Predictive Maintenance

Publisher Summary This chapter discusses the application of Grid technologies to the challenging and broadly important problem of computer-based fault diagnosis and prognostic (DP). It describes a UK e-Science project that is applying Grid technologies to the problems of diagnosing faults in Rolls-Royce aircraft engines based on sensor data recorded during flight. Fault diagnosis is fundamentally based on the monitoring and analysis of sensor data through the application of declarative and procedural knowledge. Data from sensors must be captured, often in real time, and made available to analysis systems, either remote or local. Root cause determination and prognosis may require integrating data from several different systems to build a pattern or a case that is reusable in subsequent diagnoses. To extend the capabilities of fault DP systems, it is also beneficial to archive data such that an operational log of system performance or fault conditions can be maintained.