Model-based Prognostics for Prediction of Remaining Useful Life

Abstract: Prognostics and health management (PHM) is an engineering discipline that aims to maintain the system behaviour and function, and assure the mission success, safety and effectiveness. Health management using a proper condition-based maintenance (CBM) deployment is a worldwide accepted technique and has grown very popular in many industries over the past decades. These techniques are relevant in environments where the prediction of a failure and the prevention and mitigation of its consequences increase the profit and safety of the facilities concerned. Prognosis is the most critical part of this process and is nowadays recognized as a key feature in maintenance strategies, since estimation of the remaining useful life (RUL) is essential. PHM can provide a state assessment of the future health of systems or components, e.g. when a degraded state has been found. Using this technology, one can estimate how long it will take before the equipment will reach a failure threshold, in future operating conditions and future environmental conditions. This thesis focuses especially on physics-based prognostic approaches, which depend on a fundamental understanding of the physical system in order to develop condition monitoring techniques and to predict the RUL. The overall research objective of the work performed for this thesis has been to improve the accuracy and precision of RUL predictions. The research hypothesis is that fusing the output of more than one method will improve the accuracy and precision of the RUL estimation, by developing a new approach to prognostics that combines different remaining life estimators and physics-based and data-driven methods. There are two ways of acquiring data for data-driven models, namely measurements of real systems and syntactic data generation from simulations. The thesis deals with two case studies, the first of which concerns the generation of synthetic data and indirect measurement of dynamic bearing loads and was performed at BillerudKorsäs paper mill at Karlsborg in Sweden.In this study the behaviour of a roller in a paper machine was analysed using the finite element method (FEM). The FEM model is a step towards the possibility of generating synthetic data on different failure modes, and the possibility of estimating crucial parameters like dynamic bearing forces by combining real vibration measurements with the FEM model. The second case study deals with the development of prognostic methods for battery discharge estimation for Mars-based rovers. Here physical models and measurement data were used in the prognostic development in such a way that the degradation behaviour of the battery could be modelled and simulated in order to predict the life-length. A particle filter turned out to be the method of choice in performing the state assessment and predicting the future degradation. The method was then applied to a case study of batteries that provide power to the rover. Keywords:  Prognostics, Degradation, FEM, Modelling, Particle Filter, CBM, RUL