Nehodí sa? Žiadny problém! U nás môžete do 30 dní vrátiť
S darčekovým poukazom nešliapnete vedľa. Obdarovaný si za darčekový poukaz môže vybrať čokoľvek z našej ponuky.
30 dní na vrátenie tovaru
This work introduces an on-line§particle-filtering-based framework for fault§diagnosis and failure prognosis in nonlinear,§non-Gaussian systems. This framework considers hybrid§state-space models of the system under analysis (with§unknown time-varying parameters) and§particle-filtering (PF) algorithms to estimate the§current probability density function (pdf) of the§state, enabling on-line computation of the§conditional fault probability (fault diagnosis§module) and the pdf of the remaining useful life§(RUL) in the case of a declared fault condition§(failure prognosis module). The proposed method§allows to use the state pdf estimate of the diagnosis§module as initial condition for the prognosis module,§improving the accuracy of RUL estimates at the early§stages of the fault condition. This framework§provides information about precision and accuracy of§long-term predictions, RUL expectations, and 95%§confidence intervals for the condition under study.§Ground truth data from a seeded fault test are used§to validate the proposed approach.