Research On Distribution Network Planning Containing Distributed Generation

With the gradual depletion of traditional energy,the increasingly serious environmental pollution and the continuous progress of science and technology of electric power,with its flexible,environmental protection and high efficiency,distributed generation(DG)has been paid more and more attention.However,a large number of distributed generations are integrated in distribution network,the power flow,the node voltage and network loss and so on of the distribution network will be affected seriously.And its influence on distribution network is related to the locations and capacities of DG.So it is necessary to make a reasonable planning for the allocation of DGs in distribution network.In this paper,the planning of distribution network with distributed generation is studied,mainly in two aspects: containing DG in the distribution network layout and containing DG in the distribution network expansion planning.A bi-objective optimization model,containing distribution network active power loss minimum and node voltage deviation minimum,is established about the layout planning of DG in distribution network.In view of the dimension difference between multiple objective functions,and even the problem of conflict,using the fuzzy set theory to deal with the problem,the bi-objective optimization problem is transformed to a single objective optimization problem,and the overall performance of the system is greatly promoted by the overall degree of satisfaction.At the same time,aiming at the problem that the particle swarm optimization algorithm is easy to fall into the local optimum and premature convergence,an improved multi-population particle swarm shuffled optimization algorithm is proposed,and the model is solved by the algorithm.The algorithm combines particle swarm optimization algorithm and shuffled frog leaping algorithm,maintaining the diversity of population and enhancing algorithm jump out of local optimal solution to avoid premature convergence.Finally,with the simulation of the IEEE33 node distribution network test system,the rationality and validity of the model and algorithm are verified.Containing the DG of distribution network expansion planning,the optimization model based on the minimum comprehensive cost of distribution network is established.The mathematical model of the plan considers the cost of the new line,the cost of the network loss,the operation cost of the investment of the DG,the cost of electricity and the reliability benefit brought by the DG.In order to solve the problem of planning better,this paper introduces the theory of bi-level planning,the mathematical model is transformed into a bi-level planning model.The upper planning for grid expansion planning problem,the lower level planning is optimization of the location and capacity of DG in the upper planning grid,the lower planning is feedback to the upper planning in the form of the optimal value,and the interaction between the upper and lower layers is realized.According to the characteristics of genetic algorithm has good global searching ability,combined with the binary particle swarm optimization algorithm,proposed an improved binary particle swarm shuffled optimization algorithm,and the algorithm is applied to the upper network expansion planning problem.The lower level planning problem is solved by the improved multi-population particle swarm shuffled optimization algorithm.Finally,a numerical example is analyzed to verify the feasibility of the bi-level planning model and algorithm.