摘要
提出了求解电力系统最优潮流问题新的非线性多中心校正内点算法。该算法采用仿射方向作为预测方向,在校正方向上增加了权系数,并通过线性搜索方法确定权系数的最优值,在预测方向和校正方向的组合方向上获得最大的迭代步长值;同时通过检验校正后的方向是否落在中心轨迹的对称邻域内来保证算法的收敛性。算法能够通过单次校正获得较大的计算步长,从而提高了计算的速度。该算法与预测校正内点法相比具有鲁棒性好、收敛快速的优势,特别是在计算过程中互补对差值较大的恶劣条件下。通过对多个测试系统的仿真,结果验证了算法有效性。
A novel nonlinear multiple centrality corrections interior point algorithm is presented to solve optimal power flow problems of power systems in this paper. The original affine-scaling direction is used as the predictor direction and a weighting coefficient is added into the corrector direction. To obtain the largest step-length in the combination of the predictor direction and the corrector direction, the optimal value of the coefficient is chosen by line search method. The convergence of the algorithm is guaranteed by checking whether the corrected directions are in the symmetric neighborhood of the centering direction. Since the proposed algorithm can obtain a larger computing stepsize by only single corrector, the computing time is saved. Comparing with the predictor-corrector interior point method, this algorithm is faster and robuster especially under the bad circumstance of large difference complementarity pairs during the solving process. The proposed method was tested on five systems and the numeric results demonstrated its validity.
出处
《电工技术学报》
EI
CSCD
北大核心
2007年第12期133-139,共7页
Transactions of China Electrotechnical Society
基金
国家自然科学基金重点项目(50337010)
广东省自然科学基金资助项目(06025630)
关键词
最优潮流
非线性内点法
多中心校正
预测方向
校正方向
Optimal power flow, nonlinear interior point method, multiple centrality correction, predictor direction, corrector direction
