面向智能电网的大数据可信度量方法研究

发布时间：2018-03-21 19:45

本文选题：智能电网　切入点：可信度量　出处：《华北电力大学》2017年硕士论文　论文类型：学位论文

【摘要】：大数据技术作为国内外学术界的研究热点,主要集中在大数据采集、预处理、分析及挖掘、展现等方面。目前数据预处理中大数据可信计算方法研究较少,亟需研究大数据可信性度量新方法、新技术。本文在大数据的可信性和质量问题上展开研究。研究了数据源依赖关系、数据源之间的行为特点以及计算数据源之间的可信度值方法,提出了可信虚拟网络概念,这种网络不同于传统互联设备构成的网络,其通过度量数据源之间的可信度而建立的层次化虚拟网络,权重值为数据源之间的可信度值。本文构造了可信虚拟网络生成算法,创新性地提出基于数据源依赖关系的层次化可信虚拟网络模型。本文研究了基于大数据的数据源之间的可信性度量、数据源的可信性度量以及数据的可信性度量,它们之间相互关联、相互制约构成一个整体,创新性地提出大数据可信性度量模型。数据源间的可信度取决于数据源间的本地可信度与全局可信度。数据源的可信度是由数据源历史数据的可信度期望值与推荐可信度期望值组合而成的。数据的可信度是通过计算不可靠数据对立事件的概率得出。数据源之间的可信度受数据源的可信度制约,数据源的可信度受数据的可信度和数据源之间的可信度双重制约,数据的可信度受数据源的可信度和数据源之间的可信度双重制约。本文将大数据可信度量模型引入智能电网中,给出智能电网大数据可信性度量方法。以电力系统大数据为例,验证可信性度量模型的有效性,满足智能电网大数据的可信需求,为电力系统信息控制与决策提供可靠的数据支持。形成具有自主知识产权的大数据可信性度量方案,为我国今后大数据技术和可信性研究达到国际领先水平提供技术支撑。
[Abstract]:Big data technology, as a hot research topic in academic circles at home and abroad, mainly focuses on the collection, preprocessing, analysis, mining and presentation of big data. At present, there are few researches on trusted computing methods of large data in data preprocessing. There is an urgent need to study new methods and techniques of big data credibility measurement. This paper presents the concept of trusted virtual network, which is different from the network composed of traditional interconnected devices, based on the behavior characteristics of data sources and the method of calculating the reliability value between data sources. The hierarchical virtual network is built by measuring the credibility between data sources, and the weight value is the credibility value between the data sources. In this paper, a trusted virtual network generation algorithm is constructed. This paper proposes a hierarchical trusted virtual network model based on data source dependency. This paper studies the credibility metrics between data sources based on big data, the credibility metrics of data sources and the credibility metrics of data sources. They are interlinked and mutually constrained to form a whole, The credibility between data sources depends on the local and global credibility of the data sources. The credibility of the data source is derived from the reliability expectation and the extrapolation of the historical data of the data source. The reliability of the data is obtained by calculating the probability of the opposing events of unreliable data. The credibility between the data sources is restricted by the credibility of the data sources. The credibility of a data source is constrained by both the credibility of the data and the credibility of the data source. The reliability of data is restricted by the reliability of the data source and the credibility between the data source and the data source. In this paper, big data's credibility model is introduced into the smart grid, and then a method to measure the trustworthiness of the smart grid is given. Verify the validity of credibility measurement model, satisfy the credible demand of big data in smart grid, and provide reliable data support for power system information control and decision-making. To provide technical support for big data technology and credibility research to reach the international leading level in the future.
【学位授予单位】：华北电力大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TM76;TP311.13

【参考文献】