Approaches to Dynamic Tribosystem Modelling

Predictive models for friction and wear are among the most important issues setting guidelines for tribological research [5]. The analysis of published models of tribological phenomena reveals, that the achievement of satisfactory predictive accuracy of the proposed models is still problematic [10, 12]. Practically speaking the mathematical models mentioned have a limited usability for the purpose of designing new machines. Two groups of models used in tribology can be distinguished: design practice oriented and fundamental [10]. The model belonging to the first group is built so it calculates the expected behaviour of the tribological system on the basis of input data. The core of the model is a ‘black box’ the characteristics of which is based on experience. Data gathered in laboratory and field experiments is used by the designer to predict friction and wear on the design stage of a real life tribosystem. Such a model will nowadays have a form of a computer program allowing to calculate wear rates etc. Whereas the fundamental model quantifies various physical processes influencing friction and wear in the analysed system.