|
Learnable Evolution Model for Optimization (LEM)
Grunt, Pavel ; Vašíček, Zdeněk (referee) ; Schwarz, Josef (advisor)
My thesis is dealing with the Learnable Evolution Model (LEM), a new evolutionary method of optimization, which employs a classification algorithm. The optimization process is guided by a characteristics of differences between groups of high and low performance solutions in the population. In this thesis I introduce new variants of LEM using classification algorithm AdaBoost or SVM. The qualities of proposed LEM variants were validated in a series of experiments in static and dynamic enviroment. The results have shown that the metod has better results with smaller group sizes. When compared to the Estimation of Distribution Algorithm, the LEM variants achieve comparable or better values faster. However, the LEM variant which combined the AdaBoost approach with the SVM approach had the best overall performance.
|
|
Production of wing rib
Ivančo, Radek ; Peterková, Eva (referee) ; Císařová, Michaela (advisor)
The thesis presents a proposal of production of a molded part from the aluminum alloy of group 2024 (dural). Based on the theoretical study and the requirements of the company was designed a cutting tool with a rounded edge for cutting the unfolded shape of the part on the eccentric press LEN 40 C. A new forming tool from textite material for fluid press Quintus QFC was designed. The parts of the cutting tool were designed primarily from normalized (purchased) parts. Manufactured punch and shearing die are made of commonly available tool steels and processed on the attached drawings. The result of the thesis is a recommendation to invest in a specialized workplace for the production of a unfolded shape and reduction of laboriousness adjusting after forming.
|
|
Learnable Evolution Model for Optimization (LEM)
Weiss, Martin ; Vašíček, Zdeněk (referee) ; Schwarz, Josef (advisor)
Numerical optimization of multimodal or otherwise nontrivial functions has stayed around the peak of the interest of many researchers for a long time. One of the promising methods that appeared is the hybrid approach of the Learnable Evolution Model that combines the well-established ways of artificial intelligence and machine learning with recently popular and efective methods of evolutionary programming. In this work, the method itself was reviewed with respect to what has been already implemented and tested and several possible new implementations of the method were proposed and some of them consequently implemented. The resulting program was then tested against a set of chosen nontrivial real-valued functions and its results were compared to those achieved with EDA algorithms.
|
|
Production of wing rib
Ivančo, Radek ; Peterková, Eva (referee) ; Císařová, Michaela (advisor)
The thesis presents a proposal of production of a molded part from the aluminum alloy of group 2024 (dural). Based on the theoretical study and the requirements of the company was designed a cutting tool with a rounded edge for cutting the unfolded shape of the part on the eccentric press LEN 40 C. A new forming tool from textite material for fluid press Quintus QFC was designed. The parts of the cutting tool were designed primarily from normalized (purchased) parts. Manufactured punch and shearing die are made of commonly available tool steels and processed on the attached drawings. The result of the thesis is a recommendation to invest in a specialized workplace for the production of a unfolded shape and reduction of laboriousness adjusting after forming.
|
|
Learnable Evolution Model for Optimization (LEM)
Weiss, Martin ; Vašíček, Zdeněk (referee) ; Schwarz, Josef (advisor)
Numerical optimization of multimodal or otherwise nontrivial functions has stayed around the peak of the interest of many researchers for a long time. One of the promising methods that appeared is the hybrid approach of the Learnable Evolution Model that combines the well-established ways of artificial intelligence and machine learning with recently popular and efective methods of evolutionary programming. In this work, the method itself was reviewed with respect to what has been already implemented and tested and several possible new implementations of the method were proposed and some of them consequently implemented. The resulting program was then tested against a set of chosen nontrivial real-valued functions and its results were compared to those achieved with EDA algorithms.
|
|
Learnable Evolution Model for Optimization (LEM)
Grunt, Pavel ; Vašíček, Zdeněk (referee) ; Schwarz, Josef (advisor)
My thesis is dealing with the Learnable Evolution Model (LEM), a new evolutionary method of optimization, which employs a classification algorithm. The optimization process is guided by a characteristics of differences between groups of high and low performance solutions in the population. In this thesis I introduce new variants of LEM using classification algorithm AdaBoost or SVM. The qualities of proposed LEM variants were validated in a series of experiments in static and dynamic enviroment. The results have shown that the metod has better results with smaller group sizes. When compared to the Estimation of Distribution Algorithm, the LEM variants achieve comparable or better values faster. However, the LEM variant which combined the AdaBoost approach with the SVM approach had the best overall performance.
|
|
Reproduction of \kur{Achillea millefolium} agg. and \kur{Achillea ptarmica} in meadows and verges
TOMŠOVÁ, Pavla
Several changes in land use during the last 50 years of 20th century had a significant impact on the composition of traditional meadows. The abundance of many plant species typical for traditionally managed meadows has declined. But some of these species have found a refuge in habitats such as field margins and road verges. The aim of this study was to describe how the reproduction success of two related Achillea species A. millefolium agg. and A. ptarmica depends on (i) the particular habitat in which they grow (meadow/verge); (ii) the abundance of pollinators in the study site; and (iii) timing of flowering within the season. Moreover, the longevity of individual flowers of the two taxa has been studied in order assess the width of the time frame the reproduction takes place in. These objectives were achieved by means of measuring plant total seed production and germination as proxies of reproductive success both in meadow and verge populations at the beginning, peak and end of the flowering season of the two species in 2012. The plants were chosen at plots, where concurrently a pollinator survey has been conducted as the part of the broader project. The durations of the male and female phases of individual flowers have been studied in separate experiments.
|