Applying Extreme Value Theory for alarm and warning levels setting under variable operating conditions
Abstract »Alarm configuration is one of the main challenges of power generation and associated industries. Another challenge is related to the increasing number of machines being operated under variable conditions. Change in operational condition i.e. speed or torque affects the vibration response. Thus, if the data used to determine the alarm and warning threshold levels characterises only limited range of operational conditions a false alarm may be triggered indicating onset of fault while only the operational regimes have changed. Another possibility is the fault to be masked by change in the operational condition which leads to misdetection. Central to determining the alarm and warning threshold levels is establishing the type of the data distribution. The distributions are usually assumed Gaussian or a number of possible distributions are considered in search of the best fit. Incorrect distribution fit may result in sub-optimal alarm configuration. In the present paper instead of considering the whole data set only maxima will be taken into account as likely to reveal an outlier. The Generalised Extreme Value distribution is the only possible limit distribution for the maxima. In order to take into account the effect of the variable speed, Extreme Value Theory for non-stationary processes will be applied. The suggested approach is validated on data from an experimental gearbox.
Biography: Dr. Daniela Filcheva received a MSc. degree in Acoustics and Audio Engineering in 1996 and PhD degree in Control System Engineering in 2008. In 2012 she started working on a project, which was part of British Government Initiative "Knowledge Transfer Partnership". The objective of the project between University of Bristol and Beran Instruments was to develop new intelligent health monitoring solutions. In 2014 Daniela started developing Automatic Alarm Setup algorithms for power generation industry. Currently she works on data quality assessment.
Affiliation: University of Bristol BS8 1TR Bristol United Kingdom 7780876276 07780876276Lieven, NicholasLieven, Nicholas
Affiliation: University of Bristol Bristol United KingdomMorrish, PeterMorrish, Peter
Affiliation: Beran Instruments Great Torrington United KingdomHutchinson, PaulHutchinson, Paul email@example.com
Affiliation: Beran Instruments EX38 7HP Great Torrington United Kingdom