基于自适应算法的电力系统可靠性评估

针对电力系统可靠性评估中Monte Carlo方法存在的计算效率低下的问题,提出应用自适应算法对系统状态进行概率分析。该算法采用变分运算分析和区间拟合的方法,实现在最优密度函数下抽样,降低计算方差。应用该方法对IEEE-RTS标准系统的发电部分进行了可靠性评估,并与采用常规抽样方法和重要抽样方法的评估结果做了比较,表明该算法在保证计算精度的前提下分别减少了87%和68%的抽样次数,对大型电力系统可靠性评估具有实用价值。

Power system reliability evaluation based on an adaptive algorithm

In view of the low efficiency of Monte Carlo simulations of power system reliability evaluation,an adaptive algorithm was developed for probabilistic system simulations.The algorithm samples the system status from the optimal probability density and reduces the sample variance by using the variation principle and interval fitting.The method was used to evaluate the reliability of the power generation section in the IEEE-RTS test system.Comparison with a conventional sampling method and the importance sampling methods show that for the same calculational accuracy,the current method requires 87% less samples than the conventional sampling method and 68% less samples than the importance sampling method,which verifies its applicability to power system reliability evaluations.