Optimal Research Of Distribution Network Including Distributed Generation Considering The Environmental Benefit

With the development of science and technology, the energy crisis and environmental issues are also increasingly prominent. At the same time, traditional power generation not only consume a lot of fossil fuels but also emit large quantities of polluting gases. In addition, large grid interconnection system which is centralized power generation and transports long-distance has so many defects such as line loss, lack of reliability, high cost and so on. Compared to traditional power generation, distributed generation(DG) has so many advantages, for example, relieves environmental pollution, reduces of investment costs, reduces of network losses, improves of reliability and saves of fossil fuels and so on. In the above context, the combination of DG and centralized power generation has become one of the important measures for energy conservation in many countries, but also the direction of the electric power industry. But When DG meets distribution network, there is a closely relationship between the followed effect and DG’s sites and sizes, therefore it is necessary to make same research about that.Firstly, a method based on sensitivity of power loss and stability of voltage was presented to choose DG’s sites, and compared with the method only considered only one case, results show the superiority of the method which determines DG’s sites proposed by this paper. Then, considering multiple types DGs’ environmental benefit, a multi-objective optimization model about DG’s sizes which takes comprehensive polluted gas emission, total cost of the distribution system and system’s voltage deviation as three targets is established, meanwhile a kind of improved multi-objective particle swarm optimization(IMOPSO) is proposed. In IMOPSO the nonlinear mutation operator is combined to keep the population diversity and avoid falling into Pareto front end which is local optimum. Dynamic adaptive grid is used to maintenance the set of feasible solutions and make the distribution of the non-dominated set more even. Finally, fuzzy selection method with subjective and objective weights is utilized to obtain final scheme from the non-dominated set, and compared with the scheme only considered subjective or objective weight, results show the rationality of this paper’s decisive method. the example’s results show that with the method of this paper can get a reasonable configuration DG. It is verified that the proposed model and algorithm are effective and feasible.