National Repository of Grey Literature 4 records found  Search took 0.01 seconds. 
Acceleration of Neural Networks in FPGA
Krčma, Martin ; Strnadel, Josef (referee) ; Kaštil, Jan (advisor)
This thesis deals with a training of the FPNN structures. It focuses on the ways of direct conversion of the pretrained arti cial neural networks to FPNNs. This is useful when original training data set is not reachable.
Acceleration of Neural Networks in FPGA
Krčma, Martin ; Strnadel, Josef (referee) ; Kaštil, Jan (advisor)
This thesis deals with a training of the FPNN structures. It focuses on the ways of direct conversion of the pretrained arti cial neural networks to FPNNs. This is useful when original training data set is not reachable.
Monitoring of fatigue crack growth in riveted aircraft structures by acoustic emission
Chlada, Milan ; Převorovský, Zdeněk
In principle, the acoustic emission (AE) represents one of NDT and structure health monitoring (SHM) methods for the detection and identification of growing material defects. Different types of signals detected by the AE method are supposed to correspond to different types of defects or operational noises. Monitoring of structures by acoustic emission needs new robust and fast methods for emission source location and classification. Recently proposed AE source location method using so called signal arrival time profiles and artificial neural networks (ANN) was applied for monitoring of growing defects during long term fatigue testing of riveted aircraft wing flange. The potentialities of the method regarding the on-line monitoring of dangerous crack growth in selected critical parts of aircraft structures are discussed.
Detection of AE Sources During Long-term Fatigue Tests of Riveted Aircraft Wing Flange
Chlada, Milan ; Převorovský, Zdeněk
Fracture is the primary threat to the integrity, safety, and performance of nearly all highly stressed mechanical structures, e.g., aircrafts, building units or pressure vessels. Contemporary exacting demands on reliability and safety of material structures are not realizable without effective means of NDT and continuous state monitoring. AE monitoring of structures needs new robust and fast methods for emission source location and classification. Recently proposed AE source location method using so-called signal arrival time profiles and artificial neural networks (ANN) was applied for monitoring of growing defects during long-term fatigue testing of riveted aircraft wing flange. The potentialities of the method regarding the on-line monitoring of dangerous crack growth in selected critical parts of aircraft structures are discussed.

Interested in being notified about new results for this query?
Subscribe to the RSS feed.