Lifetime model of the inverter-fed motors secondary insulation by using a design of experiments

Numerous parameters (environmental conditions, material and equipments design) affect the ageing (and consequently the lifetime) of the Electrical Insulation System (EIS) of motors supplied by static converters. In order to find a modeling of the lifetime of such materials, accelerated tests have to be considered. The problem is that if we consider all the relevant parameters, these tests become costly and take a lot of time. Consequently the topic of this study is to use a statistical method called: the `Design of Experiments' (DoE) to obtain the lifetime modeling by only a `reasonable' number of experiments. Only electrical and thermal parameters will be modified in this study (vibrations and thermal cycling are not taken into account). In this work, the ageing of the secondary insulation of a motor fed by an inverter is supposed to be originated from partial discharges. In this kind of machine, if a partial discharge activity begin to grow, the first insulation link of the EIS to be affected by these discharges will be the secondary one (i.e.: the impregnation varnish). Consequently, this study is focused only on this secondary insulation. The results obtained by using the DoE method show that it is possible to classify the different ageing parameters and to find their effects on the insulation lifetime. The interactions between all the parameters will also be shown by this method. Finally, a mathematical model of the insulation lifetime taking into account all the parameters and their interactions will be presented and discussed.