An effective neuro-fuzzy paradigm for machinery condition health monitoring

Systèmes d’inférence floue auto-évolutifs. Apprentissage incrémental pour la reconnaissance de gestes manuscrits

“…La plupart des algorithmes d'apprentissage incrémental de structure et de paramètres sont basés sur le principe de l'algorithme de clustering ART (Carpenter et al, 1988), tel que (Carpenter et al, 1992), (Sadri et al, 2006), (Gary G. Yen, 2001) et . Le problème principal de ces systèmes est qu'ils sont sensibles à la sélection du paramètre de vigilance (qui correspond au seuil de distance qui contrôle la création d'un nouveau cluster), aux niveaux de bruit dans les données d'apprentissage et à l'ordre dans lequel les données d'apprentissage sont présentées.…”

An Approach to Online Identification of Takagi-Sugeno Fuzzy Models

“…If in addition to the previous condition (the potential of the new data point is higher that the potential of all the previously existing centers) the new data point is close to an old center (22) then the new data point replaces this center . This mechanism for rule-base adaptation called modification ensures a replacement of a rule with another one built around the projection of the new data point on the axis .…”

“…This could be a new operating mode of the plant or reaction to a new disturbance. In reality, many regimes and process states cannot be practically included into the training data set (such as faulty process behavior), but states close to them could well appear during the process run [22].…”

Structural Health Monitoring Using Index Based Reasoning For Unmanned Aerial Vehicles

“…Most of these non-destructive evaluation methodologies are impractical to be implemented onboard, and cannot operate automatically for in situ monitoring (Yen and Meesad 2001). Moreover, many NDE methods only can be applied to identify surface damages; even though some methods can reveal inner problems, they lack accuracy in providing quantitative information (Bayraktara et al 2008).…”

“…At the same time, a downsized workforce, a declining budget and a shorter interim operating period have further complicated the task of maintenance and sustainment (Yen and Meesad 2001). To solve this problem, SHM has been developed during the past several decades.…”