Cálculo de la vida útil remanente mediante trayectorias móviles entre hiperplanos de máquinas de soporte vectorial

A novel remaining useful life (RUL) prediction method inspired by support vector machines (SVM) classifiers is proposed. The historical instances of a system with life-time condition data are used to create a classification by SVM hyper planes. For a test instance of the system whose RUL is to be estimated, degradation speed is evaluated by computing the minimal distance defined based on the degradation trajectories, i.e. the system approach to the hyperplane that segregates good and bad conditions data at different time horizons. Therefore, the final RUL of a specific component can be estimated and global RUL information can then be obtained by aggregating the multiple RUL estimates using a density estimation method. The degradation process of a system may be affected by many unknown factors that complicate the degra-dation behavior and also make it difficult to collect quality data. Due to lack of knowledge and incomplete measurements, certain important context information of the collected data might be missing. Therefore, historical data of the system with a large variety of degradation patterns is mixed together. With such data, learning a global model for RUL prediction becomes extremely hard. This has led to look for advanced RUL prediction techniques beyond the traditional prediction models. The proposed model develops an effective RUL prediction method that addresses multiple challenges in complex system prognostics. Similarities between degradation trajectories can be checked in order to enrich existing methodologies in prognostics applications. Existing condition monitoring data for bearings is used to validate the model.