基于风光数据驱动的概率电压稳定评估算法

Probabilistic voltage stability evaluation algorithm based on data-driven of wind speeds and solar radiation

针对风速和光照强度的相关性难以准确描述和概率电压稳定计算耗时长等问题,提出一种基于风光数据驱动的概率电压稳定评估算法。以风速、光照强度等历史数据作为输入,通过贝叶斯网络深入挖掘输入数据之间的相关关系并建立基于数据驱动的输入概率模型;采用随机响应面法构建电压稳定分析的替代模型,基于替代模型进行大规模概率电压稳定计算,进而大幅提升概率分析效率。实验结果表明:所提算法能准确考虑风光之间的复杂相关性,且计算效率高。

To tackle the difficulty of accurately describing the correlation between wind speed and solar radiation and the time-consuming of probabilistic voltage stability computation,a probabilistic voltage stability evaluation algorithm based on data-driven wind speed and solar radiation is proposed.The historical records of wind speed and solar radiation are applied as the input data,and the correlation between the input data is deeply mined by using Bayesian network,and then a data-driven input probability model is established.A substitution model for voltage stability analysis is constructed by using stochastic response surface method,then the substitution model is used in massive probabilistic voltage stability computation,thus greatly improving the computational efficiency of probability analysis.The experiment results show that the proposed algorithm is effective and can accurately describe the complex correlation between wind speed and solar radiation with high computational efficiency.