Distribution Network Planning Considering Distributed Generation And Ring Main Unit

Distribution network is an important part of the power system. Improving thedistribution network planning technique level can effectively guide the construction andreconstruction of medium voltage distribution network, and improve the reliability ofpower supply.In urban medium-voltage distribution network, cable rings formed by ring main unit(RMU) is widely used. It can improve the network structure and reduce the main cablebreaking with appropriate position and number of RMU in a ring. Simultaneously, thedistributed power generation for its many advantages has become a trend in the energyfield in recent years, which will be sure to bring new challenge to distribution networkplanning. Therefore, this paper study on the distribution network planning consideringdistributed generation and ring main unit. Specific contents are as follows:①Aiming at urban medium-voltage distribution network planning with whole cablelaying, a new model is proposed with the consideration of connection mode withRMU.The model is to optimize the position and number of RMU and the ascription ofcommon load between rings. Compared to previous work in this area, this modelinclude the3-level struction of ‘HV substation-RMU-distribution transformer’, whichwill be more practical. According to the characteristics of the new model, geneticalgorithm with a multi-information matrix encoding strategy and the correspondinggenetic manipulation is presented.②Based on a distribution network with distributed wind power generation, acoordinated planning model of substation expansion and battery energy storage system(BESS) sizing is proposed, fully considering the BESS charging and discharging benefitto defer substation expansion. The objective of the proposed model is to minimize thepower purchase costs and the investment and operation costs of BESS and substation.The charging and discharging characteristics of BESS and the network securityconstraints are included. For it is a multi-period nonlinear mixed-integer programmingproblem with complementarities constraints, the applicable decomposed primal-dualinterior point method is studyed.③Considering both DG locating/sizing and distribution network expansion planning,a multiobjecitive model of distribution network planning with DG is proposed. Threeobjectives, i.e. the investment and operation cost of DG and line, network loss cost, and environmental cost of power generation are considered. Network connectivity,radiativity, penetration rate of DG and network security constraint are included.Compared with the existing research, this model includes tie-lines for DG, without therequrement of capability of DG to control requency/voltage, hence it has a betterpracticability. A fast and elitist multiobjective genetic algorithm (NSGA-II) to solve thisproblem is studyed. Corresponding encoding strategy and genetic manipulation isdesigned. The fuzzy set theory is used to get the best compromise solution from thenon-dominated solutions.