|
|
Nonlinear Control of Complex Systems by Utilization of Evolutionary Approaches
Minář, Petr ; Ošmera, Pavel (referee) ; Oplatková,, Zuzana Komínková (referee) ; Matoušek, Radomil (advisor)
Control theory of complex systems by utilization of artificial intelligent algorithms is relatively new science field and it can be used in many areas of technical practise. Best known algorithms to solved similar tasks are genetic algorithm, differential evolution, HC12 Nelder-Mead method, fuzzy logic and grammatical evolution. Complex solution is presented at selected examples from mathematical nonlinear systems to examples of anthems design and stabilization of deterministic chaos. The goal of this thesis is present examples of implementation and utilization of artificial algorithms by multi-objective optimization. To achieve optimal results is used designed software solution by multi-platform application, which used Matlab and Java interfaces. The software solution integrate every algorithms of this thesis to complex solution and it extends possible application of those approaches to real systems and practical world.
|
|
|
Parallel genetic algorithm
Trupl, Jan ; Kobliha, Miloš (referee) ; Jaroš, Jiří (advisor)
The thesis describes design and implementation of various evolutionary algorithms, which were enhanced to use the advantages of parallelism on the multiprocessor systems along with ability to run the computation on different machines in a computer network. The purpose of these algorithms is to find the global extreme of function of $n$ variables. In the thesis, there are demonstrated various optimization problems, and their effective solution with the help of evolutionary algorithms. There are also described interface libraries MPI(Message Passing Interface) and OpenMP, in the extent needed to understand the problematic of parallel evolutionary algorithms.
|
|
|
Toolbox for the cooperation of MATLAB and external simulation programs
Moravec, Petr ; Sekanina, Lukáš (referee) ; Oliva, Lukáš (advisor)
In this Master's thesis scripting interface of two programs CST Microwave studio and Ansoft HFSS for the purpose of analysis of electromagnetic structures is described. The work is focuses control of these programs with help of scripting languages and system's interface of MS Windows XP. Next the process of connecting programs with MATLAB is shown on commented scripts together with an example of complete analysis of a chosen problem, and the import and export of results results in MATLAB. Further the functions which form programming interface between MATLAB and simulation programs are designed and implemented. The interconnection layer makes the complete control of simulating programs possible using the function description published in the official documentation of used simulation programs. The layer is described in reference manual in detail and it is used for optimization with use of Particle swarm optimalization (PSO) of planar antenna model. Then there is presented another usage of the layer for an implementation of global optimization methods - SOMA and DE including suggestion of process for comparison efficiency of optimization algorithms on simple electromagnetic models.
|
|
|
Parallel genetic algorithm
Trupl, Jan ; Kobliha, Miloš (referee) ; Jaroš, Jiří (advisor)
The thesis describes design and implementation of various evolutionary algorithms, which were enhanced to use the advantages of parallelism on the multiprocessor systems along with ability to run the computation on different machines in a computer network. The purpose of these algorithms is to find the global extreme of function of $n$ variables. In the thesis, there are demonstrated various optimization problems, and their effective solution with the help of evolutionary algorithms. There are also described interface libraries MPI(Message Passing Interface) and OpenMP, in the extent needed to understand the problematic of parallel evolutionary algorithms.
|
|
|
Nonlinear Control of Complex Systems by Utilization of Evolutionary Approaches
Minář, Petr ; Ošmera, Pavel (referee) ; Oplatková,, Zuzana Komínková (referee) ; Matoušek, Radomil (advisor)
Control theory of complex systems by utilization of artificial intelligent algorithms is relatively new science field and it can be used in many areas of technical practise. Best known algorithms to solved similar tasks are genetic algorithm, differential evolution, HC12 Nelder-Mead method, fuzzy logic and grammatical evolution. Complex solution is presented at selected examples from mathematical nonlinear systems to examples of anthems design and stabilization of deterministic chaos. The goal of this thesis is present examples of implementation and utilization of artificial algorithms by multi-objective optimization. To achieve optimal results is used designed software solution by multi-platform application, which used Matlab and Java interfaces. The software solution integrate every algorithms of this thesis to complex solution and it extends possible application of those approaches to real systems and practical world.
|
|
|
Nonlinear Control of Complex Systems by utilization of Evolutionary Approaches
Minář, Petr ; Matoušek, Radomil (advisor)
Control theory of complex systems by utilization of artificial intelligent algorithms is relatively new science field and it can be used in many areas of technical practise. Best known algorithms to solved similar tasks are genetic algorithm, differential evolution, HC12 Nelder-Mead method, fuzzy logic and grammatical evolution. Complex solution is presented at selected examples from mathematical nonlinear systems to examples of anthems design and stabilization of deterministic chaos. The goal of this thesis is present examples of implementation and utilization of artificial algorithms by multi-objective optimization. To achieve optimal results is used designed software solution by multi-platform application, which used Matlab and Java interfaces. The software solution integrate every algorithms of this thesis to complex solution and it extends possible application of those approaches to real systems and practical world.
|
|
|
Toolbox for the cooperation of MATLAB and external simulation programs
Moravec, Petr ; Sekanina, Lukáš (referee) ; Oliva, Lukáš (advisor)
In this Master's thesis scripting interface of two programs CST Microwave studio and Ansoft HFSS for the purpose of analysis of electromagnetic structures is described. The work is focuses control of these programs with help of scripting languages and system's interface of MS Windows XP. Next the process of connecting programs with MATLAB is shown on commented scripts together with an example of complete analysis of a chosen problem, and the import and export of results results in MATLAB. Further the functions which form programming interface between MATLAB and simulation programs are designed and implemented. The interconnection layer makes the complete control of simulating programs possible using the function description published in the official documentation of used simulation programs. The layer is described in reference manual in detail and it is used for optimization with use of Particle swarm optimalization (PSO) of planar antenna model. Then there is presented another usage of the layer for an implementation of global optimization methods - SOMA and DE including suggestion of process for comparison efficiency of optimization algorithms on simple electromagnetic models.
|
guest ::
