We propose a novel algorithm for extracting samples from a data set supporting the extremal points in the set. Since the density of the data set is not taken into account, the method could enable adaptation to gradually changing data characteristics (e.g. machine wear). Using knowledge about the clustering structure of the data (obtained with multidimensional scaling techniques), the complexity...

Increased railway patronage worldwide is putting pressure on rolling stock and infrastructure to operate at higher capacity and with improved punctuality. Condition monitoring is seen as a contributing factor in enabling this and is highlighted here in the context of rolling stock being procured with high capacity data buses, multiple sensors and centralised control. This therefore leaves scope...

Novelty detection has been developed into a state-of-the-art technique to detect abnormal behavior and trigger alarm for in-field machine maintenance. With built-up models of normality, it has been widely applied to several situations with normal supervising dataset such as shaft rotating speed and component temperature available meanwhile in the absence of fault information. However, the resea...

A National Grid funded research project at Southampton has investigated possible methodologies for data acquisition, transmission and processing that will facilitate on-line continuous monitoring of partial discharges in high voltage polymeric cable systems. A method that only uses passive components at the measuring points has been developed and is outlined in this paper. More recent work, fun...

Aiming at the prevention of global warming, energy conservation and the reduction of power generation cost, demonstration and full-scale production of power generation devices using renewable energy are well underway. In Japan, the feed-in tariff program for renewable energy (photovoltaic, wind, hydro, geothermal and biomass power) started in July 2012 and their deployment is expected to accele...

Condition monitoring techniques are described in this chapter. Two aspects of condition monitoring process are considered: (1) feature extraction; and (2) condition classification. Feature extraction methods described and implemented are fractals, Kurtosis and Melfrequency Cepstral Coefficients. Classification methods described and implemented are support vector machines (SVM), hidden Markov mo...

Structural control and health monitoring as condition are some essential areas that allow for different system parameters to be designed, supervised, controlled, evaluated during the system’s operation in processes, such those used machinery, structures, physical variables mechanical, chemical, electrical, aeronautical, civil, electronics, mechatronics, agricultural engineering applications, am...

Power electronics is becoming pervasive in electricity-based applications, making power device reliability an increasingly important topic. Maintenance operations are necessary to ensure the functioning of electronic apparatuses [1]. Guessing time for maintenance challenging due related tradeoff between costs and improvements. Lifetime prediction techniqu...