Real time aero engine signal analysis

For many years the analysis of dynamic signals obtained from aero engine mounted transducers has been performed either by using real time spectrum analysers within the test facility, or by making high quality tape recordings during engine tests and analysing the data via an off-line main frame computer. Although real time analysers produce the information where it is most needed, they provide no history of events and the results are operator dependent. Analysis from tape recordings enables information extraction algorithms to be performed and tables and graphs of notable events to be printed. However much of this information could be more effectively utilised if produced within the test facility and in real time. This thesis describes the design and development of a real time data acquisition, signal processing and information extraction system ideally suited for engine health and performance monitoring within test facilities. The thesis begins with a detailed description of the problems encountered in dynamic signal analysis in the field of aero engine performance testing, and with an overview of digital signal processing and the latest technology signal processing micro processors that have made this project possible. It then describes the problems encountered and the subsequent solutions found during the design and development of the hardware and software needed for the·high bandwidth data acquisition and fast signal processing algorithms. The fast Fourier transform has been used for very many years in the field of spectrum analysis, ... however this technique has limitations which are overcome by some of the more modern spectrum estimation techniques. This thesis makes an assessment of some of these techniques, noting particularly their performance on aero engine type signals. The results of these tests are recorded and the possible use of the techniques in aero engine analysis is discussed.