Economic Dispatch Considering Wind Power Penetration Based On Conditional Risk Method

With the problem of energy shortage and environmental pollution has become increasingly prominent, for less polluting and more mature technology advantages, wind power and new energy are becoming useful complement to the conventional power generation. Wind power is a clean renewable energy which require no fuel costs, however, the uncertainty of wind power generation has brought new challenges to the economic dispatch. Therefore, using the method for processing stochastic problems in power system economic dispatch with wind power has important significance.At first, this paper elaborates the development of wind power at home and abroad, as well as the impact of the wind power integration on power system on economic dispatch, and from the point of view of certainty and uncertainty analyzing the research status of power system economic dispatch with wind farm, and describe in detail and compare the different methods of dealing with uncertainty. Secondly, according to the randomness of wind speed and the function of wind speed and wind power, the probability distribution model of wind speed and wind power are proposed. With the uncertainty of wind power output, the deviation between scheduled wind power and available wind power is involved, and taking the punishment to consider the deviation of the costs incurred as overestimation and underestimation of the wind power penalty costs are included in the overall goal of the economic dispatch, then, the economic dispatch model is established. IEEE30node system for the calculation example, it analyzed the impact of the cost coefficient of the overestimation and underestimation of available wind power on the total system cost and the scheduled wind power, and observe the changes of the scheduled wind power by changing the scale parameter of Weibull distribution.In order to study the uncertainty impact on system cost, on the basis of the previous model, Conditional Value-at-risk (CVaR) method is introduced to calculating the penalty cost of the overestimation and underestimation of available wind power, and constructing a new economic dispatch model of power system. Taking the scheduled wind power as the optimization variables, defining the penalty cost function of wind power, and using the aggregate function for smoothing solving the maximum problem of the penalty cost function, the penalty cost function is transformed into a continuously differentiable convex function. Because it is difficult to describe the analytical expression of the integral of the probability density function of the available wind power, Monte Carlo stochastic simulation method is utilized to transform random variables into discrete points instead of the integral function. Finally, introducing auxiliary variables, the calculation can be simplified with transforming the model into convex programming model.Taking IEEE30node system for example, it analyses and compare the results of the scheduling of the above two models. Taking the CVaR method, IEEE30-bus system simulations show that the scheduling decision-makers can regulate the different confidence level to develop a rational scheduling scheme. And in a given confidence level, the changes in the total generation cost and the scheduled wind power relate to the cost coefficient of the overestimation and underestimation of available wind power, and scheduling decision-makers can take it as a theoretical basis, based on actual demand to set the relevant parameters, reasonable arrangements output plan for the wind turbine and other conventional unit. So, the effectiveness and accuracy of the proposed strategy can be verified. This method can be used as a new stochastic optimization method and models to solve the economic dispatch problem in wind power integrated system.