Dissertations / Theses on the topic 'Robust model validation'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 26 dissertations / theses for your research on the topic 'Robust model validation.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
Davis, Robert Andrew. "Model validation for robust control." Thesis, University of Cambridge, 1995. https://www.repository.cam.ac.uk/handle/1810/251990.
Full textChai, Li. "Multirate periodic systems : robust model validation and stabilization /." View Abstract or Full-Text, 2002. http://library.ust.hk/cgi/db/thesis.pl?ELEC%202002%20CHAI.
Full textAl-Takrouri, Saleh Othman Saleh Electrical Engineering & Telecommunications Faculty of Engineering UNSW. "Robust state estimation and model validation techniques in computer vision." Publisher:University of New South Wales. Electrical Engineering & Telecommunications, 2008. http://handle.unsw.edu.au/1959.4/41002.
Full textSakouvogui, Kekoura. "Robust Capital Asset Pricing Model Estimation through Cross-Validation." Thesis, North Dakota State University, 2018. https://hdl.handle.net/10365/29019.
Full textMadden, Ryan J. "Development of Robust Control Techniques towards Damage Identification." Cleveland State University / OhioLINK, 2016. http://rave.ohiolink.edu/etdc/view?acc_num=csu1460986638.
Full textHuang, Chao-Min. "Robust Design Framework for Automating Multi-component DNA Origami Structures with Experimental and MD coarse-grained Model Validation." The Ohio State University, 2020. http://rave.ohiolink.edu/etdc/view?acc_num=osu159051496861178.
Full textMaillard, Guillaume. "Hold-out and Aggregated hold-out Aggregated Hold-Out Aggregated hold-out for sparse linear regression with a robust loss function." Thesis, université Paris-Saclay, 2020. http://www.theses.fr/2020UPASM005.
Full textIn statistics, it is often necessary to choose between different estimators (estimator selection) or to combine them (agregation). For risk-minimization problems, a simple method, called hold-out or validation, is to leave out some of the data, using it to estimate the risk of the estimators, in order to select the estimator with minimal risk. This method requires the statistician to arbitrarily select a subset of the data to form the "validation sample". The influence of this choice can be reduced by averaging several hold-out estimators (Aggregated hold-out, Agghoo). In this thesis, the hold-out and Agghoo are studied in various settings. First, theoretical guarantees for the hold-out (and Agghoo) are extended to two settings where the risk is unbounded: kernel methods and sparse linear regression. Secondly, a comprehensive analysis of the risk of both methods is carried out in a particular case: least-squares density estimation using Fourier series. It is proved that aggregated hold-out can perform better than the best estimator in the given collection, something that is clearly impossible for a procedure, such as hold-out or cross-validation, which selects only one estimator
Corbier, Christophe. "Contribution à l’estimation robuste de modèles dynamiques : Application à la commande de systèmes dynamiques complexes." Thesis, Paris, ENSAM, 2012. http://www.theses.fr/2012ENAM0041/document.
Full textComplex dynamic systems identification remains a concern when prediction errors contain innovation outliers. They have the effect to damage the estimated model if the estimation criterion is badly chosen and badly adapted. The consequence is the contamination of the distribution of these errors; this distribution presents heavy tails and deviates of the normal distribution. To solve this problem, there is a robust estimator's class, less sensitive to the outliers, which treat the transition between residuals of very different levels in a softer way. The Huber's M-estimators belong to this class. They are associated to a mixed L2 - L1 norm, related to a disturbed Gaussian distribution model, namely gross error model. From this formal context, in this thesis we propose a set of estimation and validation tools of black-box linear and pseudo-linear models, with extension of the noise interval to low values of the tuning constant in the Huber's norm. We present the convergence properties of the robust estimation criterion and the robust estimator. We show that the extension of the noise interval reduces the sensitivity of the bias of the estimator and improves the robustness to the leverage points. Moreover, for a pseudo-linear model structure, we present a new context, named L-FTE, with a new method to determine L, in order to linearize the gradient and the Hessien of estimation criterion and the asymptotic covariance matrix of the estimator. From these expressions, a robust version of the FPE validation criterion is established and we propose a new decisional tool for the estimated model choice. Experiments on simulated and real systems are presented and analyzed
Wroblewski, Adam C. "Model Identification, Updating, and Validation of an Active Magnetic Bearing High-Speed Machining Spindle for Precision Machining Operation." Cleveland State University / OhioLINK, 2011. http://rave.ohiolink.edu/etdc/view?acc_num=csu1318379242.
