Real-time hardware implementation of symbolic health monitoring for aircraft engine components

Symbolic dynamics based fault detection and condition monitoring is a new research area that provides a new set of tools for data analysis, fault detection and investigation of the dynamical systems. The partitioning concept is used to obtain a finite number of non-overlapping segments for a low-dimensional representation of time series. This analysis method for fault progression can be constructed from observed data sequences. This paper presents hardware implementation of symbolic dynamic filtering technique for real-time health monitoring of aircraft engine components using piezoelectric sensors. The signal processing tools are implemented in a two-step procedure using MATLAB/Simulink. The Simulink implementation gives a direct code generation capability that enables the generated code to be executed on the microprocessor for real-time applications.