Method and system for calculating capacity of wind farm energy storage device

Abstract

The embodiment of the present invention provides a method and system for calculating the capacity of a wind farm energy storage device. The method comprises the steps of: obtaining a standard reference distribution function and all approximate reference distribution functions of the wind power output according to the historical data of wind power output of the wind farm; obtaining the KL distance between each approximate reference distribution function and the standard reference distribution function, and constituting an uncertain set composed of all the approximate reference distribution functions having KL distances not exceeding a preset KL distance; based on the uncertain set, establishing a distributed robust energy storage device planning model of the wind farm; acquiring the capacity of the energy storage device according to the distributed robust energy storage device planning model of the wind farm. The embodiment of the invention establishes the distributed robust energy storage device planning model. The distributed robust optimization does not require an accurate probability distribution function, so the result is more robust in the statistical sense and avoids a situation of putting a large quantity of resources to deal with an extreme scene with an extremely low probability, thereby achieving a low conservativeness.

Description

technical field [0001] The invention relates to the technical field of energy utilization, in particular to a method and system for calculating the capacity of a wind farm energy storage device. Background technique [0002] In the past few decades, clean and renewable energy represented by wind and light has developed rapidly. As of the end of 2017, my country's wind power and photovoltaic installed capacity reached 164 million kilowatts and 130 million kilowatts, respectively, ranking first in the world. However, the long construction period of the transmission channel, the insufficient peak-shaving (frequency modulation) capacity of conventional units, and the potential risks of wind and wind power generation fluctuations to the operation of the power grid have led to the problem of new energy consumption. [0003] In 2017, my country's abandoned wind power was as high as 41.9 billion kWh, and the abandoned wind rate reached 12%, resulting in a serious waste of resources...

AI Technical Summary

Problems solved by technology

The stochastic programming method needs to assume in advance that uncertain factors (such as wind power output) obey a certain probability distribution, but in practice, the empirical distribution is often not accurate due to the lack of sufficient historical data

[0006] In addition, even if a certain reference distribution is estimated based on historical data, when the real distribution deviates from the reference distribution in actual situations, the optimality of the random programming results will often be greatly reduced

Robust optimization ignores the distribution statistics in the historical data of uncertain factors, and only considers the optimization results in the worst case. In practice, the probability of the worst scenario is often very low, resulting in the results of robust optimization tending to keep

Images