Application of statistical methods in the diagnosis of electric plastic insulating materials used in electric machines

Abstract

As we know, the use of electricity is one of the most important issues today. The requirements set in this regard are related to the reliability, physical characteristics and cost of various types of electrical machines and devices. These listed requirements depend fundamentally on the materials used in their preparation and their technical resources. It is shown in the work that in the production of Electric machines (EM), mainly, the test-diagnostic method is used, and in their operation, both functional (in working mode) and test (after repair) diagnostic methods are used. "Destructive" methods are accepted as the most reliable and correct method during test-diagnostics.

The article examines the shortcomings of "destructive" methods. It was mainly shown that the majority of punctures in plastic insulation materials occurred during handover tests. This, in turn, is due to the excessively high selection of the test voltage.

The specified procedures of the isolation system allow to predict their reliability. In the article, as a result of testing a limited number of samples (selection tests), the issues of statistical control of the quality of the object were investigated and some mathematical models were given for stationary states of the controlled parameter.