show that the proposed methodhas good convergence and global search capability.