A Fault Diagnosis Model for Power Transformer Using Association Rule Mining-Based on Rough Set

Abstract

With the development of on-line monitoring technology of electric power equipment, and the accumulation of both on-line monitoring data and off-line testing data, the data available to fault diagnosis of power transformer is bound to be massive. How to utilize those massive data reasonably is the issue that eagerly needs us to study. Since the on-line monitoring technology is not totally mature, which resulting in incomplete, noisy, wrong characters for monitoring data, so processing the initial data by using rough set is necessary. Furthermore, when the issue scale becomes larger, the computing amount of association rule mining grows dramatically, and its easy to cause data expansion. So it needs to use attribute reduction algorithm of rough set theory. Taking the above two points into account, this paper proposes a fault diagnosis model for power transformer using association rule mining-based on rough set.

Transformer Health Status Evaluation Method Based on the Association Rule and Analytic Hierarchy Process

“…At the same time, due to the complexity process of power transformer fault, a reflection of the measured data and its potential pitfalls transformer failure probability affiliated comprehensive analysis is also very important. Beyond the above analysis, an assessment based on transformer state association rule and AHP method was proposed [7]. Through individual state quantity of classified management, individual state quantity will contact coupling integrated into a number of comprehensive state quantity, the state quantity as an integrated whole state power transformer evaluation elements [8].…”