|
|
Dynamics of a cantilever beam with piezoelectric sensor: Parameter identification
Cimrman, Robert ; Kolman, Radek ; Musil, Ladislav ; Kotek, Vojtěch ; Kylar, Jaromír
The piezoelectric materials are electroactive materials often applied for real-time sensing or structural health monitoring, both important research topics in dynamics. Mathematical models of such structures have to allow also for the external electrical circuits and contain several material parameters that need to be identified from experiments. We present a model of a simple experiment involving dynamics of a cantilever beam with an attached piezoelectric sensor excited by a suddenly removed weight. The external circuit can be taken into account as having either a finite or infinite resistance. We also outline the parameter identification procedure based on automatic differentiation and present the experimental and numerical results.
|
|
|
Parameter Identification of Induction Motor Equivalent Circuits
Karpenko, Oleksii ; Cipín, Radoslav (referee) ; Toman, Marek (advisor)
In the control of an asynchronous motor, it is important to determine the parameters of its replacement connection with the greatest accuracy. This work focuses on the identification of these parameters by various methods, which are represented in the form of a -network and a T-network. The bachelor thesis is divided into five parts. The first part focuses on the theory of the asynchronous motor, its use, distribution and properties. The second part deals with replacement connections of the asynchronous motor, their specifics and the possibility of its conversion. The third chapter focuses on the theoretical procedures of parameter identification using various methods as well as the theory of engine tests. In the last two chapters, the theoretical findings from the previous parts are experimentally verified on a real engine sample. Subsequently, the results obtained by different methods were compared with each other.
|
|
|
Simulation of nonlinear response of construction materials and problem of identitifaction of material parameters
Hokeš, Filip ; Kala, Zdeněk (referee) ; Krejsa,, Martin (referee) ; Kala, Jiří (advisor)
The issue of parameter idenetification within numerical simulations can be considered a necessary step in the use of mathematical models, such as complex material models of building materials utilized in finite element method. The problem is particularly epmhasized in the case of anisothropic materials but also in the case of concrete, which behaves differently in tension and compression and which shows different response under rapid-dynamic and long-term loading. Correct capture of the concrete response in the computation requires usage of plasticity theory, damage theory, visco-elasticity and visco-plasticity or their mutual combinations. It results in development of material models and rheological schemes with large amount of parameters that are not commonly available in standards. The disseration presents application of meta-heuristic algorithm Particle Swarm for parameter identification. The success of the method is demonstrated in a pair of cases, first on the identification of elastoplastic material model parameters from stress strain diagram and then to identify the values of stiffness constants and viscosities of the rheological scheme from a creep curve.
|
|
| |
|
|
On the Optimal Initial Conditions for an Inverse Problem of Model Parameter Estimation - a Complementarity Principle
Matonoha, Ctirad ; Papáček, Š.
This contribution represents an extension of our earlier studies on the paradigmatic example of the inverse problem of the diffusion parameter estimation from spatio-temporal measurements of fluorescent particle concentration, see [6, 1, 3, 4, 5]. More precisely, we continue to look for an optimal bleaching pattern used in FRAP (Fluorescence Recovery After Photobleaching), being the initial condition of the Fickian diffusion equation maximizing a sensitivity measure. As follows, we define an optimization problem and we show the special feature (so-called complementarity principle) of the optimal binary-valued initial conditions.
|
|
|
Control Rotational Inverse Pendulum Laboratory Model
Šnajder, Jan ; Bastl, Michal (referee) ; Brablc, Martin (advisor)
Aim of this bachelor thesis is modeling, identification and control of rotational inverted pendulum. The thesis describes parts of real system and its control. The work also contains creation of simulation model in Matlab/Simulink. Simulation model is respresented by mathematical equations created by second type Lagrange equations. Estimation of parameters is based on reduced systems and these parameters have been validated. Control of system is based on state-space model. The paper includes description of linearization, design of state-space control created by LQR method and creation of state-space observer. The last part of the study is focused on state observer experiments, friction compensation by few static friction models and by adding noise to compensation.
|
|
|
Experimental analysis of torsion vibration of hard rubbers under large deformation
Šulc, Petr ; Pešek, Luděk ; Bula, Vítězslav ; Cibulka, Jan ; Košina, Jan
The paper deals with a stress analysis of hard rubber under large torsion deformations. This study was motivated by effort to enhance the theoretical background for experimental evaluation of material behaviour of hard rubbers on our test rig. First the Mooney Rivlin model (MRM) for shear case of loading was developed and then MRM constants were attained by fitting of the MRM to the experimental torsiondeformation curve. Then the tuned MRM cylindrical model was tested under torsion loading for evaluation of stress state. Besides the radial distribution of shear stress and strain the attention was paid to evaluation of axial stresses. It could help to assess the influence of the tension stresses on the tangential deformations of the test sample during large torsion.
|
|
|
BLDC drive control
Havlíček, Jiří ; Buchta, Luděk (referee) ; Otava, Lukáš (advisor)
The bachelor thesis is focused on problems of sensor control of Brushless DC Electric motor (BLDC) using hall-effect sensors. The BLDC motor is connected to 3-Phase low-voltage board which was produced by Freescale Semiconductor, Inc. The board is adapted to control electrical motors by MC56F8006 signal microcontroller. The next part is devoted to implementation of interconnection of the machine with the Matlab/Simulink programme using the MF 624 multifunction I/O card. Cascaded control, current and speed loops are used for the final configuration. The former loop is realized in the signal microcontroller, the latter one in the Simulink programme.
|
|
|
Frequency method for motor parameters identification
Zimmert, Martin ; Buchta, Luděk (referee) ; Pohl, Lukáš (advisor)
This bachelor´s thesis deals with improved frequency method for motor parameters Identification. The goal of this thesis is to get an overview about identification method and simulate it on motor model in the Simulink and then to verify functionality of the method on the real motor. Firstly there are informations about a motor with permanent magnets and its control. Then there is an application of identification method for complete motor model. In the next part is measurement of the real motor. In the end is design of controller and discrete filter of identification signal.
|
|
|
Fixed-wing UAV simulation modeling
Příleský, Libor ; Mašek, Petr (referee) ; Krejsa, Jiří (advisor)
Nowadays number of UAVs is still increasing. If we want to design their autopilot system, we need a simulation model to test and tune our autopilot design. This thesis is about creating such model fo small fixed wing UAV. This thesis contain information how to create the model, which equations to use and what various parameters mean. One of the possibilities is parameter identification from measured data. In later part of this paper is a small research of this topic and implementation of two of them. Virtual model, which is obtained at the end of this work, accomplished its primary goal and was used to initial autopilot design. Implemented identification methods worked as well, but we didn't succeed in upgrading the model parameters due to the defects during the data measurement.
|
guest ::