Full textAlvarado, Christiam Segundo Morales. "Estudo e implementação de métodos de validação de modelos matemáticos aplicados no desenvolvimento de sistemas de controle de processos industriais." Universidade de São Paulo, 2017. http://www.teses.usp.br/teses/disponiveis/3/3139/tde-05092017-092437/.
Full textLinear model validation is the most important stage in System Identification Project because, the model correct selection to represent the most of process dynamic allows the success in the development of predictive and robust controllers, within identification technique finite number and around the operation point. For this reason, the development of linear model validation methods is the main objective in this Thesis, taking as a tools of assessing the statistical, dynamic and robustness methods. Fuzzy system is the main component of model linear validation system proposed to analyze the results obtained by the tools used in validation stage. System Identification project is performed through operation real data of a pH neutralization pilot plant, located at the Industrial Process Control Laboratory, IPCL, of the Escola Politécnica of the University of São Paulo, Brazil. In order to verify the validation results, all modes are used in QDMC type predictive controller, to follow a set point tracking. The criterions used to assess the QDMC controller performance were the speed response and the process variable minimum variance index, for each model used. The results show that the validation system reliability were 85.71% and 50% projected for low and high non-linearity in a real process, respectively, linking to the performance indexes obtained by the QDMC controller.
Kartowisastro, Iman Herwidiana. "Quantitative model validation of manipulative robot systems." Thesis, City University London, 1991. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.280671.
Full textCavalcanti, santos João. "Model Predictive Tracking Control of Cable-Driven Parallel Robots : From Concept to Real-Time Validation." Thesis, Montpellier, 2020. http://www.theses.fr/2020MONTS017.
Full textThis thesis addresses the position tracking control of Cable-Driven Parallel Robot (CDPR) within the framework of the European H2020 project named Hephaestus. The main goal of this project is to develop a robotic solution for the installation of curtain wall modules on building facades. An essential requirement is that the CDPR should safely operate close to the system constraints. It was observed that state-of-the-art control schemes do not cope with this requirement. The control strategies used in the design of such schemes are not able to consider system constraints as an integral part of the main controller.Since Model Predictive Control (MPC) is one of the few control strategies able to explicitly handle the system constraints, this thesis is focused on the design and analysis of MPC schemes for position tracking of CDPRs. Two approaches are then proposed: a linear MPC and a nonlinear MPC (NMPC).The proposed linear MPC is based on a linear approximation of the CDPR dynamic model. The Experimental tests proved that the linear MPC may safely operate close to system constraints. This capability is validated by applying a desired trajectory that cannot be performed without violating the cable tension limits. In this case, the proposed linear MPC scheme is able to perform a trajectory tracking as best as possible while satisfying the cable tension bounds. Conversely, state-of-the-art control schemes are not able to suitably respond under such conditions. Comparing the behavior obtained with the proposed linear MPC and a state-of-the-art control scheme, one may conclude that the capability to operate close to the system constraints represents an important result related to the safety of the operation of CDPRs.Nevertheless, it was noted that the proposed linear MPC may be sensitive to increased nonlinearities. The precision of positioning tracking may be deteriorated for trajectories presenting relatively high velocities. Accordingly, an NMPC able to consider the system nonlinearities is proposed. In contrast to its linear counterpart, the stability of the resulting closed-loop system could be analyzed. Details on its numerical implementation are presented and the improved performance is validated through simulations.In addition to the design of MPC control schemes, this thesis also presents contributions related to the kinematic model of CDPRs and the control of cable tensions. A Forward Kinematic (FK) algorithm considering the pulley kinematics is proposed. An explicit expression for the differential kinematics enabled the implementation of a numerical solution of the nonlinear least-squares system representing the FK problem. Its convergence capabilities are evaluated experimentally and numerically.It is worth noting that the algorithms and control schemes proposed in this thesis were implemented in an industrial software, which demonstrates the applicability of the proposed solutions in commercial applications
Söderdahl, Fabian. "A Cross-Validation Approach to Knowledge Transfer for SVM Models in the Learning Using Privileged Information Paradigm." Thesis, Uppsala universitet, Statistiska institutionen, 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-385378.
