大电网多目标动态优化调度的解耦算法及并行计算  

Decomposition Algorithm for Multi-objective Dynamic Optimal Dispatch of Large-Scale Power Systems and Parallel Computing

作　　者：杨柳青[1] 林舜江[1] 刘明波[1] 

机构地区：[1]华南理工大学电力学院,广州510640

出　　处：《电工技术学报》2016年第6期177-186,共10页Transactions of China Electrotechnical Society

基　　金：国家高技术研究发展计划(863计划)(2012AA050209);国家自然科学基金(51207056);中央高校基本科研业务费(2015ZM106)资助项目

摘　　要：针对风电规模化接入,兼顾发电总燃料耗量、污染气体排放量和购电费用最小的大电网多目标动态优化调度模型,提出基于法线边界交叉(NBI)法和原对偶内点法的多目标解耦算法及并行计算。该算法首先根据NBI法的基本原理将多目标优化问题转换为一系列单目标优化问题,再根据内点法求解这类单目标优化问题时形成的修正方程系数矩阵的特殊结构,对修正方程进行解耦降阶处理,实现静态变量和动态变量的解耦分离,最后借助Matlab并行计算平台实现并行计算,快速有效地得到一系列均匀分布的Pareto最优解,为运行人员提供决策指导。在某省级电力系统上的计算结果表明,所提算法能够快速、有效地获取多目标动态优化调度问题的Pareto最优解集,具有很好的实用价值。This paper proposes a multi-objective decomposition algorithm and parallel computing, based on normal boundary intersection （NBI） method and prime-dual interior point method, to minimize fuel consumption, emission of atmospheric pollutants and power purchase for large-scale power systems considering wind power penetration. The proposed algorithm firstly transforms a multi-objective optimization problem into a series of single objective optimization problems, and then decouples and reduces the correction equations of these single objective optimization problems based on the special structure of their coefficient matrices, at last decouples static and dynamic variables . Finally, a series of uniform distributed Pareto optimal sets, providing guidance for operators, is obtained rapidly by Matlab parallel computing platform. Computational results on a real provincial power system demonstrate that the proposed algorithm can obtain the Pareto optimal sets rapidly and efficiently to the multi-objective dynamic optimal dispatch problem.

关 键 词：动态优化调度 多目标 法线边界交叉法 内点法 解耦 并行计算 

分 类 号：TM734[电气工程—电力系统及自动化]