Wyniki wyszukiwania - BazTech

1

Feature extraction using indicators of cyclostationarity for a mechanical diagnosis purpose

Raad A., Antoni J., Sidahmed M.

Computer Assisted Mechanics and Engineering Sciences

|

2005

|

Vol. 12, No. 2-3

223-230

EN

This paper focuses on features extraction based on cyclostationarity for diagnosis purpose. The objective is to derive new indicators for the diagnosis of rotating machinery. These indicators are based on cyclic higher order statistics and generalize some existing ones for the second order statistics. A comprehensive methodology is proposed for obtaining a diagnosis objective; a crucial example is presented, relating to vibration signals of a gearbox. Results demonstrate the effectiveness of these features to detect spalling in gearbox.

2

Multimedia and distant learning of condition monitoring

Sidahmed M.

Vibrations in Physical Systems

|

2002

|

Vol. 20

58--67

EN

We discuss in the paper education and training in the domain of signal processing for the acoustic and vibration diagnosis of rotating machines on the basis of CD-ROM that have been developed by various European partners in the frame of the Leonardo da Vinci EC Program. This project was also supported by the Region Picardie (France). The CD-ROM is organised in three modules: methods of signal processing, vibrations of rotating machines and related symptoms, and a module devoted to real signals analysis. It contains interactive consultation modes, with several levels of input, complexity, and purpose. The signal processing makes use of interactive exercise in order to ‘understand’ and demonstrate various aspects of the theory and the possibilities and limitations of the methods. Real signals are extensively use. There is an intensive use of real signals taken from a data base of real industrial installations.

3

Monitoring of rotation machines by vibration analysis using time-frequency methods

Laurent H. H., Sidahmed M.

Prace Naukowe Instytutu Górnictwa Politechniki Wrocławskiej. Konferencje

|

1999

|

Vol. 86, nr 24

172-177

EN

This paper deals with the problem of detecting spectral changes in non stationary noisy signals using non parametric tools. Belonging to this class, the time frequency representations (TFRs) allow a better insight into complicated processes by displaying the content of the signal in a 2-dimensional space. We propose in this article to extract exploitable knowledge from the time-frequency domain by the use of distance measures between TFRs. After presenting the different resulting stationarity indices, we show their performances in a monitoring context : the early detection of gear faults. The analysis of vibration signals which are representative of physical processes in gearboxes finally demonstrates the potential monitoring capability of the proposed methods.

4

Cyclostationary and bilinear approaches for gears vibrating signals application to helicopter gear box diagnosis

Bouillaut L., Sidahmed M.

Prace Naukowe Instytutu Górnictwa Politechniki Wrocławskiej. Konferencje

|

1999

|

Vol. 86, nr 24

93-102

EN

Our study first deals a comparison between cyclostationarity and bilinearity, and then with different applications of these approaches. Using synthetic and indudtrial signals, we underline that cyclostationarity tools make it possible to determine non-linear links between two frequencies (Quadratic Phase Coupling). Then , we present different applications of cyclostationarity and bilinearity to helicopter noise. These methods enable us to obtain much more interesting results than classical methods such as Fourier, Spectrum, time-frequency studies. First, these techniques allow us to detect and to determine precisely a fault on the system. Moreover, the PSD presents a frequency which is not linked to any physical property of the engine. A cyclostationary study makes it possible to explain precisely the apparition of the frequency. Finally, we underlined a unusual modulation phenomenon. Usually, during a meshing phenomenon, the rotating frequency modulaters the meshing frequency; this is characterised by a spectrum with lateral bands around meshing harmonics. Here we encountered a meshing frequency which modulated another meshing frequency.