Study On Electricity Power Purchase Optimization In Day-ahead Market With Consideration For The Uncertainty Of Wind Power Output

In the traditional electrical power system,the purchased power in the day-ahead market is equal to the load forecast due to the small load forecast error.When the actual load power is different from the predicted,it just needs to dispatch the reserve capcity and spend corresponding energy cost to keep the powe balanced.With the increase of wind capacity integrated into the grid,the unpredictability of the wind makes the wind curtailment problem more serious.In order to improve the wind acceptance level and reduce the curtailmented wind power,the reasonable power purchase plan should be made combining of the reserve capacity market and real market with consideration for the uncertainty of the wind.The main content of this paper is as follow:Firstly,the price of the real market is in the form of interval.And the union market robust optimization model is built consideration of the day-ahead market and real market using robust optimal method.The improved quantum-behaved particle swarm algorithm(QPSO)is proposed by analyzing the statistic characteristic of the geneal QPSO to show the limitation of the algorithm.And the improved QPSO is used to solve the robust optimization model.By comparing the result of the union market and separated market,it shows that the power puschase plan of the union market is better the the separated.What's more,with the increase of the wind capacity integrated into grid,the more deviation power,the better economy the union market got.Secondly,the uncertainty of wind power is in the form of interval because the probabilistic model needs a large amount of historical data.The multi-objective interval model for day-ahead energy and reserve market is built,which contains two obj ective functions consisting of the minimum clearing cost of the day-ahead market and real-time imbalance power of the real market.The interval optimization model is converted into double-layer nonlinear pessimistic and optimistic models.The improved multi-objective quantum particle swarm optimization(MOQPSO)is developed by analyzing the deficiency of general MOQPSO.The result of the analysis example indicates that the more real Pareto front can be got with the improved QPSO.The decision makers of the power purchase can select an appropriate compromise solution for reference about the terminal decision of the day-ahead market.Finally,the uncertainty of wind power and the price of real market are all represented of interval.The power purchase model is established consideration of day-ahead energy market,reserve capacity market and real time energy market.The real market balanced cost can't be determined so that the power purchase strategy is made with minimizing the roubust regret consideration of the regret of decision maker in day-ahead market.The result of the example verifies that the proposed power purchase strategy can get better purchase plan than the traditional by comparision of two purchase strategy.