A Fault Information Diagnosis System In Relayprotection Based On Data Mining

Relay protection and fault diagnosis is an important guarantee to safe running of power system. With the wide application of digital technology in power system, data of power system is increased in exponential speed. Therefore, the paper introduces the fault information diagnosis system in relay protection based on data mining.After introduces the background of the question of study, this paper discusses the principal and characteristic of DW and DM at first. And then, it expatiates how to construct DW based relay protection information, from system scheme to ETL process.The data collected from reality often contain noisy, uncertainty, and incompleteness. It is not ideal when supposing with pure mathematics to eliminate uncertainty. Rough set as a classify theory of data mining is a strong mathematic tool dealing with uncertainty and incompleteness, and it can overcome the difficulty in getting knowledge existed before. Therefore, rough set is the key technology of the system. Arithmetic of data mining based rough set, in fact, it is the process of reduction to decision table. So, we provide the description and realization of reduction arithmetic.At last, it illuminates details of the system, such as, connection of mining system and data ware, discretization, of sequential data, automatics creation of decision table, and the pick-up of diagnosis rules etc.Data Mining research are on the stage of developing. A lot of problems is worthy of discussing about Data Mining itself and its application in relay protection fault information. The study of this paper is just to have a try, and relevant work needs further study.