大规模校正性安全约束最优潮流问题的近似牛顿方向分解协调算法  

An Approximate Newton Directions Based Decomposition and Coordination Algorithm for Large Scale Corrective Security-constrained Optimal Power Flow

作　　者：赵启飞[1] 

机构地区：[1]柳州职业技术学院,广西柳州545006

出　　处：《广西电力》2017年第6期24-30,共7页Guangxi Electric Power

基　　金：广西高校中青年教师基础能力提升项目(No.KY2016YB637;No.2017KY1059)

摘　　要：提出了在大规模安全约束最优潮流(Security-constrained Optimal Power Flow,SCOPF)问题下应用近似牛顿方向(Approximate Newton Directions,AND)分解协调算法。该算法利用近似牛顿方向直接对原问题Karush-Kuhn-Tucker(KKT)条件解耦的思想,将校正性安全约束最优潮流问题分解为一个正常运行状态子问题和N个故障状态子问题,N为预想故障数。分解的每个子问题只迭代1次而不用求其最优解,避免了参数的人为调整,极大地提高了计算效率。对1 047节点424个预想故障、高达4 559 128个原-对偶变量的实际系统进行计算,结果表明所提算法具有良好的适应性和稳定性,不仅显著地减少了占用内存,而且在串行求解时CPU时间下降了3~6倍,在并行计算条件下可获得10~30倍甚至200倍以上的加速比,并保证所得最优目标值与准确值的误差在10^(-8)以下,确保了分解协调结果的最优性。The application of approximate newton directions （AND） based decomposition and coordination algorithm in large scale security-constrained optimal power flow （SCOPF） problems is proposed. The algorithm, which uses the thought of decoupling the Karush-Kuhn-Tucker （KKT） condition of the original problem by approximate newton directions, decomposes the corrective security- constrained optimal power flow problem into a sub problem under normal operation condition and N sub-problems under fault conditions, in which N is the number of anticipated faults. Each sub problem decomposed only needs single iteration without searching for the optimal solution, which avoids the manual adjustment of parameters and improves significantly computational efficiency. An actual system with 1 047 buses, 424 anticipated faults and up to 4 559 128 primal-dual variables is calculate, and the results show that the proposed algorithm has good adaptability and stability which not only reduces significantly memory usage, but also decreases 3-6 times of CPU time under serial calculation. Under the condition of parallel calculation, more than 10-30 times even 200 times speedup ratio can be obtained, and meantime it can guarantee that the error between calculated optimal objective value and accurate value is less than 10-8, which guarantees the optimality of the decomposition and coordination result.

关 键 词：校正控制 安全约束最优潮流 分解协调 近似牛顿方向 

分 类 号：TM744[电气工程—电力系统及自动化]