"Mammendorf" . "RIV/68407700:21220/06:02120817" . . "20"^^ . . "P(2B06023)" . "Real-time Fault Diagnosis for Non-linear Stochastic Systems"@en . . "Advanced Technologies, Research-Development-Applications" . "RIV/68407700:21220/06:02120817!RIV07-MSM-21220___" . . . "Real-time Fault Diagnosis for Non-linear Stochastic Systems"@en . . "2"^^ . "1"^^ . "Pro Literatur Verlag" . . "393;412" . "fault detection; fault diagnosis; probabilistic model"@en . . . . "Diagnostika poruch v re\u00E1ln\u00E9m \u010Dase pro neline\u00E1rn\u00ED syst\u00E9my"@cs . . . "Real-time Fault Diagnosis for Non-linear Stochastic Systems" . "Diagnostika poruch v re\u00E1ln\u00E9m \u010Dase pro neline\u00E1rn\u00ED syst\u00E9my"@cs . "Garajayewa, G." . "The problem of real-time fault diagnosis is addressed by means of probability distributions. The last describes the relationships of the measured data and the fault variable, which acts as a pointer to one of all considered modes of the system. The main advantage of the presented FDI approach is the ability to provide real-time supervised training, so that the probabilistic model is able to learn new changes in the process behaviour in real time. Application to a laboratory process is presented."@en . . "496561" . . "Real-time Fault Diagnosis for Non-linear Stochastic Systems" . "Diagnostika poruch je \u0159e\u0161ena pro neline\u00E1rn\u00ED stochastick\u00FD syst\u00E9m vyu\u017Eit\u00EDm pravd\u011Bpodobostn\u00EDho modelu, kter\u00FD popisuje vztah mezi m\u011B\u0159en\u00FDmi daty a n\u00E1hodnou prom\u011Bnnou slou\u017E\u00EDc\u00ED k odhadu poruch. Uveden\u00E1 metodika umo\u017E\u0148uje v re\u00E1ln\u00E9m \u010Dase dopl\u0148ovat nov\u00E9 informace. Uvedena aplikace t\u00E9to techniky na laboratorn\u00ED \u00FAloze."@cs . . "Hofreiter, Milan" . "21220" . "The problem of real-time fault diagnosis is addressed by means of probability distributions. The last describes the relationships of the measured data and the fault variable, which acts as a pointer to one of all considered modes of the system. The main advantage of the presented FDI approach is the ability to provide real-time supervised training, so that the probabilistic model is able to learn new changes in the process behaviour in real time. Application to a laboratory process is presented." . "3-86611-197-5" . "[5B17F19BF691]" .