|
|
Evolutionary Optimization of Control Algorithms
Weisser, Roman ; Šeda, Miloš (referee) ; Zelinka,, Ivan (referee) ; Ošmera, Pavel (advisor)
The dissertation thesis deals with Evolution optimization of control algorithms. The first part of the thesis describes the principles and partial methods of evolution optimization methods especially those used in two-level transplant evolution method. Later the grammatical evolution method is described, which modified algorithm became impulse for creation of transplant evolution method. The transplant evolution method and its two-level modification are new evolutionary algorithms proposed in this work, which were used for optimization of structure and parameters of general controllers control algorithms. The transplant evolution algorithm and its extended two-level modification are described in detail in next chapters. The proper settings of evolutionary algorithms are important for minimization the time of optimization and for finds results approaching the global optimum. For proper setting the parameters of differential evolution was created meta-evolution algorithm that is described in chapter named meta-evolution. The basic concepts of control, chosen methods of system identification and controller parameters settings are described in next part. This part describes algorithms of digital controllers and some specific methods uses in digital control. The demonstrations of control algorithm optimizations of various types of controllers are showed in experimental part. The optimized algorithms of general controllers are compared with various types of PSD controllers which were set by various algebraic methods or differential evolution for various models of systems. In the conclusion of this work is stated a recommendation for further development of evolutionary optimization of controllers are focusing on parallel and distributed computing.
|
|
|
Application of Genetic Algorithms and Data Mining in Noise-based Testing of Concurrent Software
Šimková, Hana ; Kofroň, Jan (referee) ; Lourenco, Joao (referee) ; Vojnar, Tomáš (advisor)
Tato práce navrhuje zlepšení výkonu testování programů použitím technik dolování z dat a genetických algoritmů při testování paralelních programů. Paralelní programování se v posledních letech stává velmi populárním i přesto, že toto programování je mnohem náročnějsí než jednodušší sekvenční a proto jeho zvýšené používání vede k podstatně vyššímu počtu chyb. Tyto chyby se vyskytují v důsledku chyb v synchronizaci jednotlivých procesů programu. Nalezení takových chyb tradičním způsobem je složité a navíc opakované spouštění těchto testů ve stejném prostředí typicky vede pouze k prohledávání stejných prokládání. V práci se využívá metody vstřikování šumu, která vystresuje program tak, že se mohou objevit některá nová chování. Pro účinnost této metody je nutné zvolit vhodné heuristiky a též i hodnoty jejich parametrů, což není snadné. V práci se využívá metod dolování z dat, genetických algoritmů a jejich kombinace pro nalezení těchto heuristik a hodnot parametrů. V práci je vedle výsledků výzkumu uveden stručný přehled dalších Technik testování paralelních programů.
|
|
|
The Shortest Graph's Pahts Finding
Jágr, Petr ; Ohlídal, Miloš (referee) ; Jaroš, Jiří (advisor)
The aim of this thesis is finding, comparing and implementation of algorithms for finding the shortest paths between each of pairs of nodes in a graph. For this task I use modifications of existing algorithms to achive the lowest time consumption of the computation. Modifications are established on Dijkstra's and Floyd-Warshall's algorithm. We also familiarize with Bellman-Ford algorithm.
|
|
|
The Investment Models in an Environment of Financial Markets
Repka, Martin ; MSc, Martin Volko (referee) ; Budík, Jan (advisor)
This thesis focuses on automated trading systems for financial markets trading. It describes theoretical background of financial markets, different technical analysis approaches and theoretical knowledge about automated trading systems. The output of the present paper is a diversified portfolio comprising four different investment models aimed to trading futures contracts of cocoa and gold. The portfolio tested on market data from the first quarter 2013 achieved 46.74% increase on the initial equity. The systems have been designed in Adaptrade Builder software using genetic algorithms and subsequently tested in the MetaTrader trading platform. They have been finally optimized using sensitivity analysis.
|
|
| |
|
|
Evolutionary Algorithms in Convolutional Neural Network Design
Badáň, Filip ; Vašíček, Zdeněk (referee) ; Sekanina, Lukáš (advisor)
This work focuses on automatization of neural network design via the so-called neuroevolution, which employs evolutionary algorithms to construct artificial neural networks or optimise their parameters. The goal of the project is to design and implement an evolutionary algorithm which can be used in the process of designing and optimizing topologies of convolutional neural networks. The effectiveness of the proposed framework was experimentally evaluated on tasks of image classification on datasets MNIST and CIFAR10 and compared with relevant solutions. The results showed that neuroevolution has a potential to successfully find accurate and effective convolutional neural network architectures.
|
|
|
Process Control with Dynamic Resource Scheduling
Šinkora, Jan ; Kočí, Radek (referee) ; Janoušek, Vladimír (advisor)
This project pursues issues on the border of information technologies and process optimization. Previously published concepts of~modeling projects and shared resources with object-oriented Petri nets are presented and further expanded. The possibilites of~the use of~genetic algorithms for dynamic realtime optimization of the resource schedules are explored. The resource constrained project sheduling problem is presented and it is shown, how instances of the problem can be implemented. A more complex model that is inspired by real production systems is then created. Next, a control agent, which monitors a running production system and allows for it's dynamic optimization is designed. The whole system is implemented in the Squeak Smalltalk environment with the use of the tool PNtalk, which is an experimental implementation of the object oriented Petri nets paradigm.
|
|
|
Design and Optimization of Automated Trading System
Ondo, Ondrej ; Gancarčík, Lukáš (referee) ; Budík, Jan (advisor)
This thesis focuses on automated trading systems for foreign exchange markets. It describes theoretical background of financial markets, technical analysis approaches and theoretical knowledge about automated trading systems. The output of the thesis is set of two automated trading systems built for trading the most liquid currency pairs. The process of developing automated trading system as well as its practical start up in Spartacus Company Ltd. is documented in the form of project documentation. The project documentation captures choosing necessary hardware components, their installation and oricess of ensuring smooth operation, as well as the selection and installation of the necessary software resources. In the Adaptrade Builder enviroment there has been shown the process of developing strategies and consequently theirs characteristics, performance, as well as a graph showing the evolution of the account at the time. Selected portfolio strategy has been tested in the MetaTrader platform and in the end of the thesis is offered assessing achievements and draw an overall conclusion.
|
|
|
Simulation of Entities Collective Behavior in Virtual World
Vymazal, Tomáš ; Žák, Pavel (referee) ; Láník, Aleš (advisor)
Theme of this work is to evaluate and compare aviable paradigms for entity control in virtual worlds, and to implement one of these paradigms in application. Dynamic finite state machine upgraded using genetic algorithms has been chosen. This paradigm should make agent's behavior better and adapt agent to required state: i.e. make agent harvest resources in virtual world. Output of this work is application for running evolution and application for 3D view of agent's behavior.
|
|
|
The Use of Artificial Intelligence for Optimization of Production
Svoboda, Radovan ; Martinec, Petr (referee) ; Dostál, Petr (advisor)
This paper deals with the problem of optimization of a production plan by using genetic algorithms. It contains a brief overview of the principles behind genetic algorithms in scope of evolutionary algorithms and artificial intelligence in general. It also takes a closer look on the challenge of production planning and control and all activities connected to it. This is followed by description of the modification of genetic algorithms that needed to be done in order to implement it into a computer program, which is used to create and optimize the production plan, and is a result to the issue that this paper deals with. Incorporated is detailed escription of principles and functions of the program, that it offers to its users.
|
guest ::
