考虑风电不确定性的短期合同电量协同分解优化模型及算法

Model and Algorithm of Cooperative Optimization Decomposition for Short-Term Contract Electricity Considering Wind Power Uncertainty

现有合同电量分解方法大多没有考虑风电不确定性的影响及未与发电计划协同优化,导致到期合同电量往往未能够得到充分执行。提出一种考虑风电不确定性和检修计划影响的短期合同电量协同分解优化新模型及算法,使得到期时合同电量能够公平合理地充分执行。首先,以发电成本、合同偏差成本及风电品质风险成本最小为目标,构建短期合同电量分解到天的优化模型及算法,得到每天预计完成的合同电量。然后,基于日前及日内短期及超短期负荷及风功率预测信息,分别构建考虑合同完成度的日前鲁棒发电计划及日内重调度优化模型及算法,在保证对负荷尽可能供电及系统安全约束满足的前提下,实现风功率充分消纳及每日合同电量的充分执行。在此基础上,对每日未完成的合同电量,通过后续日合同电量的滚动修正进一步加以实现,从而保证到期时短期合同电量能够得到充分执行。算例证实了该模型及算法的可行性和先进性。

Most of the existing contract electricity decomposition methods do not take into account the impact of wind power uncertainty and do not coordinate with the generation plan for optimization,resulting in their insufficient execution when the contract is due.A new cooperative optimization decomposition model and algorithm of short-term contract electricity is therefore proposed with consideration of the wind power uncertainty and maintenance plans.Firstly,a cooperative optimization decomposition model and algorithm of short-term contract electricity is constructed with the objective of minimization of the generation cost,contract deviation cost and low-quality wind power risk cost,to obtain the decomposed daily contract electricity.And then,based on the dayahead and intraday short-term and ultra-short-term load and wind power forecast information,the day-ahead robust generation plan model and algorithm and the intraday redispatch model and algorithm are constructed respectively with consideration of the contract completion degree,to realize the full accommodation of wind power and full execution of the daily contract electricity with the premise of power load to be fully supplied and the security constraints to be met.Furthermore,the uncompleted daily contract electricity is fulfilled through rolling updating the subsequent daily contract electricity to ensure the full execution of the short contract electricity when the contract is due.The results of case study have demonstrated the feasibility and superiority of the proposed model and algorithm.This work is supported by National Natural Science Foundation of China(No.51177107).