Data quality assessment of automatic wheel profile measurement systems

Projekt:

JVTC

Sammanfattning:
The aim of this paper is to present a method for the quality assessment of data from a condition monitoring system for rolling stock wheels to ascertain if the data have the right quality to be used for further analyses. This quality assessment will also show if there are variations between different measurement units for the same system, and if there are relations between different wheel parameter measurements, speed and time. The assessment of data is accomplished using the quality dimension freedom of error. There are two different data sources, namely an automatic wheel profile measurement system and a manual wheel profile measurement device. The manual measurements of wheel profiles are used for verifying the accuracy of the automatic wheel profile measurements, which constitute the larger data set. The proposed method for evaluating the data quality is demonstrated using the data from a specific condition monitoring system. The results show some inconsistencies indicating that this system lacks quality in the dimension of freedom of error and that there is need for internal calibration or self-adjustment of the studied system for quality reasons.

Författare: Matthias Asplund ; Stephen Mayowa Famurewa ; Matti Rantatalo
Utgivare: Luleå tekniska universitet
Utgivningsdatum: 2015-06-03
Diarienummer: TRV 2011/58769
Antal sidor: 10
Språk: Engelska
Kontaktperson: Per Olof Larsson Kråik, UHjbs