Optimal-Scheduling Model And Algorithm Of Power System Based On Confidence Margin Spaqce Of Wind-photovoltaic Power Generation With Randomness

Wind power generation and photovoltaic power generation have the characteristics of randomness and volatility.The dispatching pressure of power system will be increased by the output of scenery of grid-connected wind and photovoltaic system.Constant scenery output threshold was used as optimization dispatch model in the traditional research.The randomness of Wind power generation and the volatility of photovoltaic power generation were not considered.There is the phenomenon of abandoned wind and light when the threshold of traditional model is low and the phenomenon of indequate power supply when the threshold of traditional model is high.The stability and safety of power system could be endangered.The probability model of regional wind power considering random fluctuation has been researched based on cumulant method.The active power margin on the different condition of confidence levels has been studied to form the confidence margin space based on wind power randomness on the full time axis.The power restrain condition of dynamically adjusting and the optimization of traditional model has been implemented by building the confidence margin space model.The essence of optimal dispatching problem is nonlinear optimization problem.Traditional optimal dispatching problem commonly used interior point algorithm.A initial point is picked in feasibility domain to calculate.The method of dividing the optimization into prediction part and correction part has been proposed to improve convergence performance.But the convergence speed could be reduced if there is over correction or error correction.A kind of correction algorithm of adaptive weighted factor has been proposed to find optimal solution based on standard predictive-correction interior point method.At last,the dispatching model of confidence margin space has been verified which could effectively improve the phenomenon of abandoned wind and light and indequate power supply by numerical examples.At the policy node of IEEE14,IEEE30 and IEEE118,the convergence time and number of correction algorithm ofadaptive weighted factor reduced.The average convergence times are generally increased by about 25% of standard prediction-correction algorithm.The accuracy and feasibility of the correction algorithm of adaptive weighted factor has been verified Combined with optimization objectives.