Research Of Distribution Network Dispatch Method Considering The Uncertainty Of Photovoltaic Power And Electrical Vehicles’ Charging Power

With the development of human society, the demand of energy source is increasing. Traditional fossil energies such as coal and oil play a significant role in development of human civilization, which would continue to be the indispensable parts of energy sources in the near future. In the traditional power system, its uncertainty mainly attribute to the stochastic fluctuation of load demand. According to load prediction and day-ahead unit commitment, traditional units can balance load demand and electrical energy supply through frequency regulation. However, with the depletion of traditional energy source and the emission of toxic and harmful waste, green and low carbon economy is becoming the hot focus of different fields. The development and utilization of renewable energies such as wind and photovoltaic energy not only abate the aggravation of the energy crisis, but also make contributions to environment protection and carbon emission reduction.Renewable energies such as wind and photovoltaic energy have the nature of randomness. The integration of these renewable energies can increase the uncertainty of power system dramatically, which will bring challenges to the operation and dispatch of power system. Referring to distribution network, traditional control measures mainly contains reactive power compensation, voltage management at the slack node. We can obtain minimum network loss through measures above. The second chapter of my dissertation indicates that distribution network node voltage will increase in order to reduce network loss through deterministic optimal method. However, on the other hand, the integration of photovoltaic generators itself will also heave the node voltage. Under the mutual influence of these two sides above, the voltage of photovoltaic power injected node may be close to its maximum allowed value, though the capacity of photovoltaic generators integrated is not very large. In this case, a large fluctuation of power may induce the occurrence of distribution network voltage violation. Therefore, the second chapter of this dissertation proposes an optimal dispatch method considering the uncertainty of load and photovoltaic power. This method can set aside a certain extent of system margin, at the same time of reducing the network power loss. Thus, this method can balance the security and economy of distribution network dispatch. In the third chapter of this dissertation, a charging model of electrical vehicles has been established. On this basis, I propose two kinds of electrical vehicles delay charging strategies. The strategy only considering the difference of arriving time of vehicles can reduce the intraday load volatility, but fail to deal with the contradiction between absorbing photovoltaic energy during noon and reducing the late peak load demand. So, the second method proposed in the third chapter considers not only the arriving time of vehicles, but also state of charge of vehicles.On the basis of the third chapter, the fourth chapter designs a joint operation model between electrical vehicles integrated into distribution network and other traditional distribution network control methods. The uncertainty of electrical vehicles’charging power is considered in this joint operation model, which is different from other electrical vehicles’ operation strategies. In this chapter, interval number is used to describe uncertainty factors (photovoltaic power, electrical vehicles’charging power, load level). A simulation example is used to certify the feasibility of the uncertainty of electrical vehicles’charging power described with interval number. In addition, the necessity of considering uncertainty of electrical vehicles’charging power is demonstrated through two typical scenarios provided in the fourth chapter.