With large-scale integration of renewable energy to power system

its uncertainty has brought great challenges to economic dispatching. To solve this problem

a data classification method for modeling temporal correlation of wind power is put forward

and the versatile-copula distribution is applied to describe the correlation. On the basis of those improvements

the chance constrains and over／under estimation punishment model are presented

and a stochastic optimization model of minimizing generating cost is given in the meantime. After that

through conversions of objective functions and constraints

the stochastic optimization model is transformed into a linear convex problem

and the successive linearization program is applied to solve it accurately. Finally

case study is carried out in modified IEEE-30 bus and IEEE-118 bus system

which verifies the effectiveness of the proposed scheduling method in terms of fitting accuracy and generating cost.