Optimalizace betonových prvků pomocí genetických algoritmů

Abstract

This abstract can include spoilers to the thesis. If you would like to enjoy the process of reading more, I kindly suggest you skipping it and getting back to it later on. In the following work, the author has developed a code that uses the Finite Element Analysis to calculate structure's characteristics, such as deflections and stress in each member. In the next step, she focused on optimizing chosen 2D truss structure to the best shape possible via Genetic Algorithm application. The optimization is performed on selected nodes locations and crosssections of each element. In the first part of the work, the author discusses methods used to reach the goal. All of them were implemented via programming language Python and programming framework PyCharm. Then a guide through the written code is provided. The main objective was to provide a reader with an excessive amount of information to create a similar code or to continue in a development of the existing one and expanding it to different usages, such as, for example, 3D structures. Further, a few examples of the work are being discussed in detail to introduce the reader to how, by adjusting different parameters, the outcome can be changed. A discussion on the implications of the model is provided and options for the future work are outlined. Author has decided to share her progress with others and have everything open-sourced to allow anyone to contribute and continue with the idea and build upon it. You can find the code on GitHub (https://github.com/margkuval/Thesis), a platform created mainly for code sharing. In case you shall have any questions related to the code, please address them using the GitHub platform. As for the thesis itself, you can find an online version on Research Gate (https://www.researchgate.net/profile/Margarita\_Kuvaldina) The author would kindly like to ask you for citing whenever you use any part of her code or thesis. Thank you for understanding.

This abstract can include spoilers to the thesis. If you would like to enjoy the process of reading more, I kindly suggest you skipping it and getting back to it later on. In the following work, the author has developed a code that uses the Finite Element Analysis to calculate structure's characteristics, such as deflections and stress in each member. In the next step, she focused on optimizing chosen 2D truss structure to the best shape possible via Genetic Algorithm application. The optimization is performed on selected nodes locations and crosssections of each element. In the first part of the work, the author discusses methods used to reach the goal. All of them were implemented via programming language Python and programming framework PyCharm. Then a guide through the written code is provided. The main objective was to provide a reader with an excessive amount of information to create a similar code or to continue in a development of the existing one and expanding it to different usages, such as, for example, 3D structures. Further, a few examples of the work are being discussed in detail to introduce the reader to how, by adjusting different parameters, the outcome can be changed. A discussion on the implications of the model is provided and options for the future work are outlined. Author has decided to share her progress with others and have everything open-sourced to allow anyone to contribute and continue with the idea and build upon it. You can find the code on GitHub (https://github.com/margkuval/Thesis), a platform created mainly for code sharing. In case you shall have any questions related to the code, please address them using the GitHub platform. As for the thesis itself, you can find an online version on Research Gate (https://www.researchgate.net/profile/Margarita\_Kuvaldina) The author would kindly like to ask you for citing whenever you use any part of her code or thesis. Thank you for understanding.