• Refine Query
  • Source
  • Publication year
  • to
  • Language
  • 3
  • Tagged with
  • 3
  • 3
  • 3
  • 2
  • 2
  • 2
  • 2
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • 1
  • About
  • The Global ETD Search service is a free service for researchers to find electronic theses and dissertations. This service is provided by the Networked Digital Library of Theses and Dissertations.
    Our metadata is collected from universities around the world. If you manage a university/consortium/country archive and want to be added, details can be found on the NDLTD website.
1

Genetinių algoritmų taikymas imituojant sistemas aprašytas agregatiniu metodu / Genetic algorithms usage to simulate the systems described in the aggregate method

Dobilas, Mindaugas 13 August 2010 (has links)
Mokslinių tyrimų sritis – genetinių algoritmų ir agregatinio metodo panaudojimas modeliuojant sudėtingas sistemas. Darbo tikslas – genetinių algoritmų taikymas formaliuose sistemų aprašymo metoduose, sistemų imitaciniame modeliavime, sistemų parametrams nustatyti. Mokslinis naujumas. Šiame darbe siūlomas naujas genetinio algoritmo ir agregatinio metodo taikymas sistemos modeliams aprašyti. Sistemos modelio parametrai genetiniame algoritme laikomi kaip individo chromosomos, o sistemos modelis tai naudingumo funkcija genetiniame algoritme. Padarytos prielaidos leidţia nustatyti sistemos parametrų optimalias reikšmes, kad sistema efektyviai dirbtų. Kitas siūlomas taikymo atvejis, kai genetinis algoritmas naudojamas perėjimo operatoriuje, nustatyti sekančios populiacijos struktūrai. Tai leidţia imituoti biologines, agentines, savireguliuojančias sistemas. / Research area - genetic algorithms approach to aggregate and use modeling complex systems. Work objective - the application of genetic algorithms in formal methods, systems imitation modeling, to find optimal settings. This work proposed new usage of aggregate method and genetic algorithm to describe system models. The system parameters of the model are used as the individual's chromosomes in genetic algorithm, and the system model used as a utility function of genetic algorithm. It also proposed other aggregate approach, the genetic algorithm used for the transition operator, followed by the population structure. This allows the simulation of biological, agent, self-regulating systems.
2

Realiojo laiko sistemų veiksenos įvertinimas / Evaluation of Real Time System Behavior

Milinis, Tadas 06 June 2006 (has links)
Complexity and variety of systems that are working in real time mode need to be specified regarding all behavior conditions. The correctness of the specification, determines whether implemented system will supply conditions that were set. To ensure that specification of the described real-time system is correct, we have to do verification and validation of the specification. Traditional verification methods do not assure full real time system inspection. The main drawback, talking about them, is impossibility of system evaluation according time. In this paper a reachable states graph and its generating algorithms are described, while two time moments are compared using linear programming, Simplex method. Therefore, method for checking equivalent state in system behavioral pathway was suggested. Also, theoretical reasoning for creating computer applications, which automates generation of reachable states graph, is given. Reachable states graph fully evaluates real time system behavior.
3

Hibridinių sistemų modeliavimas naudojant agregatinį metodą / The use of piece-linear aggregates for specification and simulation of hybrid systems

Simaitis, Laurynas 16 August 2007 (has links)
Šiame darbe pristatomas PLA formalizmo metodas, skirtas aprašyti ir imituoti hibridines sistemas, kurių tolydiniai komponentai aprašomi pirmos eilės diferencialinėmis lygtimis. Pasinaudojus PLA formalizmu suformuluosime Kvantuotų būsenų sistemos modelį (QSS). QSS modelis yra integravimu pagrįstas metodas, skirtas spręsti pirmos eilės diferencialines lygtis. Šis modelis buvo suformuluotas DEVS formalizmo aplinkoje, kuris yra giminingas PLA formalizmui. Hibridinės sistemos pavyzdžiu pasirinktas farmakokinetinis modelis. Farmakokinetika yra farmakologijos mokslo šaka, tyrinėjant vartojamų vaistų pasisavinimą ir absorbciją žmogaus kūne. Farmakokinetiniame modelyje vaistų pasiskirstymas aprašomas pirmos eilės diferencialines lygtimis. Tam, kad būtų galima aprašyti šį modelį QSS pagalba, buvo reikalinga patobulinti QSS modelį. Rezultate gautas pakeistas QSS modelis. Naudojant pakeistą QSS modelį buvo sudarytas farmakokinetinis modelis ir atliktas vaistų koncentracijos kitimo imitavimas paciento kūne, kai pacientas reguliuoja vaistų suleidimą. / This paper presents the method of using Piece Linear Aggregates formalism (PLA) for specification and simulation of hybrid systems, when continuous components are described in ordinary differential equations (ODEs). PLA is used to create Quantized State System (QSS) model. QSS is integration based method, created for ODE solving. Since QSS method first was introduced for use in Discreet Event Systems (DEVS) formalism, we adapted it for PLA formalism, which is fundamentally consanguineous to DEVS formalism. Pharmacokinetics model is used as example of hybrid system. Pharmacokinetics is a branch of pharmacology which describes administered drug absorption and distribution in human body. In pharmacokinetics drug distribution is described using ODE. In order to use drug injection in pharmacokinetics model, we have to modify standard QSS model and add some specific capabilities. In the result: modified QSS method was obtained and used for pharmacokinetic model specification. Simulation of morphine concentration in plasma was performed using real pharmacokinetics data to test Patient-Controlled Analgesia method effectiveness.

Page generated in 0.0894 seconds