一种基于拉丁超立方采样的概率最优潮流算法

在概率最优潮流的求解技术中,随机采样的蒙特卡罗法因其在大规模采样的情况下求解精确度高,而被广泛应用。本文采用拉丁超立方采样和蒙特卡罗法相结合的技术处理含多随机变量因素的概率最优潮流问题,并将其运用于分析随机变量的波动对系统发电成本影响的计算中。通过IEEE-14和IEEE-118节点测试算例的分析表明,采用拉丁超立方采样能改善采样值的分布空间,在采样规模较低的情况下能够给出精确的统计结果,较随机采样的蒙特卡罗法具有应用优势,可以替代随机采样的蒙特卡罗法,作为评价其他算法优劣的标准。

A Probabilistic Optimal Power Flow Calculation Method With Latin Hypercube Sampling

Among the solving methods of probabilistic optimal power flow, Monte Carlo Simulation (MCS) combined with random sampling is widely used because of its accuracy. In order to further improve the solving accuracy, Latin hypercube sampling (LHS) with Monte Carlo Simulation was applied to calculate the consumption of system power generating cost under many random variables. Numerical results of IEEE 14-bus and IEEE 118-bus systems show that comparing with random sampling method and Latin hypercube sampling method, the Latin hypercube sampling method provides more accurate performance in dealing with probabilistic optimal power flow under the condition of a lower sampling size, thus the Latin hypercube sampling method can replace the MCS as the standard method of other algorithms.