|
|
Dynamic model of a heat pump with plate heat exchangers and ecological refrigerant
Strmiska, Pavel ; Řehák Kopečková, Barbora (referee) ; Pokorný, Jan (advisor)
For the optimization of a heat pump, there is a need to possess its dynamic model. One of the freely available tools for creating such a model is the language Modelica and its open-source compiler OpenModelica. The thesis aims to research options for creating dynamic models of refrigeration circuits with the ecological refrigerant R290 and its components in Modelica, to create the circuit and subsequently validate it using previously measured data. Models of basic components of heat pumps - compressor, condenser, evaporator and expansion valve are created in a way described in the thesis. A simplified model of a refrigerant loop is also created
|
|
|
Validity of commercially available smartwatches when estimating energy expenditure
Kopecký, David ; Pokorný, Jan (referee) ; Řehák Kopečková, Barbora (advisor)
Metabolic rate is an energy quantity. Its value is a key input not only in thermophysiological models, but also in the calculation of heat stress indices or complex models predicting a person's thermal sensation. With the dynamic development of technology, sporttesters are able to estimate energy expenditure, among other things. It is a non-invasive method of estimating energy expenditure, or metabolic rate, using associated body functions to predict it. This work deals with the validity of the accuracy of metabolic rate estimation using the Garmin Fenix 6X Pro sporttester in comparison with indirect calorimetry. The accuracy of the metabolic rate estimation was tested for the activities of lying down, administrative work and cycling with regulated mechanical resistance of 100 and 150 W. 15 male subjects aged 28 ± 5 years, with a weight of 83.2 ± 13.5 kg and a height of 180 ± 7 cm participated in the study. In terms of the results, it is possible to observe a lower accuracy of prediction of metabolic rate for low-intensity activities, with the resulting MAPE being around 19-43 %. On the other hand, for activities with moderate activity, the prediction of the metabolic rate by the used sporttester proved to be sufficiently accurate, with a resulting MAPE in the range of 6–8 %.
|
|
|
GA 19-07635S: Outputs and Results
Rehák, Branislav
This manuscript aims to deliver a survey of results obtained during the solution of the project No. GA19-07635S of the Czech Science Foundation. The timespan dedicated to the work on this project was 1.3.2019 - 30.6.2022. The main area dealt with were\nnonlinear multi-agent systems and their synchronization, further, attention was paid to some auxiliary results in the area of nonlinear observers. This Report briefly introduces the Project, provides a summary of the results obtained and also sketches an outline how these results will be applied and extended in future.
|
|
|
A NUMERICAL METHOD FOR THE SOLUTION OF THE NONLINEAR OBSERVER PROBLEM
Rehák, Branislav
The central part in the process of solving the observer problem for nonlinear systems is to nd a solution of a partial differential equation of first order. The original method proposed to solve this equation used expansions into Taylor polynomials, however, it suffers from rather restrictive assumptions while the approach proposed here allows to generalize these requirements. Its characteristic feature is that it is based on the application of the Finite Element\nMethod. An illustrating example is provided.
|
|
|
REGULATORY NETWORK OF DRUG-INDUCED ENZYME PRODUCTION: PARAMETER ESTIMATION BASED ON THE PERIODIC DOSING RESPONSE MEASUREMENT
Papáček, Štěpán ; Lynnyk, Volodymyr ; Rehák, Branislav
The common goal of systems pharmacology, i.e. systems biology applied to the eld of pharmacology, is to rely less on trial and error in designing an input-output systems, e.g. therapeutic schedules. In this paper we present, on the paradigmatic example of a regulatory network of drug-induced enzyme production, the further development of the study published by Duintjer Tebbens et al. (2019) in the Applications of Mathematics. Here, the key feature is that the nonlinear model in form of an ODE system is controlled (or periodically forced) by an input signal being a drug intake. Our aim is to test the model features under both periodic and nonrecurring dosing, and eventually to provide an innovative method for a parameter estimation based on the periodic dosing response measurement.
|
|
| |
|
|
DCTOOL-A5
Bakule, Lubomír ; Papík, Martin ; Rehák, Branislav
DCTOOL-A5 presents draft of a manuscript, which is intended to be submitted for publication. This report presents a new method for the decentralized event-triggered control design for large-scale uncertain systems. The results are formulated and proved in terms of linear matrix inequalities. Two design problems are solved: For interconnected systems without any quantization and for interconnected systems with local logarithmic quantizers. Results are illustrated by an example.
|
|
|
DCTOOL-A4
Bakule, Lubomír ; Papík, Martin ; Rehák, Branislav
DCTOOL-A4 report presents draft of a manuscript, which is intended to be submitted for publication. The report provides a novel systematic approach to the analysis of asymptotic stability for output event-triggered uncertain centralized control systems. A class of nonlinear but nominally linear systems possessing unknown time-varying bounded uncertainties with known bounds is considered. Uncertainties are allowed in all system matrices. Original LMI-based suffi cient conditions are derived to guarantee asymptotic stability of closed-loop systems with both static output and observer-based feedback loop under even-triggered control. Both these output feedback strategies are extended to model-based uncertain control systems with\nquantized measurements. A logarithmic quantizer is considered. The Lyapunov-based approach and convex optimization serve as the main methods to derive the asymptotic LMI-based stability conditions. Bounds on the inter-event times to avoid the Zeno-effect are proved for all the cases considered. Finally, feasibility and effi ciency of the proposed strategies is demonstrated by providing numerical examples.
|
|
|
DCTOOL-A3
Bakule, Lubomír ; Papík, Martin ; Rehák, Branislav
DCTOOL-A3 is a documentation of Matlab routines developed for the design of decentralized control of large scale complex systems. The current beta version covers three areas as follows:\nReport 4.1 deals with the event-triggered control design for unstructured uncertain systems. Both non-quantized and quantized feedback is analyzed. The results are given in terms of linear matrix inequalities (LMIs). Logarithmic quantizer is used. Numerical example illustrates the effectiveness of the presented results.\nReport 4.2 presents a new decentralized overlapping wireless control design with a switched communication protocol. The method is applied by simulations on the 20-story building structure including the test of robustness of the methods against sensor failures and network node dropouts of a digital network.\nReport 4.3 presents the construction of a new decentralized wireless controller and a set of heuristic algorithms for evaluation of packet dropouts, sensor faults and actuator faults. The digital network operates at the standard frequency used in well-known widely-used industrial protocols. The results are tested at the Benchmark model decomposed into two disjoint substructures. The results are published. Thus, the details are omitted here.
|
|
| |
guest ::
RSS feed.