Ce travail étudie l’applicabilité des méthodes d’identification aux paramètres de la suspension d’un modèle multi-corps du TGV. L’objectif est de caler le modèle par rapport au système réel par une estimation des paramètres de la suspension en utilisant la réponse du véhicule mesuré. A cause du comportement nonlinéaire du système la méthode temporelle de recalage a été choisie. Elle nécessite la définition et minimisation d’une fonction coût qui décrit la distance entre le modèle et la mesure. La convergence la plus rapide est obtenue avec des méthodes gradient qui demandent le calcul des dérivés de la fonction coût par rapport á chaque paramètre de la suspension. Comme le calcul par des différences finis est coûteux et moins précis la méthode adjoint a été implémentée. L’application à un modèle simplifié d’un bogie avec une description mathématique connue permet d’identifier les paramètres de la suspension. La présence des minima locaux dans la fonction coût du modèle TGV nécessite l’utilisation des méthodes globales. Le recuit simulé et l’algorithme génétique donnent des réductions importantes de la fonction coût et des estimations pour les paramètres du modèle TGV. Dans des travaux ultérieurs, ces informations pourraient être utilisées pour d’autre application comme le « condition monitoring ». / This work investigates the applicability of identification methods to the suspension parameters of a TGV multi-body model. The aim is to adjust the model to the real system by estimating the suspension parameters from measured vehicle response data. Due to the nonlinear behavior of the system the time-domain based model updating has been chosen. It requires the definition and minimization of a misfit function in the time domain describing the distance between model and measurement. The fastest convergence is obtained by the use of gradient methods requiring the calculation of the derivatives of the misfit function relative to every parameter. Since the calculation from finite differences is time consuming and less accurate the gradients are calculated from the adjoint method. The application to a simplified bogie model with known mathematical description allows the identification of its suspension parameters. The presence of local minima in the misfit function of the TGV model requires the use of global optimization methods. The simulated annealing and the genetic algorithm method give important reductions of the misfit function and improved parameter estimations. In following work this information could be used for further applications like the condition monitoring.

Two and three-wheeled vehicles are being used in increasing numbers in many emerging countries. The dynamics of such vehicles are very different from those of cars and other means of transportation. This thesis deals with a study of the dynamics of a motorcycle and an extensively used three-wheeled vehicle, called an “auto-rickshaw” in India. The commercially available multi-body dynamics (MBD) software, ADAMS, is used to model both the vehicles and simulations are performed to obtain insight into their dynamics. In the first part of the thesis, a study of the two wheeler dynamics is presented. A fairly detailed model of a light motorcycle with all the main sub-systems, such as the frame, front fork, shock absorbers , power train, brakes, front and rear wheel including tire slips and the rider is created in ADAMS-Motorcycle. The simulation results dealing with steering torques and angles for steady turns on a circular path are presented. From the simulation results and analytical models, it is shown that for path radius much greater than motorcycle wheel base, the steering torque and angle can be described by only two functions for each of the two variables. The first function is related to the lateral acceleration and can be determined numerically and the second function, in terms of the inverse of the path radius, is derived as an analytical approximation. Various tire and geometric parameters are varied in the ADAMS simulations and it is clearly shown that steady circular motion of a motorcycle can be reasonably approximated by only two curves–one for steering torque and one for steering angle. In the second part of the thesis, a stability analysis of the three-wheeled “autorickshaw” is presented. The steering instability is one of the major problems of the “auto-rickshaw” and this is studied using a MBD model created in ADAMS-CAR .In an Initial model the frame ,steering column and rear-forks (trailing arms) are assumed to be rigid. A linear eigenvalue analysis, at different speeds, reveals a predominantly steering oscillation, called a “wobble” mode, with a frequency in the range of 5 to 6Hz. The analysis results show that the damping of this mode is small but positive up to the maximum speed(14m/s) of the three-wheeled vehicle. Experiments performed on the three-wheeled vehicle show that the mode is unstable at speeds below 8.33m/s and thus the experimental results do not agree with the model. Next, this wobble instability is studied with an analytical model, similar to the model proposed for wheel shimmy problem in aircrafts. The results of this model show that the wobble is stable at low speeds regardless of the magnitude of torsional stiffness of steering column. This is also not matching with the experimental result. A more refined MBD model with flexibility incorporated in the frame, steering column and the trailing arm is constructed. Simulation results with the refined model show three modes of steering oscillations. Two of these are found to be well damped and the third is found to be lightly damped with negative damping at low speeds, and the results of the model with the flexibility is shown to be matching reasonably well with the experimental results. Detailed simulations with flexibility of each body incorporated, one at a time, show that the flexibility in the steering column is the main contributor of the steering instability and the instability is similar to the wheel shimmy problem in aircrafts. Finally, studies of modal interaction on steering instabilities and parametric studies with payload and trail are presented.