Full textMarangé, Pascale. "Synthèse et filtrage robuste de la commande pour des systèmes manufacturiers sûrs de fonctionnement." Reims, 2008. http://theses.univ-reims.fr/exl-doc/GED00000826.pdf.
Full textThe work presented in this thesis, focuses on the control design by automatic control engineers who can have different levels of competence. The use of real plant can involve control design errors. This raises two original problems; the validation and verification of the control and the requirement to take into the designer. These problems require an approach taking into account the human part. The contributions developed in this thesis focuse on adapting the system to designer knowledge and on the system safety. On the first point, it highlights the educational importance to work on the global system rather than a part. An approach is based on the functional decomposition of the plant to adapt the specifications. This approach helps maintain the global vision of system. On the second point 2 approaches are envisaged to ensure the safety. The first, offline, is based on previous work performed. Supervisory control allows to obtain a sure, robust and without deadlock controller. The second approach, online, uses a filter. This approach allows sending to the plant only validated outputs. The main difficulties of filter lie firstly in its specifications, and secondly in its explanatory capabilities for the designer. A filter structure with 2 levels is proposed. The first one: system validation filter, assuring the plant safety. The second one: functional validation filter enables to verify if specifications are conformed. The proposed approach to ensure that the constraints are properly defined is to formally verify, using the model-checker. The verification procedure determines the sufficient constraints set to avoid deterioration situations. Two applications are presented
Marangé, Pascale. "Synthèse et filtrage robuste de la commande pour des système manufacturiers sûrs de fonctionnement." Phd thesis, Université de Reims - Champagne Ardenne, 2008. http://tel.archives-ouvertes.fr/tel-00353654.
Full textOffermann, Alexis. "Conception et validation d’un robot apte à effectuer des mesures sur ouvrage d’art." Thesis, Compiègne, 2021. http://www.theses.fr/2021COMP2603.
Full textThis PhD. applies in the field of robotics for building inspection. Buildings are understood here as huge constructions such as bridges or churches. Using photogrammetry to analyses the structure of these kind of buildings is nowadays a common technique. To extend the analysis methodologies, we propose in this document to imagine a way to bring tools directly in contact with the buildings. Multiple solution possibilities are presented for a final choice. The chosen solution consists of designing a drone able to tilt its arms to bring the building inspection tool directly onto the wall of the construction. This leads to a double challenge. Firstly, the mechanical part of designing such a complex and non-common drone. Secondly,the development and real time testing of the control algorithm. Eventually, this manuscript proposes to use a particular software architecture joining fast experimentations and validation of the control laws. This platform being very versatile, it is also possible to consider re-using it for different applications in future works. The experimentations reveal good results, promoting the usage of the drone as a carrying vector to bring the tools onto the construction
Py, Frédéric. "Contrôle d'exécution dans une architecture hiérarchisée pour systèmes autonomes." Toulouse 3, 2005. http://www.theses.fr/2005TOU30199.
Full textThere is an increasing need for advanced autonomy in complex embedded real-time systems such as robots or satellites. Still, this raises a major problem : on one side we have complex sys-tems - therefore, hard to validate - with little human intervention, on the other side these systems are used in domains where safety is critical. How can we guaranty that an autonomous system, with high level decisional capabilities, will exhibit a proper behavior and will not jeopardize the mission? The work we present here integrate an on-line execution control component for hierar-chical architectures. We first describe the role of this program. Then we introduce the R2C, our controller based on synchronous hypothesis, and the tool used to generate it. We then discuss why it is important to take into account the decisional components in our controller. We eventu-ally illustrate our contribution with some experimental results. We then conclude and give some possible future work in this area
Boopathy, Komahan. "Uncertainty Quantification and Optimization Under Uncertainty Using Surrogate Models." University of Dayton / OhioLINK, 2014. http://rave.ohiolink.edu/etdc/view?acc_num=dayton1398302731.
Full textLe, Truong Giang. "Using Event-Based and Rule-Based Paradigms to Develop Context-Aware Reactive Applications." Phd thesis, Conservatoire national des arts et metiers - CNAM, 2013. http://tel.archives-ouvertes.fr/tel-00953368.
