Power Flow Analysis And Optimization Of Distribution Network Integrated With Distributed Generation And Electric Vehicles

New power generation technology based on renewable energy sources is rapidly developing, as distributed generation (DG) integrated with network. Taking the view that electric vehicles (EVs) are the effective approach to reduce emission and sound pollution, EVs are becoming an active part of the distribution network. The integration of DG and EVs leads to lilateral power flow in distribution network, which makes the problems of power flow calculation and optimization more complicated.Given the characteristics of DG and EVs, how to take full advantage of distributed resource and alleviate the negtive impact of EVs’ integration without compromising the security and reliability of power supply is an urgent and necessary problem to be solved. While, optimal power flow provides an effective way to solve this problem.First, the physical model and mathematical model of a radial distribution network are proposed. Power flow equations of distribution network are formulated and node types integrated with DG and EVs are analyzed. The forward-backward sweep method which is suitable for radial distribution networks is used to solve the equations. On the basis of proposed method of power flow calculation, nodes voltage and branch currents are calculated with different scenarios to analyze the impact of the integration of DG and EVs on power flow of distribution networks.Second, an OPF model is proposed to determine the optimal output of distributed generation with uncoordinated charging demand of EV in a way that minimizes the active output of substation connected with upper voltage level network or power loss to make network under economical condition and maximize renewable energy penetration. By means of variable substitution and relaxation method, the original nonlinear OPF model can be transformed into a second-order cone programming model. The complex nonlinear relation between variables is transformed into a cone, which makes the model close to the linear programming, so as to simplify the model and reduce the computational complexity. The primal-dual interior-point method is adopted to solve conic programming. Meanwhile, load factor and capacity factor are introduced to acquire the optimal results of different periods in one day, which satisfies the load changing and subjects to the characteristics of output of DG.Third, considering that DG and EVs bring uncertainty to power system, a probabilistic optimal power flow model based on the theory of chance constrained programming is proposed in which Monte Carlo sampling method is used to generate a random sample of wind speed and radiation and power generation cost and operation cost are set as optimization objectives. In this model, the output of DG is still viewed as deterministic variables but the maximum outputs of DG decided by renewable energy are introduced to constrain the deterministic variables to improve the energy utilization of distributed generation and reduce energy waste. Due to the existence of underivable points in the objective function, segmented function is used. Besides, massive simulation is needed. Thus, the chance constrained second-order conic programming is employed. A primal-dual interior-point method characterized by less solution time, better convergence and more accurate results is a kind of algorithm very suitable for the proposed model.IEEE33 nodes distribution system is taken as an example in this paper to verify the proposed method of power flow calculation and perform OPF and POPF solution. Through simulation, the proposed method of power flow calculation, optimization models and solutions are reasonable and effective and can solve the problems:maximizing renewable energy penetration and reducing the negative impact of electric vehicles charging. At the same time, it provides reference and basis for the operation and planning of distribution network with DG and EVs.