Research On Large-scale Wind Power Prediction And Its Combined Dynamic Economic Dispatch With Thermal Power

The depletion of traditional fossil energy sources and their pollution to the environment make the development of renewable energy technologies inevitable.At present,the total amount of renewable energy exploration and utilization in China has reached the highest in the world.However,compared with developed countries,China's intermittent energy sources such as wind power still face high abandonment rate.When the wind farm and the thermal power plant are jointly scheduled,the traditional deterministic dispatch will not guarantee the security of the system.Therefore,this paper studies the improvement of wind power power prediction accuracy and the improvement of robust interval economic dispatch.If a correlation analysis was conducted between the object wind farm information and the information from adjacent grid points,also using the information effectively will improve short-term wind power prediction accuracy.Firstly,a monotonic nonlinear correlation coefficient index based on regression tree is proposed,in this article,and a reference wind farm with high correlation is selected according to the correlation coefficient matrix.Then,a recursive feature elimination method for nonlinear kernel functions is proposed to determine the smoothing order of meteorological data and eliminate irrelevant feature vectors to improve prediction accuracy.Finally,wind power from 10 wind farms in two regions of Australia was predicted,also compared with other methods to verify the advancement of the proposed method.For the increasingly complex power system economic dispatching,with the improvement of computer computing power,the research of parameterless intelligent optimization algorithm is very important for such complex economic dispatching.To solve the problems existing in the current constraint handling techniques,where it's easy to fall into the local optimal solutions,difficult to satisfy the equality constraints and there are also multiple control parameters,the parameter ? is designed,based on the gradient descent method and multi-objective crowding-distance theory.In this paper,? is a reflection of population constraint violation degree distribution information,which can adaptively adjust the relaxation of the constraint violation threshold ? to solve the constrained problem accurately.In addition,the ? threshold comparison criteria have been improved to increase population diversity.After the solutions of the standard constrained optimization problems(COP)of CEC2017,we compare the results with other advanced algorithms.And the comparison shows that the improved ?-constraint handling techniques can deal with complex equation constraints efficiently.The robust interval economic dispatching for combined wind power and thermal power dispatching does not consider the problem of transmission line safety constraints.So,this paper proposes a function that can greatly reduce the congestion of transmission lines caused by wind power uncertainty.In general,the value of the optimal solution is generally at the boundary of the box constraint,the equality constraint or the inequality constraint.The purpose of this function is to make the optimal solution away from this boundary to ensure the security of the constraint.Then,the objective function combined with the robust interval economic scheduling can greatly alleviate the violation of the constraint condition caused by the uncertainty of the decision variable.Finally,the IEEE118 standard test system is taken as an example,and Monte-Carlo simulation is used to verify the effectiveness of the proposed method.