摘要
目前解决不确定性潮流问题的主要方法是先对不确定量进行预测,再根据预测结果进行分析决策,但预测与决策分离可能会导致次优解。将预测过程融入求解不确定性潮流问题的决策过程中,提出基于Predictive&Prescriptive框架的鲁棒最优潮流模型。利用k近邻算法预测并构造表示风电功率不确定性的最小体积椭球集,建立考虑风电不确定性的鲁棒二阶锥最优潮流模型,并利用对偶理论将该模型转换为可求解的混合整数规划模型进行高效求解。IEEE 14和IEEE 118节点系统算例仿真结果表明,所提模型能有效降低预测、决策过程分离时最优解的劣化程度,在保证系统安全运行的前提下提高系统经济性。
The current main approach to solve uncertain power flow problems is to predict the uncertainty quantity firstly,and then analyze and make decisions based on the predicted results,but the separation of prediction and decision-making may lead to suboptimal solutions.The prediction process is embedded to the decision-making process of solving the uncertain power flow problems,and a robust optimal power flow model based on Predictive&Prescriptive framework is proposed.The k-nearest neighbor algorithm is used to predict and construct a minimum volume ellipsoid set representing the uncertainty of wind power,a ro⁃bust second-order cone optimal power flow model considering wind power uncertainty is established,and the duality theory is used to transform the model into a solvable mixed-integer programming model for effi⁃cient solution.The case simulative results of IEEE 14-bus and IEEE 118-bus systems show that the pro⁃posed model can effectively reduce the deterioration degree of the optimal solution when the prediction and decision-making processes are separated,and improve the system economy under the premise of ensuring the safe operation of the system.
作者
郑丽琴
谢东梅
白晓清
ZHENG Liqin;XIE Dongmei;BAI Xiaoqing(Guangxi Key Laboratory of Power System Optimization and Energy Technology,Guangxi University,Nanning 530004,China)
出处
《电力自动化设备》
EI
CSCD
北大核心
2023年第7期175-181,共7页
Electric Power Automation Equipment
基金
国家自然科学基金资助项目(51967001)
广西研究生教育创新计划项目(YCSW2022119)。
