Andrew Doyle/LONDON
NEW SIGNAL-processing technology able to monitor the health of mechanical systems and predict possible failures could have wide-ranging aerospace applications, according to Domain Dynamics of the UK.
The company, marketing a system called the Tespar (time-encoded signal processing and recognition), developed by the UK's Cranfield University, says that its key benefits are the ability to pinpoint specific component failures while requiring only a small amount of data processing.
It is hoped that the Tespar could be used to monitor the condition of helicopter gearboxes or turbine blades in aeroengines, simply by analysing the sounds they emit. For example, as a crack in a turbine blade propagates, the sound it produces will change.
Conventional signal processing is carried out using a mathematical method called Fast Fourrier Transformation (FFT), which calculates the individual frequencies which make up a sound wave. This method requires vast amounts of digital data to be processed.
The Tespar works by analysing "snapshots" of a sound wave against time, without the need to calculate frequencies. It compares sections of the wave against pre-determined templates and assigns a code to each, allowing the wave to be defined using a tiny fraction of the data generated by FFT.
These data can be rapidly compared with a stored reference wave, and any differences identified. The concept is particularly suitable for use with networked artificial-intelligence computers, which can "learn" rapidly the significance of deviations and attribute them to the failure of a particular component.
"Tespar is completely different to established technology," says Martin George, Domain Dynamics' marketing manager. "FFT is appropriate where you need to recognise specific frequencies. If recognition of a condition or event is required, Tespar is more efficient and requires less processing power. It is much more of a diagnostic tool," he adds.
Using a PC-based Tespar development kit, an appropriate specification can be defined for potential customers. In addition, noise which reveals nothing about the condition of a device, such as that related purely to the rotational speed of an engine shaft, can be ignored to further improve efficiency.
Other possible applications for the technology include so-called "smart" sensors and digital cockpit-voice recorders, for which the Tespar's data-compression capabilities could be be suitable.
Source: Flight International
