Research On Distribution Network Planning With Distributed Generation

The technology of distributed generation (DG) has attracted more and more attention, with the gradual failure of global primary energy and the deeper understanding of strategies for sustainable development. DG has some merits, such as clean and environment protective, small investment, the flexible way of power generation, high reliability of power supply, especially sustaining power supply for important users in the situation of electric network collapse or unexpected disasters. However, the combination of DG and large power system has influence on traditional power system in many aspects, such as the calculation of voltage node, line power flow, short-circuit current, power quality and reliability. And the degree of the influence is determined by the location and capacity of DG. The questions above bring some substantial challenges for traditional distribution network planning, which makes the power planners consider the effects brought by the DG while choosing a optimal plan.In this thesis, the author makes some analysis and research about the distribution network planning with DG as follow.Firstly, for the single planning of in the distribution network system including DG, this paper analyzes the effects on the network brought by DG, and confirms the maximum capacity of DG according to the total charge with considerations of economy of DG. In addition, it establishes multi-target optimized model whose weight of site selection and the volume determination of DG can change integrating new calculation method of tidal current under the condition of indeterminacy of the number of power sources, location and capacity.Secondly, it accomplishes the solutions of emulation examples based on the genetic algorithm (GA) and ant colony algorithm (ACG) directing at the model above on the purpose of seeking optimal planning scheme. The analysis of results have shown that the rational planning scheme by GA reduces line loss and power purchase cost effectively, whose economy efficiency is not evident. By contrast, the scheme by ACG improves the economy efficiency with lower speed of calculation and more convergence stagnation.Thirdly, the paper puts forward an improved algorithm integrating GA and ACG according to the analysis results of the two algorithms mentioned above, in which reversal variation and elite reservation strategie are introduced in GA and modified max-min ant system is utilized. What’s more, the selection method of juncture is also modified. In comparison with the GA or ACG, the advantages in economy and solution speed of the new confluent algorithm in terms of the planning scheme of the location and capacity are demonstrated by solution of the same emulation example. In conclusion, it is proved in terms of theory analysis and emulation experiment that the modified algorithm is appropriate for solving the distribution network planning including DG. It attains the expected results and improves the economy and security. The rational planning of DG will bring tremendous social and economic benefit, with more demand of energy and the reduction of cost of distribution technology.