Risk-Based Security Analysis And Preventive Control For Complex Distribution Systems

With the development of our country’s strong smart grid and the implementation of energy conservation and emission reduction policies, the promotion of new energy technology has reached a new height, which brings new opportunities to the development of modern power system. At the same time, the uncertainty and intermittent of the distributed generation and electric vehicle charging behavior makes the power grid face many potential risk factors, which brings new challenges to the safe and stable operation of power grid.Considering the stochastic characteristics of the distributed power output, electric vehicle charging and load power, a method based on the point estimation is proposed in this paper. The numerical characteristics of random variables are obtained by solving the stochastic power flow, and the probability distribution of each node voltage and branch power is calculated by using the semi invariant and Cornish-Fisher series expansion. To effectively evaluate the system security level and risk assessment, method considering the events of probability and severity, the risk severity function is defined from the two aspects of node voltage limitation and branch power flow overload, through the value at risk calculations quantify the power grid operation risk.The IEEE-33 node system simulation analyzed the impact of different proportion of distributed power and electric vehicles access to the grid system, which can provide practical reference for the operation and planning of the distributed power and electric vehicle charging station.In order to estimate the degree of risk, a risk warning method of complex distribution system based on the value of risk characteristic is proposed. The method combine CVaR method with actual situation in the power grid operation, explore the risk characteristic value calculation process, then design branch node voltage limit and current overload risk grading rules based on risk characteristic value, which can provided a theoretical basis for risk early warning. IEEE-33 node system examples show that the risk warning can be more intuitive to reflect the degree of risk, by comparing the simulation results with the risk assessment, the validity of the proposed risk warning method is verified.In allusion to the possible risk of distribution network, the risk prevention and control measures are deeply studied. In the optimization model, the system is introduced to reduce the risk of system operation, Combined with the distribution network reactive power optimization, power generation adjustment and optimal load reduction, the basic process of risk control is presented.Using quantum particle swarm algorithm to solve the optimization control model, the optimal control strategy is obtained. By comparing the system risk before and after the different control measures, the effectiveness of the proposed risk control method is verified.