具有可变规则的专家系统在输电线路故障诊断与监测中的应用研究

【作者】 乔飞；

【导师】 王钦友；

【作者基本信息】 南京理工大学 ， 控制理论与控制工程， 2007， 硕士

【摘要】 随着电力系统规模的日益扩大，对输电线路的安全运行和供电可靠性的要求越来越高。高压输电线路作为输电系统中的枢纽干线，在电力系统中起着重要作用，其故障直接威胁着现代电力系统的安全运行。当输电线路发生故障时，在很短的时间内会有大量的报警信息，如果仅凭运行人员的经验进行故障诊断，其快速性和正确性难以保证。因此研究智能化的高压输电线路故障诊断系统，对电力系统的安全运行具有重要意义。专家系统是人工智能中最重要的也是最活跃的一个应用领域。专家系统是一个智能计算机程序系统，其内部含有大量的某个领域专家水平的知识与经验，能够利用人类专家的知识和解决问题的方法来处理该领域问题。论文首先论述了课题研究的背景及意义，接着对专家系统的方法进行了概述，介绍了专家系统的定义、结构以及各组成部分的功能。然后根据输电线路故障类型的特点，在分析故障诊断专家系统各个关键技术的基础上，结合人工智能、数据库技术和模糊理论，建造了具有可变规则的输电线路故障诊断专家系统实用模型。由于输电线路故障信息存在着不确定性，论文引入了模糊规则，在不精确推理理论的基础上，提出了用隶属度来表示事实的不确定性，实现了基于规则和可信度的不确定性传播算法，并且根据诊断结果不断调整规则，实现系统的自学习能力。同时该专家系统采用面向对象的语言进行开发，用数据库存储知识，使知识库独立于推理机，并设计了良好的知识库管理界面，有效的解决了知识获取的瓶颈问题，同时该模型具有很好的模块性和扩充性。更多还原

【Abstract】 With the development of power system, the demanding on the transmission line’s safeoperation and the reliability of power supply is increasingly high. As the hub of powerdelivery system, high-volt transmission-line plays an important role so that its fault willthreaten the safety of modern power system. When a fault in transmission-line occurs therewill be so much alarm information entering into control center in a short time that itbecomes very hard for operators to identify fault merely by their own experience. In thesecircumstances, it is absolutely necessary to develop an intelligent fault diagnosis systemsfor high-volt transmission-line because for one thing, this system can realize quickdiagnosis for transmission line fault to minimize loss of power blackout, for another,intelligent system also acts as a very useful assistant to operators when dealing withcomplicated fault.One of the most important and active in artificial intelligence application areas isexpert systems(ES). ES is computer program embodying much knowledge or experience ina certain domain for solving problems related to that domain.The paper firstly outlines the background and significance of the research, thenintroduces the definition, the structure of ES and the features of its different compositions.And according to the features of the transmission line’s fault type, basing on the analysis ofthe key technologies of expert systems, and combining with artificial intelligence, databaseand fuzzy theory, this paper construct a general fault diagnosis expert system model basedon variable rules. As the transmission line fault exists uncertain information, the paperintroduces the fuzzy rules, and on the base of imprecise reasoning, the paper proposes anew approach that we can use credibility to express the uncertain of the fact, and achieveuncertainty propagation algorithm to calculate the spread of uncertainty based on the rulesand the credibility, and the same time, with diagnosis results to adjust the rules to achievethe self-learning ability. Lastly, the fault diagnosis expert systems was designed usingobject oriented programming, with the database to storage knowledge, making theknowledge management system be independent of the Inference mechanism, and highperformance knowledge acquisition module is constructed, this model has the bettermodularity and expandability.更多还原