Cálculo de la vida útil remanente mediante trayectorias móviles entre hiperplanos de máquinas de de soporte vectorial
2013 (Spanish)In: Interciencia, ISSN 0378-1844, Vol. 38, no 8, 556-562 p.Article in journal (Refereed) Published
We propose a new method for predicting remaining useful life ( RUL ) classifiers inspired by support vector machines ( SVM). Historical data of condition of a system during its lifetime are used to create a classification by hyperplanes in SVM. To estimate the RUL of a system , the degradation rate was assessed by calculating the minimum distance defined based on the degradation pathways , ie , the progressive approach hyperplane segregates information from both good and bad for different time horizons.One can estimate the final life of a specific component , or the information in the RUL of a population be calculated by aggregating multiple RUL estimates using a density estimation method . The degradation of a system is affected by many unknown factors , besides complicating degradation behaviors , difficult to collect quality data.Due to lack of knowledge and incomplete measurements , normally lacks important information on the context of the data collected . Therefore historical data of the system with a variety of degradation patterns are grouped , with the search for a global model RUL prediction is extremely difficult. This leads them to seek advanced forecasting techniques beyond traditional models. The proposed model develops an effective RUL prediction method that addresses multiple challenges in forecasting complex systems. The similarities between degradation pathways can be contrasted to enrich existing methodologies prognosis.
Place, publisher, year, edition, pages
2013. Vol. 38, no 8, 556-562 p.
Research subject Operation and Maintenance
IdentifiersURN: urn:nbn:se:ltu:diva-11090Local ID: 9ff6d85b-39b0-4eb5-9f76-fa0efd06351bOAI: oai:DiVA.org:ltu-11090DiVA: diva2:984039
Validerad; 2013; 20131211 (diegal)2016-09-292016-09-292016-10-19Bibliographically approved