|
|
Decomposition and Forecast in Time Series
Šramková, Kristína ; Holec, Tomáš (referee) ; Bednář, Josef (advisor)
Bachelor thesis is focused on statistic methods for analysis and modelling time series. In particular decomposition techniques, which decompose time series into particular components and model trend component to predict future values. First part of thesis is focused on summary of essential knowledge of regression, basic properties of time series and their decomposition models. Hereinafter are discussed models of trend curves. In conclusion the mentioned methods are applied on real data and its modeling in software Statistica.
|
|
| |
|
|
Prediction of Time Series Using Statistical Methods
Beluský, Ondrej ; Bidlo, Michal (referee) ; Schwarz, Josef (advisor)
Many companies consider essential to obtain forecast of time series of uncertain variables that influence their decisions and actions. Marketing includes a number of decisions that depend on a reliable forecast. Forecasts are based directly or indirectly on the information derived from historical data. This data may include different patterns - such as trend, horizontal pattern, and cyclical or seasonal pattern. Most methods are based on the recognition of these patterns, their projection into the future and thus create a forecast. Other approaches such as neural networks are black boxes, which uses learning.
|
|
|
Implementation of matrix decomposition and pseudoinversion on FPGA
Röszler, Pavel ; Rajmic, Pavel (referee) ; Smékal, David (advisor)
The purpose of this thesis is to implement algorithms of matrix eigendecomposition and pseudoinverse computation on a Field Programmable Gate Array (FPGA) platform. Firstly, there are described matrix decomposition methods that are broadly used in mentioned algorithms. Next section is focused on the basic theory and methods of computation eigenvalues and eigenvectors as well as matrix pseudoinverse. Several examples of implementation using Matlab are attached. The Vivado High-Level Synthesis tools and libraries were used for final implementation. After the brief introduction into the FPGA fundamentals the thesis continues with a description of implemented blocks. The results of each variant were compared in terms of timing and FPGA utilization. The selected block has been validated on the development board and its arithmetic precision was analyzed.
|
|
|
Decomposition of Volatile Organic Compounds in Gliding Arc Discharge
Grossmannová, Hana ; Kapička, Vratislav (referee) ; Pekárek,, Stanislav (referee) ; Krčma, František (advisor)
The aim of this thesis was to elaborate the issue of the decomposition of volatile organic compounds in the Gliding Arc plasma discharge at atmospheric pressure. Technologies based on nonthermal plasma could offer a good alternative to conventional techniques for the decomposition of volatile organic compounds, such as thermal and catalytic oxidation. Gliding Arc discharge (GidArc) is a widely exploited nonthermal plasma source used for many industrial applications, such as air-pollution control. The energy efficiency, reaction selectivity or production of specific species may be achieved in this kind of plasma, and thus for various chemical processes it can be much more effective then in conventional techniques. Presented experiments are linked to the previous results published in diploma thesis, which gave us the basics for construction of new reactor and optical emission spectroscopy measurements have been done to characterize the plasma. Toluene (aromatic, unsaturated), cyclohexane (aromatic, saturated) and hexane (aliphatic, saturated) were used as the model compounds for these experiments in the concentration range from hundreds to thousands ppm. Results focused on the electrical parameters of the reactor were carried out, with the aim to operate the system at a lower energy cost. In order to get the time-resolved diagnostics of the moving plasma channel, the evolution of the plasma channel was recorded continuously by using a high-speed video camera. In next part of the work, some results concerning generation of low molecular products like nitric oxide, nitrogen dioxide, hydrogen and carbon monoxide on the discharge conditions are presented. In combustion process, undesirable mixture of toxic highmolecular by-products can be formed. Samples were therefore analysed in gas chromatograph linked to mass spectrometer, to characterize the chemical transformation pathway of VOC in plasma.
|
|
|
Implementation of edge detector using wavelet transform
Pálka, Zbyněk ; Rášo, Ondřej (referee) ; Růčka, Lukáš (advisor)
This thesis is focused on edge detection in image. In theoretical part are contained genarally used methods of edge detection using first and second-order derivate and both of mentioned methods are described here. Further it’s deiscribed here continuous, descrete and two dimensional descrete wavelet transform and process of noise removing in image by descrete wavelet transform. In next part are analysed two methods of edge detection using wavelet transform and their possible realizations in program Matlab. In practical part of thesis is in detail described algorithm of program on edge detection using wavelet transform and it‘s described here individual functions of program. The main content of practical part are visual results of wavelet edge detector and their comparison with Canny, Prewitt and Sobel edge detector.
|
|
|
ON THE PASNIK
Siegelová, Kateřina ; Vlková, Veronika (referee) ; Klímová, Barbora (advisor)
The topic of my thesis On the Pasník a time and decomposition. I have produced small houses from natural materials and placed them in a forest. The subject of my work is to record the flow of time. The outcome of my work is time based large-formal drawing. The buildings on the drawing gradually collapse and disappear. Thus I have created the time drawing.
|
|
|
Hydrological Time Series Analysis for Chosen Sub Catchment in Morava River Basin
Černý, Vojtěch ; Benáčková, Kateřina (referee) ; Marton, Daniel (advisor)
Current climate conditions raise questions about how climate change should affect the hydrological regime of the Czech Republic. The main questions are: I) Is climate change serious and is it already tangible in hydrological series? II) Are there significant changes in the climatic conditions, or in the flow of the river network? III) How does climate change affect hydrological conditions in the catchment? IV) Are there non-stationary tendencies in the hydrological series caused by climate change? The aim of this work was to perform a basic analysis of hydrological time series and estimated its possible changes. The analysis was done with using autocorrelation function, spectral density, moving averages methods, statistical characteristics, empirical line of probability of exceeding and last but not least by decomposing the time series. The practical application was applied on time series of average monthly flows determined in the hydrometric profile of Dalečín in the Svratka river basin.
|
|
|
Static image enhancement using wavelet transform
Candrák, Matúš ; Rajmic, Pavel (referee) ; Smékal, Zdeněk (advisor)
In tomography and ultrasound signal processing, there is the noise build-up into the processing. Bachelor's thesis deals with static images highlighting, with denoising using wavelet transformation and edge detection with basic operators. This work describes some types of wavelts used for denoising of image and basic operators for edge detection in the image. The last part deals with a particular application for image processing, which was created in MATLAB.
|
|
|
Water Management Analysis of Storage Capacity of the Vranov Reservoir
Vítková, Lucie ; Havlík, Martin (referee) ; Marton, Daniel (advisor)
The current persistent drought and changes in the climate system have raised water managers expert questions about how to manage water resources in the future. The manifestation of climate change in hydrological series and their influence on the magnitude of threats to the storage functions of reservoirs are more often investigated. Already today, long-term shortages of storage capacity in reservoirs lead to the introduction of special manipulations on water structures. The aim of the thesis is to perform the analysis of time series, respectively decomposition of hydrological series average annual and monthly discharges. Create extended hydrological bases using synthetic discharge series generators and develop a comprehensive analysis of storage volume without considering losses even with the introduction of losses from the water surface vapor in the UNCE RESERVOIR program. The created discharge series are compared and evaluated on the basis of statistical characteristics and reservoir storages results with the real discharge series. The practical app is conducted on the Vranov reservoir in the Dyje River Basin.
|
guest ::
