Implementing a novel fault prognosis technique based on nonlinear fault observer and online parameters estimation

<span lang="EN-US">This research presents a methodology for predicting errors of parameters, as the algorithm tries to monitor parameters in order maintain or replace them when needed avoid excessive expenses. The presented implementation mechanism is based on monitoring according specific number batches and each batch consists iterations, which turn are samples. proposed involves designing new nonlinear observer writing secondary parameter estimation online recursive least squares associated with states. In addition, an attempt find relationship between error states state by creating function determine weight four components; changes, output residuals, diagnosed observer. also includes introducing probability form weights using kernel density average maximum batch. Finally, relying results, it possible take appropriate decision change shown non-linear direct current motor model case study.</span>