Full textWåhlin, Peter. "Enhanching the Human-Team Awareness of a Robot." Thesis, Mälardalens högskola, Akademin för innovation, design och teknik, 2012. http://urn.kb.se/resolve?urn=urn:nbn:se:mdh:diva-16371.
Full textAnvändningen av autonoma robotar i vårt samhälle ökar varje dag och en robot ses inte längre som ett verktyg utan som en gruppmedlem. Robotarna arbetar nu sida vid sida med oss och ger oss stöd under farliga arbeten där människor annars är utsatta för risker. Denna utveckling har i sin tur ökat behovet av robotar med mer människo-medvetenhet. Därför är målet med detta examensarbete att bidra till en stärkt människo-medvetenhet hos robotar. Specifikt undersöker vi möjligheterna att utrusta autonoma robotar med förmågan att bedöma och upptäcka olika beteenden hos mänskliga lag. Denna förmåga skulle till exempel kunna användas i robotens resonemang och planering för att ta beslut och i sin tur förbättra samarbetet mellan människa och robot. Vi föreslår att förbättra befintliga aktivitetsidentifierare genom att tillföra förmågan att tolka immateriella beteenden hos människan, såsom stress, motivation och fokus. Att kunna urskilja lagaktiviteter inom ett mänskligt lag är grundläggande för en robot som ska vara till stöd för laget. Dolda markovmodeller har tidigare visat sig vara mycket effektiva för just aktivitetsidentifiering och har därför använts i detta arbete. För att en robot ska kunna ha möjlighet att ge ett effektivt stöd till ett mänskligtlag måste den inte bara ta hänsyn till rumsliga parametrar hos lagmedlemmarna utan även de psykologiska. För att tyda psykologiska parametrar hos människor förespråkar denna masteravhandling utnyttjandet av mänskliga kroppssignaler. Signaler så som hjärtfrekvens och hudkonduktans. Kombinerat med kroppenssignalerar påvisar vi möjligheten att använda systemdynamiksmodeller för att tolka immateriella beteenden, vilket i sin tur kan stärka människo-medvetenheten hos en robot.
The thesis work was conducted in Stockholm, Kista at the department of Informatics and Aero System at Swedish Defence Research Agency.
Vestin, Albin, and Gustav Strandberg. "Evaluation of Target Tracking Using Multiple Sensors and Non-Causal Algorithms." Thesis, Linköpings universitet, Reglerteknik, 2019. http://urn.kb.se/resolve?urn=urn:nbn:se:liu:diva-160020.
Full textMa, Wenjing. "Semi-blind robust identification and model (in)validation." 2007. http://etda.libraries.psu.edu/theses/approved/WorldWideIndex/ETD-2342/index.html.
Full textYang, Shuowei Schuster Eugenio. "Tokamak resistive wall model validation and robust stabilization strategies." 2009. http://gateway.proquest.com/openurl?url_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:dissertation&res_dat=xri:pqdiss&rft_dat=xri:pqdiss:1469572.
Full textKAO, SHIH-YING, and 高世穎. "A robust analysis of the Cox model under the validation sampling." Thesis, 2014. http://ndltd.ncl.edu.tw/handle/18998964625444988616.
Full text國立臺北大學
統計學系
102
This work is to study the Cox proportional hazards model in cohort study under the validation sampling setting. We analyse the robust properties of regression parameter estimates in which the event times are assumed from the Cox model with individuals explanatory variables. However, only part of individuals are able to observe these variables, and the proxy covariates are available for the cohort study. Therefore, we consider two types of relationships between the covariates and its proxy variables, and the simulation illustrates the statistical properties of estimates with a full scale of senarios. The results provide a way to utilize the validation sample for the Cox model.
Canales, Ángel Giancarlo Miranda, and 白立文. "Advanced Model Validation and Robust Power Management of Hybrid Green Energy Systems." Thesis, 2014. http://ndltd.ncl.edu.tw/handle/40108489115367779912.
