TY - RPRT
T1 - EU-project Mantis: D4.5 Final report on wear-out models to predict remaining life time
AU - Hooghoudt, Jan Otto
AU - Yi, Yang
PY - 2018/5/1
Y1 - 2018/5/1
N2 - The report identifies and describes wear-out models used for predicting the remaining useful life time of
the various components and assets considered in the various use-cases. The deliverable is part of Task 4.2
which has the aim to identify and develop methods for forecasting the lifetime of an asset by modelling
its operational dynamics and by extrapolating this data into the future. When the extrapolated data
reaches a certain threshold that marks the end-of-life of an asset, that point in time can be used as the
forecasted remaining useful life. Simultaneously, the extrapolated data gives an indication of the expected
wear and tear, which can be compared with the observed wearing out to monitor the behavior of an asset.
Based on the predicted remaining lifetime, suitable maintenance tasks can be planned, in order to prevent
unscheduled system down time. First a classification system of RUL prediction models is stated and for
each use-case the developed RUL modeling techniques and algorithms are then presented according to
the class of model they mainly apply to. The class of models consists of: Physical Modeling, Artificial
Neural Networks, Life Expectancy Models and Expert Knowledge Systems. Each use-case presents their
developed RUL modeling techniques and algorithms by the following outline: 1) Introduction and
Objectives, 2) Methodology, 3) Data & Data preparation, 4) Results, and 5) Conclusions.
AB - The report identifies and describes wear-out models used for predicting the remaining useful life time of
the various components and assets considered in the various use-cases. The deliverable is part of Task 4.2
which has the aim to identify and develop methods for forecasting the lifetime of an asset by modelling
its operational dynamics and by extrapolating this data into the future. When the extrapolated data
reaches a certain threshold that marks the end-of-life of an asset, that point in time can be used as the
forecasted remaining useful life. Simultaneously, the extrapolated data gives an indication of the expected
wear and tear, which can be compared with the observed wearing out to monitor the behavior of an asset.
Based on the predicted remaining lifetime, suitable maintenance tasks can be planned, in order to prevent
unscheduled system down time. First a classification system of RUL prediction models is stated and for
each use-case the developed RUL modeling techniques and algorithms are then presented according to
the class of model they mainly apply to. The class of models consists of: Physical Modeling, Artificial
Neural Networks, Life Expectancy Models and Expert Knowledge Systems. Each use-case presents their
developed RUL modeling techniques and algorithms by the following outline: 1) Introduction and
Objectives, 2) Methodology, 3) Data & Data preparation, 4) Results, and 5) Conclusions.
M3 - Report
BT - EU-project Mantis: D4.5 Final report on wear-out models to predict remaining life time
ER -