A bootstrap data methodology for sequential hybrid engine model building

The current evolution in on-board Propulsion Health Management (PHM) systems aimed at performing real-time engine diagnostics and prognostics has placed a greater demand on model accuracy and implementation speed. The complexities in assembling accurate physics based models for real-time operation has placed greater focus on the use of hybrid engine models employing some form of empirical modeling. A practical consideration for implementing a hybrid engine model that incorporates both physics-based and empirical components, involves the application of sequential model building for the construction of the empirical elements. This arises for the reason that sufficient engine data required to model the entire flight regime for a given engine/aircraft application is never available from one flight alone and may takes days or weeks to assemble. This paper discusses a methodology for constructing the empirical model in a sequential manner without the requirement for storing all of the original data. The method involves sequentially developing and storing a compact statistical and parametric representation of the data, as it is collected and generating representative pseudo-data samples from these models to be used in a piecewise model building process to bootstrap the model building process.