Full text國立清華大學
動力機械工程學系
102
Green energy power sources and storage devices are an ever growing market and research area for vehicles, portable and stationary applications. In green energy vehicles, the power may come from an exclusive main power source or combined with a secondary auxiliary supply. In this research, the main and secondary power source may be composed of a slow dynamics high power throughput energy source such as a fuel cell and a “free” energy source solar cell module. The storage device is chosen to be a lithium-ion battery pack, but may as well be a super-capacitor. This generality is the special characteristic of the proposed power management strategy. The parameter identification is performed by dynamic load conditions and experimental spectroscopy analysis techniques. For the dye sensitized solar cells, a new identification is performed by obtaining the dark-current impedance under different temperatures and then extending this model to the solar cell mode impedance for a reliable representation. The dynamic load method is employed to obtain a high power integrated battery model for the lithium-ion battery pack. This advanced parameter identification technique results in a lightweight but accurate model which can further enhance state of charge Kalman filter based estimators or other advanced techniques. A systematic analysis of an active topology power system with numerical simulations that take into account realistic experimental conditions will also be performed to accurately predict the steady-state effects of two types of power converters on the energy storage and energy source devices; fuel cells, solar cells, olivine batteries, etc. In order to achieve this, it is chosen to (i) employ the Euler-Lagrange (EL) framework to propose new non-ideal models comprising conduction losses (CL) and dead-time (DT) effects of two types of power converters. (ii) Demonstrate the vast improvement of the steady state error reduction for the derived inner current and outer voltage nonlinear passivity based controllers (PBCs) based on the CL+DT models with dynamic simulations. (iii) Provide analytic steady state operation solutions that result from the CL+DT models and their respective PBC controllers which can be employed to predict the current required from the energy source to aid in the design and sizing of hybrid green energy power systems. Without the new CL+DT PBC controllers, the output power is seen to be vastly reduced with respect to the demanded power even with dead-time periods as small as 0.1961 μs if the CL model and ideal model (IM) PBCs are employed. (iv) Modularly unify the system under passivity based nonlinear control laws for a seamless design of the power management system ensuring accurate inner current control tracking during operation. (v) Embed the ability to include power source protection instructions based on advanced model characterizations. With these initiatives, a convenient modularity to adapt to different hybrid power arrangements is proposed and thus, can easily be designed to protect each one of them. The proposed management strategy in this research could be applied to a green power air conditioner, whose load is composed of thermoelectric chips. However, because of the high computational demands of the power converters, the power system cannot be fully simulated alongside. In addition, the time scales of the cabin, DSSCs and LiBS are in the seconds scales, while the power converters time scales are in the microsencond range. Therefore, a novel idea to derive the steady state efficiency of the power converters using fundamental theory from the PBCs will be proposed. The relationship between the cabin dynamics and the environment, as well as the DSSCs and LiBs will be fully simulated with realistic dynamics. A feasibility study will be performed in order to determine scenarios where the solid-state air conditioner may be used and will include guidelines for its success. A MATLAB/Simulink environment will predict the efficiency and performance of the air conditioner in the cabin of a vehicle. Finally, by importing the efficiency relationships from fundamental theory, the electrical efficiency of the system will be designed for the first time.
CHEN, HUI-TING, and 陳惠婷. "Validation on Digital Twin of Six-Axis Robots with Modal Analysis." Thesis, 2019. http://ndltd.ncl.edu.tw/handle/jenebm.
Full text國立高雄科技大學
機械與自動化工程系
107
In this study, the goal of building a six-axis robot's Digital Twin is to achieve the off-line trajectory planning in machining operation by the ability of the digital avatar to synchronize with real state of the robot. In order to realize the equivalent model of the six-axis robot, the actual robot was subjected to impact test using Experimental Modal Analysis (EMA), after obtaining the frequency response function (FRF) by impact hammer and sensor, the modal analysis software (STAR 7) is used to perform curve fitting on FRF to obtain the natural frequency and mode shape. In finite element modal analysis, 3D model of the robot is analyzed by simulating boundary conditions of the actual fixed state. Finally, the results of two experiments are verified to confirm the correctness of model. This study found that boundary conditions have a significant impact on the results of modal analysis. For this reason, the experimental modal analysis conducted by six-axis robot from low-rigid aluminum extruded chassis to high-rigidity Vibration isolation table. In the finite element analysis, boundary conditions are set to fix the entire bottom surface and position of fixing screws and relax the joints for analysis, simulating the boundary condition setting close to the actual robot. In this way, by setting approach condition of boundary conditions, the difference between natural frequency values in the experimental and simulation results can be effectively improved.