Machine health prognostics using the Bayesian-inference-based probabilistic indication and high-order particle filtering framework

摘要: Abstract Prognostics is much efficient to achieve zero-downtime performance, maximum productivity and proactive maintenance of machines. intends assess predict the time evolution machine health degradation so that failures can be predicted prevented. A novel prognostics system developed based on data-model-fusion scheme using Bayesian inference-based self-organizing map (SOM) an integration logistic regression (LR) high-order particle filtering (HOPF). In this system, a baseline SOM constructed model data distribution space healthy under assumption predictable fault patterns are not available. probability (BIP) derived from as quantification indication degradation. BIP capable offering failure for monitored machine, which has intuitionist explanation related state. Based those historic BIPs, LR its modeling noise constitute Markov process (HOMP) describe propagation. HOPF used solve HOMP estimation in form density function (PDF). An on-line update adapt changes dynamics quickly. The experimental results bearing test-bed illustrate potential applications proposed effective simple tool prognostics.