电网数据可信性度量模型研究

Research on Credibility Measurement Model of Power Grid Data

针对电力系统中不可信数据将导致电力系统状态估计结果的准确性降低,分析了电网中数据的可信性需求,研究了电力系统不良数据的辨识和数据可信分析理论,构建了层次化、动态化的电网数据可信分析模型。该模型引入时间因子、惩罚因子等权重参数,将电力数据的可信性分析问题归结为数据源、数据源之间及数据传播网络路径的组合问题,即通过计算数据源之间的可信度、数据源的可信度、数据的可信度,动态地构建了基于数据源依赖关系的可信虚拟网络,最后通过所构建的可信虚拟网络来评估出电力数据的可信度。仿真实验结果表明,该模型能较好地满足电网数据的可信性评估需求,为可信性评估方法进一步研究提供了解决思路。

The untrusted data in power system reduces the accuracy of the state estimation results. The credibility requirement of data in the power grid is analyzed and identification of bad data in the grid and the theories of credible data are studied. On the basis of these studies,a hierarchical and dynamic credibility analysis model for grid data was established which takes into account weight factors such as time factor and penalty factor. Credibility analysis of power data can be summarized as combinatorial problems including data source,relations between different data sources and network path of data dissemination. The virtual network of the credibility which was on the base of dependent relationship of data source was dynamically established by calculating the credibility between data sources,the credibility of data and the credibility of data source. And the overall credibility of data is normally evaluated by the virtual network.The simulation results show that the model can satisfy the requirement of the credibility estimation of grid data,and also provide ideas to solve the problem of the credibility measurement for further research.