A RECONFIGURABLE EMBEDDED PROGNOSTICS PLATFORM FOR MACHINERY PERFORMANCE MANAGEMENT

摘要

This paper presents the design and implementation of a reconfigurable embedded prognostics platform (REPP) for machine performance management. The platform contains (1) a toolbox of computational prognostic algorithms for predicting degradation of devices and systems, (2) agent-based adaptive software architecture and a reconfigurable hardware platform for different production and field service applications. An industrial case is presented to illustrate the machine performance management application of the proposed platform. This platform has also been tested on several industrial test-beds and can be reconfigured for diverse machine prognostics applications.