. . "The main subject of the project is an identification of dynamically loaded structures using methods of artificial intelligence in order to analyze damage of structure and to determine a variability of material properties (stiffness) along the structure. As an identification technique the method based on coupling of artificial neural networks and stochastic analysis of structure for preparation of appropriate training set is used. Sensitivity analysis of the input parameters is the integral part of the method. The proposed methodology will be extended towards the dynamic response of structures when results of modal or time domain analysis are the input parameters (frequencies, mode shapes, assurance criteria MAC, FRAC etc.). Limit cycles and chaotic behavior will be possibly considered too. The verification of the methodology and results will be carried out using experimental data from laboratory tests of reinforced concrete beams and from results of health monitoring performed on several bridges."@en . . "Inverzn\u00ED anal\u00FDza dynamicky nam\u00E1han\u00FDch konstrukc\u00ED za pou\u017Eit\u00ED metod um\u011Bl\u00E9 inteligence" . . . "13"^^ . . . "13"^^ . . " artificial neural networks" . "GP103/07/P380" . . "2015-01-22+01:00"^^ . "Inverse analysis; identification; artificial neural networks; Latin Hypercube Sampling; dynamic resp"@en . " Latin Hypercube Sampling" . . . "Inverse analysis" . "http://www.isvav.cz/projectDetail.do?rowId=GP103/07/P380"^^ . "V\u00FDsledky projektu jsou pln\u011B v souladu s p\u016Fvodn\u00EDm pl\u00E1nem. Hlavn\u00EDm c\u00EDlem bylo vypracov\u00E1n\u00ED a implementace metodiky identifikace dynamicky nam\u00E1han\u00FDch soustav na z\u00E1klad\u011B kombinace um\u011Bl\u00FDch neuronov\u00FDch s\u00EDt\u00ED a stochastick\u00E9 anal\u00FDzy. Tohoto c\u00EDle bylo prokazateln\u011B d"@cs . . . " identification" . "2009-04-22+02:00"^^ . "Hlavn\u00ED n\u00E1pln\u00ED projektu je vyu\u017Eit\u00ED metod um\u011Bl\u00E9 inteligence p\u0159i identifikaci dynamicky nam\u00E1han\u00FDch konstrukc\u00ED s c\u00EDlem lokalizovat poru\u0161en\u00ED konstrukce a stanovit prom\u011Bnlivost materi\u00E1lov\u00FDch vlastnost\u00ED (tuhost) po konstrukci. K identifikaci bude pou\u017Eita metodakombinuj\u00EDc\u00ED um\u011Bl\u00E9 neuronov\u00E9 s\u00EDt\u011B a stochastickou anal\u00FDzu konstrukce slou\u017E\u00EDc\u00ED pro vytvo\u0159en\u00ED vhodn\u00E9 u\u010D\u00EDc\u00ED mno\u017Einy. Ned\u00EDlnou sou\u010D\u00E1st\u00ED metody je i citlivostn\u00ED anal\u00FDza vstupn\u00EDch parametr\u016F. Vytvo\u0159en\u00E1 metodika bude rozv\u00EDjena sm\u011Brem k\u00A0dynamick\u00E9 odezv\u011B konstrukc\u00ED, kdy vstupn\u00EDmi parametry identifikace budou v\u00FDsledky mod\u00E1ln\u00ED \u010Di \u010Dasov\u00E9 anal\u00FDzy (frekvence, vlastn\u00ED tvary, srovn\u00E1vac\u00ED krit\u00E9ria MAC, FRAC apod.). Uva\u017Euje se i se zahrnut\u00EDm limitn\u00EDch cykl\u016F a chaotick\u00E9ho chov\u00E1n\u00ED. Verifikace metodiky a v\u00FDsledk\u016F bude provedena na experiment\u00E1ln\u011B zji\u0161t\u011Bn\u00FDch datech z\u00A0laboratorn\u00EDch test\u016F \u017Eelezobetonov\u00FDch nosn\u00EDk\u016F a z v\u00FDsledk\u016F dynamick\u00E9ho m\u011B\u0159en\u00ED na n\u011Bkolika mostn\u00EDch konstrukc\u00EDch. Tato identifika\u010Dn\u00ED metoda p\u0159isp\u011Bje k\u00A0n\u00E1sledn\u00E9mu stanoven\u00ED spolehlivosti konstrukce a jej\u00ED" . . "Inverse analysis of dynamically loaded structures using artificial intelligence methods"@en . . "0"^^ . "2009-12-31+01:00"^^ . . . . "."@en . "0"^^ . "1"^^ . . "2007-01-01+01:00"^^ .