|
| |
|
|
Optimization Algorithm with Probability Direction Vector
Pohl, Jan ; Pokorný, Miroslav (referee) ; Matoušek, Radomil (referee) ; Jirsík, Václav (advisor)
This disertation presents optimization algorithm with probability direction vector. This algorithm, in its basic form, belongs to category of stochastic optimization algorithms. It uses statistically effected perturbation of individual through state space. This work also represents modification of basic idea to the form of swarm optimization algoritm. This approach contains form of stochastic cooperation. This is one of the new ideas of this algorithm. Population of individuals cooperates only through modification of probability direction vector and not directly. Statistical tests are used to compare resultes of designed algorithms with commonly used algorithms Simulated Annealing and SOMA. This part of disertation also presents experimental data from other optimization problems. Disertation ends with chapter which seeks optimal set of control variables for each designed algorithm.
|
|
|
Cross_section and reinforcement optimization of roof truss
Odvárka, Miroslav ; Šimůnek, Petr (referee) ; Laníková, Ivana (advisor)
The thesis is focused on optimization of cross-section shape of reinforced concrete truss. An optimization algorithm was created in Microsoft Office Excel which optimizes the shape and reinforcement of the reinforced concrete truss. The truss is as cheap as possible while complying with the ultimate limit state and serviceability limit state - deflection. The algorithm uses the Solver tool which is a supplement program Microsoft Office Excel. Two truss variants were created. One has a cross-sectional shape of a truss designed according to empirical relationships and the second comprises an optimized cross-sectional shape using the algorithm formed. The internal truss forces were detected using Scia Engineer. Both options were statically assessed. When comparing the variants, that a considerable part of the material cost can be saved when optimizing the cross-sectional shape.
|
|
|
QRS detection using zero crossing counting
Hanus, Rostislav ; Smital, Lukáš (referee) ; Vítek, Martin (advisor)
This master’s thesis deals with the detection of QRS complex detection using zero crossing counts. QRS detection is an important part of the analysis of ECG signal. From the point of determining the R wave detection is based on the other waves and intervals necessary for the diagnosis of heart. This method is very effective even for very noisy signals. Implementation of the method in Matlab, and the success of detection is tested on the CSE and MIT-BIH database. The optimization algorithm is an optional value for the detector.
|
|
|
Advanced Evolutionary Optimisation of TSP-Based Problems
Hladyuk, Vadym ; Vašíček, Zdeněk (referee) ; Bidlo, Michal (advisor)
This paper solves the traveling salesman problem using an evolutionary algorithm, specifically a genetic algorithm. It is a hybrid of the genetic algorithm, using a local search algorithm and other enhancements that further improve the results obtained. The traveling salesman problem will be solved from 20 cities to 25,000 cities. In the experiments chapter, I have determined the best settings for all the parameters in the program and properly tested their appropriateness. In the next part of the experiments chapter, I found out the performance of the full version of the genetic algorithm and its variants. In the last section, I compared the evolution of fitness values of different variants of genetic algorithms and different variants of crossover operators, I also compared the time consumption. I suggested further possible improvements either to the local search algorithm or to another approach to solve the TSP.
|
|
| |
|
|
Web application for creating a cyber security profile
Stejskal, Michal ; Sikora, Marek (referee) ; Dzurenda, Petr (advisor)
This Bachelor thesis describes design and implementation of optimalization program designed to find best combination of courses. The goal in this thesis was to get acquainted with European Cybersecurity Skills Framework (ECSF) in the area of cyber security, optimalization algorithms and with project Cybersecurity Skills Alliance – A New Vision for Europe (REWIRE). This thesis also includes web design and working with the database. Program is developed in the Python, PHP and Javascript programming languages. Between critical requirements of the program belongs optimality of found solution and time in which is the solution found. Program is connected with web interface, in which it is possible to search for optimal combinations based on input constraints.
|
|
|
Cross_section and reinforcement optimization of roof truss
Odvárka, Miroslav ; Šimůnek, Petr (referee) ; Laníková, Ivana (advisor)
The thesis is focused on optimization of cross-section shape of reinforced concrete truss. An optimization algorithm was created in Microsoft Office Excel which optimizes the shape and reinforcement of the reinforced concrete truss. The truss is as cheap as possible while complying with the ultimate limit state and serviceability limit state - deflection. The algorithm uses the Solver tool which is a supplement program Microsoft Office Excel. Two truss variants were created. One has a cross-sectional shape of a truss designed according to empirical relationships and the second comprises an optimized cross-sectional shape using the algorithm formed. The internal truss forces were detected using Scia Engineer. Both options were statically assessed. When comparing the variants, that a considerable part of the material cost can be saved when optimizing the cross-sectional shape.
|
|
|
Two Algorithms for Risk-averse Reformulation of Multi-stage Stochastic Programming Problems
Šmíd, Martin ; Kozmík, Václav
Many real-life applications lead to risk-averse multi-stage stochastic problems, therefore effective solution of these problems is of great importance. Many tools can be used to their solution (GAMS, Coin-OR, APML or, for smaller problems, Excel), it is, however, mostly up to researcher to reformulate the problem into its deterministic equivalent. Moreover, such solutions are usually one-time, not easy to modify for different applications. We overcome these problems by providing a front-end software package, written in C++, which enables to enter problem definitions in a way close to their mathematical definition. Creating of a deterministic equivalent (and its solution) is up to the computer. In particular, our code is able to solve linear multi-stage with Multi-period Mean-CVaR or Nested Mean-CVaR criteria. In the present paper, we describe the algorithms, transforming these problems into their deterministic equivalents.
|
|
| |
guest ::
