Research on Vibration Test Conditions in Particle Impact Noise Detection

Pattern recognition approach to identify loose particle material based on modified MFCC and HMMs

“…Towards achieving this objective, sealed electronic devices are manufactured and tested in accordance with the controls and requirements of applicable military standards and specifications. One critical factor that can cause short-circuit, device malfunction, or even catastrophic system failure is loose particles in the form of foreign material, such as the wire pieces, chip dregs, aluminum scraps and tin splatters, being left within sealed electronic devices during pre-seal handing and assembly process [1][2][3]. Therefore, it is critical to investigate loose particle detection technology including loose particle detection and material recognition.…”

Detecting loose particle signals in multichannel recordings with transductive confidence predictor

“…To develop a quantitative and easy-to-use non-destructive method for detecting and classifying loose particles, much effort has been made in different aspects. For example, in order to determine the best vibration conditions on freeing the loose particle, a dynamical mathematical model has been established for particle collision in Zhang et al (2004Zhang et al ( , 2006Zhang et al ( , 2008a. For automatic particle detection and material identification, Scaglione (1994) and Ma et al (2003) used neural networks to improve the identification ratio of loose particles.…”