Wyniki wyszukiwania - BazTech

1

About diagnosing of impact of railway vehicles on the track

Cherniak G., Domin R., Nozhenko O., Mostovych A.

Prace Naukowe Politechniki Warszawskiej. Transport

|

2016

|

z. 112

59--69

EN

The report presents the results of the research aimed at justification of expediency of creating a system of experimental evaluation of indicators of impact of the rolling stock on the rails. The proposed system is based on the measurements of the vertical and horizontal accelerations of the rails under a movement of the rolling stock on the raił track section. This system for monitoring rail accelerations that are emerging from the impact of the rolling stock in the track structure is expected to be used to identify vehicles with high impact on the rails due to their design and technical condition of the running gears, including the defects of the wheels. Some results of the pilot testing of the proposed system for the monitoring of the accelerations, which were registered by passing the train consisting of five units, are presented. And the computational algorithm of the specialized pre-processing of the multidimensional signal that is recorded by this system is described. The offered algorithm in automatic mode allows dividing the signal on the separate segments that are associated with the particular wheelsof a running train for subsequent analysis of each segment. This algorithm does not reąuire input data about the train speed on the track section that is eąuipped with the monitoring system, about the number of wheelsets of the running train, and about the distances between them. These data are calculated as a result of the algorithm execution, sińce they are absent under the operating conditions of the proposed system.

2

Signal segmentation for operational regimes detection of heavy duty mining mobile machines - a statistical approach

Wyłomańska A., Zimroz R.

Diagnostyka

|

2014

|

Vol. 15, No. 2

33--42

EN

In the paper an automatic procedure for diagnostic signals segmentation is proposed. The purpose of the procedures is to detect/identify part of the signal, that is related to stationary operation regime of machine. Detection and parameterization of such events might help to improve efficiency of machine usage, for example to minimize number of segments and their duration for machine operation under idle mode or overload mode. It is proposed to use a procedures that are based on statistical analysis to estimate the critical point of the division in a structural change in a time series. Two measures have been proposed for critical points detection: the first one is based on testing of empirical moment of order two for time subseries with length k, second one is related to analysis of second order moment moving along the signal. These techniques have been validated using simulations and then applied to real data acquired from on board monitoring system developed for mobile mining machines (loaders are considered here). Results of application are discussed in the paper.

PL

W artykule przedstawiono procedury segmentacji sygnałów drganiowych. Cel opracowanych procedur jest związany z detekcją/identyfikacją fragmentów sygnału (segmentów) które związane są z różnymi trybami pracy maszyny jak na przykład bieg jałowy czy przeciążenie. Detekcja i parametryzacja tych segmentów pozwoli poprawić efektywność pracy tj. minimalizować czas pracy na biegu jałowym czy zapobiegać przyspieszonym procesom degradacji wynikającym z nieprawidłowego użytkowania maszyny. Zaproponowano dwie procedury bazujące na statystykach, które estymują punkt dzielący szereg czasowy na podprocesy o znaczących różnicach statystycznych. Pierwsza ze statystyk bazuje na badaniu zachowania empirycznego momentu rzędu 2 danej próby w oknie o długości k, druga jest analizą kroczącego drugiego momentu z próby. W pracy przedstawiono wyniki walidacji metody na sygnałach symulacyjnych oraz wyniki zastosowania procedur do sygnałów zarejestrowanych na górniczych maszynach samojezdnych pracujących w kopalni podziemnej.

3

Representation of ECG signals using the segmentation technique

Gacek A., Jeżewski J.

Biocybernetics and Biomedical Engineering

|

2003

|

Vol. 23, no. 4

33-43

EN

In this study, we are concerned with the segmentation of ECG signals and the use of the resulting constructs (segments) in the development of a vocabulary of generic signal descriptors. The resulting space of the segmentation parameters is discussed in detail. It is shown how their representatives (prototypes) are constructed via fuzzy clustering. Numerical examples using the MIT-BIH database ECG signals are provided.