Diagnostics of the technical condition of electric network equipment based on fuzzy expert estimates

Abstract

The paper describes a new possible method of diagnostics of the current technical condition of equipment using a mathematical model based on fuzzy expert estimates and the theory of fuzzy sets. The specifics of the task is determined mainly by the type of the obtained estimates, namely: causal relationships between the controlled parameters of the transformer equipment and defects that could entail their change and the possibility of further operation of the facility. At the same time, attention is paid to the problem of the degree of consistency of expert opinions that affects the quality of the assessment of the current technical condition of the studied object. The paper provides a comparative analysis of the arithmetic mean estimates and median estimates of the consistency of expert opinions. It is shown that the significant drawback of the arithmetic mean approach is its instability towards outliers of individual opinions moving the resulting value under the influence of the “dissident expert opinions”. On the other hand, the median estimate is free of such shortage; it is more outlier-resistant and simply discards a part of radically outlying expert opinions. For the first time, the Kemeny median has been used for technical diagnostics. Kemeny median is based on the introduction of a metric to the set of expert opinions, and axiomatic introduction of the distance between them. Also, the paper formulates a criterion on how to determine the optimal number of experts in the group.