Peter Myler

This paper aims to use machine learning methods to predict the service life of rotary lip seals to aid manufacturers and users improving the current maintenance procedures. Seals are widely used in most engineering applications. The knowledge of condition of seals throughout their working life is important due to the fact that they are often used on high value engineering products. As the current material properties of the seal and the working environment various, it is difficult to predict useful life of the rotary lip seal. In this paper, the factors relating to life of rotary lip seals are investigated and discussed. The application of machine learning methods using actual testing data in order to estimate the useful life of the seals has been presented. The early results show good agreement between actual and predicted values.