Innovative quality control methods for rotating machines using artificial intelligence methods

Informazioni relative al progetto

INQUAC

ID dell’accordo di sovvenzione: BRPR980631

Progetto chiuso

Data di avvio 1 Settembre 1998

Data di completamento 31 Agosto 2001

Finanziato da

• Specific research and technological development programme in the field of industrial and materials technologies, 1994-1998

Costo totale

Nessun dato

Contributo UE

Nessun dato

Coordinato da

MEDAV Digitale Signalverarbeitung GmbH
Germany

CORDIS fornisce collegamenti ai risultati finali pubblici e alle pubblicazioni dei progetti ORIZZONTE.

I link ai risultati e alle pubblicazioni dei progetti del 7° PQ, così come i link ad alcuni tipi di risultati specifici come dataset e software, sono recuperati dinamicamente da .OpenAIRE .

Risultati finali

ANOVIS prototype system for fault diagnosis in the production lines

The prototype system consists of a PC for operating, classification, archiving, communication to the environmental computer network etc., and a digital signal processor based system for signal acquisition, signal processing and data reduction. The system can be operated in the production line. Noise and vibration signals are used for the assessment of the test specimen. Order analysis is the most important signal analysis method used for feature extraction. The dual classifier is used for a go - no go decision and the fault diagnosis (identification of defective components or special defects). One module of the dual classifier is a tolerance scheme. The interpretation of the feature vector component is supported by the evaluation of the test specimen specific component / cinematic list. A trainable system (kNN based method) improves the diagnosis output. The specific advantage of the dual classifier is that it can apply in the production line "with the first product under test" and the maintenance of the tolerance values can be decreased by the use of the trainable system.