Petri网结合粗糙集的变电站故障诊断方法的研究

发布时间：2018-06-14 12:11

本文选题：变电站 + 故障诊断　；参考：《华北电力大学》2015年硕士论文

【摘要】：变电站(substation)是电力系统中对电能的电压和电流进行变换、集中和分配的场所,在输配电系统中具有极其重要的地位。因此当变电站发生故障时,需要相关运维人员能够快速准确地定位故障区域,识别真正的故障元件并将其隔离,从而快速恢复非故障区域的正常运行。然而,由于系统发生复杂的多重故障、断路器或保护存在较多误动或拒动、信道干扰造成的信息丢失等因素的影响,报警信息在后台监控中的显示存在很多问题,不利于故障的判断。因此,研究出一种受错误信息干扰小、容错力强、准确快速的变电站故障诊断方法具有重要的意义。本文在对以往的变电站以设备故障的诊断方法和技术进行详细的归纳总结,并且在总结基础上,通过对Petri网(以提出者Carl Adam Petri命名的理论)、粗糙集理论、人工神经网络等理论的研究,结合目前我国的变电站的故障诊断技术发展趋势提出了粗糙集和Petri网相互弥补的新型变电站故障诊断技术。其次,由于大部分变电站都存在设备分期建设投产的情况,并且一次扩建投产通常只涉及到一个电压等级,因此可以根据电压等级将变电站划分为数个相对独立的区域。在变电站的接线结构或保护配置改变时,针对发生变化的区域重新构建诊断模型,其他区域不受影响。然后对于各个区域分别建立故障决策表,进行粗糙集决策表的属性和属性值约简,从而获得最小的诊断规则。因为Petri网具有推理搜索快速化、诊断过程数学化、结构表示图形化的特点,本文采用Petri网对诊断规则进行描述,从而建立各个区域的诊断模型,实现了变电站故障的高效诊断。最后,将以110kV仁舍变电站作为研究的案例,采用Petri网和粗糙集结合的分区域故障诊断方法,建立各个区域的故障诊断模型。然后,进行了模型的样本测试和算法的分析比较。结果表明：这种方法在处理问题的速度方面和对于变电设备的故障诊断效率方面都有着极大的优势,能够提高处理效率,保障变电站运行的安全运行。
[Abstract]:Substation is the place where the voltage and current of electric energy are transformed, centralized and distributed in power system, which plays an extremely important role in the transmission and distribution system. Therefore, when the substation fails, it is necessary for the relevant operators to locate the fault area quickly and accurately, identify and isolate the real fault elements, so as to quickly restore the normal operation of the non-fault area. However, there are many problems in the display of alarm information in background monitoring because of the complex multiple faults of the system, the maloperation or rejection of circuit breaker or protection, the loss of information caused by channel interference, and so on. It's not good for fault judgment. Therefore, it is of great significance to develop an accurate and fast fault diagnosis method for substations with small error information interference, strong fault tolerance and high accuracy. In this paper, the methods and techniques of fault diagnosis for substation equipment are summarized in detail, and on the basis of the summary, the rough set theory, named after the name of Adam net, is introduced. According to the development trend of substation fault diagnosis technology in China, a new fault diagnosis technology based on rough set and Petri net is proposed. Secondly, because most substations are built and put into production by stages, and only one voltage level is usually involved in an extension, the substation can be divided into several relatively independent regions according to the voltage level. When the connection structure or protection configuration of the substation is changed, the diagnostic model is re-constructed for the changed area, and the other areas are not affected. Then the fault decision table is established for each region, and the attribute and attribute value reduction of the rough set decision table is carried out to obtain the minimum diagnosis rule. Because Petri nets have the characteristics of fast reasoning search, mathematical diagnosis process and graphical structure representation, the diagnosis rules are described by Petri nets, and the diagnosis models of each region are established, and the efficient diagnosis of substation faults is realized. Finally, taking 110kV Renshe substation as a case study, the fault diagnosis model of each region is established by using Petri net and rough set combined sub-region fault diagnosis method. Then, the sample test of the model and the analysis and comparison of the algorithm are carried out. The results show that this method has great advantages in dealing with the problem speed and fault diagnosis efficiency of substation equipment. It can improve the processing efficiency and ensure the safe operation of substation operation.
【学位授予单位】：华北电力大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TM63;TP18

【参考文献】