Health and Usage Monitoring, Condition-based Maintenance and Prognostics Symposium, UK, 15-16 April 2008
Venue: DEFENCE ACADEMY OF THE UNITED KINGDOM
CRANFIELD UNIVERSITY-Defence College of Management and Technology
Course on the Use of Data- Driven Methods for Prognostics
Presented by: Michael Pecht, Center for Advanced Life Cycle Engineering (CALCE), University of Maryland.
Prognostics is a process of assessing the extent of deviation or degradation of a product from its expected normal operating conditions, and then based on continuous monitoring, predicting the future reliability of the product. If one can monitor key control signals and loads, this data can be used in conjunction with precursor reasoning algorithms and stress-and-damage models to enable prognostics. By being able to determine when a product will fail, procedures can be developed to provide advanced warning of failures, reduce life cycle costs, and improve the design, qualification of fielded and future systems.
This short course will discuss data-driven methods that enable prognostics using control and load data. Data-driven diagnostic and prognostics methods are being used today to assess parameters that are precursors to impending failure, such as shifts in performance parameters. Proven methods for analysis will be explained and example applications and case studies will be given.