Quantifying uncertainties in measurements of railway vibration

© Springer International Publishing AG, part of Springer Nature 2018. Predictions of railway-induced vibration levels should be accompanied by the caveat that uncertainties exist regardless of the prediction technique used—whether it be empirical methods based on in situ measurements, or numerical models with measured or estimated parameter inputs. This paper investigates sources of uncertainty present in field measurements. Using data from an extensive in situ measurement campaign, variability due to changes in receiver position, the averaging of multiple hammer impacts and time lapse between measurements are quantified. Results are used to conservatively estimate the following uncertainty levels: 1–2 dB for installation of measuring equipment; 5 dB for temporal changes of 14 months; 5 dB at below 10 Hz when using an average of 20 hammer impacts to determine transfer functions; and at least 2–5 dB variation for 10 m longitudinal offset in measurement location.

Publication year:2018