Papers by Keyword: Novelty Detection

Paper TitlePage

Authors: Yong Gui Dong, Ensheng Dong, Huibo Jia, Wener Lv
Abstract: In case of mechanical system health monitoring, a need to develop normal-knowledge based novelty detection techniques is increasing. The negative selection algorithm, which is inspired from the operation mechanism of human immune system, is one of such approaches. Our approach is to apply the idea for the anomaly detection in the vibration time series of the rotor system. A real-valued negative selection algorithm based on Euclidean distance, as well as cosine similarity, has been implemented. By means of adding the corresponding coverage radius to each antibody elements, the detection efficiency of each antibody element is increased. The detection efficiency is evaluated with simulated data as well as vibration signal sampled from one rotor system. The results indicate that the algorithm can efficiently detect the anomaly in time series data. Moreover, the number of detectors in antibody set is less enough for potential application in online signal monitoring.
Authors: David A. Clifton, Peter R. Bannister, Lionel Tarassenko
Abstract: A novelty detection approach to condition monitoring of aerospace gas-turbine engines is presented, providing a consistent framework for on- and off-line analysis, each with differing typical implementation constraints. On-line techniques are introduced for observing abnormality in engine behaviour during aircraft flights, and are shown to provide early warning of engine events in real-time. Off-line techniques within the same analysis framework are shown to allow the tracking of single engines and fleets of engines from ground-based monitoring stations on a flight-by-flight basis. Results are validated by comparison to conventional techniques, in application to aerospace engines and other industrial high-integrity systems.
Authors: Nikolaos Dervilis, A.C.W. Creech, A.E. Maguire, Ifigeneia Antoniadou, R.J. Barthorpe, Keith Worden
Abstract: Reliability of offshore wind farms is one of the key areas for the successful implementation of these renewable power plants in the energy arena. Failure of the wind turbine (WT) in general could cause massive financial losses but especially for structures that are operating in offshore sites. Structural Health Monitoring (SHM) of WTs is essential in order to ensure not only structural safety but also avoidance of overdesign of components that could lead to economic and structural inefficiency. A preliminary analysis of a machine learning approach in the context of WT SHM is presented here; it is based on results from a Computational Fluid Dynamics (CFD) model of Lillgrund Wind farm. The analysis is based on neural network regression and is used to predict the measurement of each WT from the measurements of other WTs in the farm. Regression model error is used as an index of abnormal response.
Authors: Keith Worden, Graeme Manson, Cecilia Surace
Abstract: The object of this paper is to illustrate the use of novelty detection techniques in Structural Health Monitoring (SHM) by the consideration of a number of case studies of varying complexity, from a simple lumped-mass system to an FE model of an offshore structure to an experimental study of an aircraft wing.
Authors: Nikolaos Dervilis, Robert J. Barthorpe, Keith Worden
Abstract: The central target of this work is to provide an alternative to machine learning approaches to structural health monitoring with one of robust multivariate statistic novelty detection. Damage detection and identification is a procedure that is hierarchical in nature. At its most sophisticated, diagnosis of the damage could include localisation, classification and severity assessment and even go so far as to estimate the time-to-failure of the structure. In this paper, robust multivariate statistics were investigated focused mainly on a high level estimation of the outliers which determines only the presence or absence of novelty - something that is of fundamental interest. These methods allow a diagnosis of deviation from normality and the option of identifying the presence of masking effects caused by multiple outliers. This paper is trying to introduce a new scheme for damage detection by adopting simple measurements and exploiting robust multivariate statistics.
Authors: Miao Li, Wei Xin Ren
Abstract: The vibration features are affected by damage in structure and environmental conditions while the bridges are in the operation. Environment effects should not be ignored in making correct diagnoses of structures. Negative selection algorithm inspired by immune system has the capability for self-nonself discrimination. Temperature effect on natural frequency is analyzed in the paper, and the algorithm based on Euclidean distance is applied to natural frequencies of structures under temperature variations. The results indicate that negative selection algorithm using natural frequency passes the false-positive tests, and effectively detect the anomalous condition of structure under varying temperature.
Authors: Amin Al-Habaibeh, A. Al-Azmi, N. Radwan, Yang Song
Abstract: Condition monitoring systems of manufacturing processes have been recognised in recent years as one of the key technologies that provide the competitive advantage in many manufacturing environments. It is capable of providing an essential means to reduce cost, increase productivity, improve quality and prevent damage to the machine or work-piece. Turning operations are considered one of the most common manufacturing processes in industry. Despite recent development and intensive engineering research, the development of tool wear monitoring systems in turning is still on-going challenge. In this paper, force and acoustic emission signals are used for monitoring tool wear in a feature fusion model. The results prove that the developed system can be used to enhance the design of condition monitoring systems for turning operations to predict tool wear or damage.
Showing 1 to 10 of 10 Paper Titles