ABSTRACT. A sys tem for automatic, reliable, semi-instanta n eo us estim ation of avalan che activi ty is presen ted in an a ttempt to ch eck determinis ti c models and im prove the surveill a nce of risk zon es. Th e principle is the seismic d etection of avalanch es. Two exp erim en tal si les eq uipped with sta nd ard seismol ogical eq uipmen t a re d escribed. A vala n ch e se ismi c sig n a ls are reco rd ed , as well as m a n y extra neo us sig nals of natural or hum a n ori gin. After several years of a posteriori id entification of lhe sign als, we a re now able to set up criteria for th e automatic recognition of nonava la n che signals. vV e have therefore d eveloped a n a utom a ti c analysis/d ecision system to discrimin ate betwee n ava la n c h e signals a nd others. Thi s sys tem worked sa ti sfac toril y in pre-operational conditions durin g th e winter of . Th e res ults a r e presented a nd co mp ared with other d a ta related to ava lanch e activity. Although th ere is still room for improve ment, our sys tem seem s to be able to es timate ava lanc h e activily belter than hum a n visual observations.
In this paper, we propose to perform early fault diagnosis using high-resolution spectral analysis of the stator current to detect bearing faults in electrical induction machine. While most research works focus on mechanical vibration analysis, the originality of our work relies on the use of highresolution methods to detect modulations in the stator current. We present the results obtained for real data to detect inner and outer raceway bearing defects made articially as well as bearing defects obtained through on-site ageing. The obtained results show that the proposed method yields better detection than classical spectum analysis.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.