适合于分析广义负荷序列间相关关系的最优延位法

传统的相关性分析方法局限于获得明显的相关关系,难以挖掘序列之间潜在的相关性信息,使得电力系统调度运行的参考信息受损。针对此问题,提出一种用于分析广义负荷序列之间潜在相关关系的最优延位法,该方法通过对序列进行适当延位,挖掘出广义负荷序列之间的间接相关关系。首先,以获得序列间最大Pearson相关系数为目标函数,将位移范围与序列数据的时间单调性作为约束条件,建立最优延位模型;然后,提出了模型的解算策略;最后,以德国2016年区域数据为例,分别对单日、每月和全年数据的相关关系进行分析计算,统计得到最大相关系数与需要的位移时间。分析结果表明,与传统的相关性分析方法相比,所提最优延位法可以发现序列之间潜在的延位相关关系,完善了风电、光伏和负荷等广义负荷序列之间的相关性分析方法。

Optimal Time-delay Method for Analyzing Correlation Between Generalized Load Sequences

The traditional method of correlation analysis is limited to obtaining the superficial correlation while incapable of mining the potential correlation information between the sequences, but leading to the loss of the reference information provided to power system for operation and dispatching. For this reason, an optimal time-delay method is proposed to analyze the potential correlation between generalized load sequences. By delaying the sequences properly, the proposed method digs out the indirect correlation between generalized load sequences. Firstly, the time shift range and time monotonicity of sequences are regarded as constraints, then the objective function is built to obtain the maximum Pearson correlation coefficient. Secondly, the calculation strategy of the model is proposed. Finally, the data of a German region is taken to act as an example to analyze and calculate the correlation of each day, each month and whole year respectively, and then the maximum correlation coefficient and the required shift time are statistically obtained. The analysis results show that, compared with the traditional analysis method, the proposed method can dig out the potential delay correlation between sequences. As a result, the method of correlation analysis of generalized sequences, which includes wind energy, photovoltaic energy, load and etcetera, is perfected and provides more comprehensive reference information to power system for operation and dispatching.