SIGNIFICANCE OF UNCERTAINTY QUANTIFICATION IN PROGNOSTICS AND HEALTH MONITORING

摘要

This paper discusses the significance of uncertainty quantification in prognostics and health monitoring, which are important for failure prevention and risk-informed decision-making with respect to the operation of engineering systems. Since there are several sources of uncertainty that affect the future loading and operating conditions of these engineering systems, it is important characterize and quantify these sources of uncertainty, and quantify their combined effect on future prediction. Failure threshold levels are defined and used to quantify the remaining useful life, which is also uncertain due to the aforementioned sources of uncertainty. It is important to quantify the overall uncertainty in the remaining useful life using principles of probability, since this information is directly useful for risk-informed decision-making. Methods for such uncertainty quantification are discussed and illustrated using a numerical example.