The main purpose of this diploma thesis is to design turbocharger rotor using rolling element bearings. The diploma thesis is compiled from two specialized search parts dealing with turbocharger rotor and rolling element bearing computation, respectively. The application of the particular rotor using the multi body system Adams – View is described in the following chapter. In the last chapter, one can find the comparison between rotor using bearing with steal and hybrid ceramic rolling element. Also, there is compariosn between rolling element bearing and journal bearing. In the maximum rotor speed, the decrease of more than 50 % in power loss, due to usage of rolling element bearing, resulted from this comparison.

This thesis, within a long-term project at the Institute of automotive engineering at FME BUT, aspires to prepare a motorcycle for analysis of its dynamic riding behavior. A multi-body model of a motorcycle is created based on the measured parameters and a system for measuring driving conditions is designed. The necessary motorcycle parameters were measured using a 3D scanner and through experiments aided by computer technology and CAD software. This laid the foundation for the follow-up research and development work.

This Master’s thesis deals with designing a measuring device that would record driver’s movements during dynamic riding states known as localization, and based on the recorded data creating a simulation in multi-body software. The first part reviews the current state of knowledge in the field of photogrammetry, sensors, and motorcycles’ characteristics. Moreover, the second part focuses on constructing the measuring device, practical aspects of data collection, processing, and evaluation in multi-body software.

The master’s thesis deals with building mechatronics model of machine tool head delivered by company TOSHULIN,a.s. for purpose of damping vibration. Mechanical part of the model is based on modeling of system of flexible bodies. Components are modeled in FEM program ANSYS and subsequently reduced using Craig-Bampton method. The system of flexible bo-dies is created in MBS program ADAMS. Model is linearized and subsequently reduced. As a element of active damping is used linearized model of magnetic bearing. For control is desig-ned PID controller. The comparison of damped and undamped model is conducted in the end.

The thesis solves vibration of simple combustion engine model as a basic step for evaluating the noise of engine. On the basis of computational model realized in software MBS ADAMS, which contains all significant input parameters, there are found normal velocities of the one cylinder four stroke engine block surfaces. The outputs are then processed into the integral characteristics that define the level of possible emissions for whole acoustic spectre. Two variants of the engine block were solved and in the end the summary of solution approach was made.

While human presence in cislunar space continues to expand, so too does the demand for `lightweight' automated on-board processes. In nonlinear dynamical environments, computationally efficient guidance strategies are challenging. Many traditional approaches rely on either simplifying assumptions in the dynamical model or on abundant computational resources. This research employs reinforcement learning, a subset of machine learning, to produce a controller that is suitable for on-board low-thrust guidance in challenging dynamical regions of space. The proposed controller functions without knowledge of the simplifications and assumptions of the dynamical model, and direct interaction with the nonlinear equations of motion creates a flexible learning scheme that is not limited to a single force model. The learning process leverages high-performance computing to train a closed-loop neural network controller. This controller may be employed on-board, and autonomously generates low-thrust control profiles in real-time without imposing a heavy workload on a flight computer. Control feasibility is demonstrated through sample transfers between Lyapunov orbits in the Earth-Moon system. The sample low-thrust controller exhibits remarkable robustness to perturbations and generalizes effectively to nearby motion. Effective guidance in sample scenarios suggests extendibility of the learning framework to higher-fidelity domains.

<div>Indiana University–Purdue University Indianapolis student team Jaguar has been participating in the electric Formula SAE (FSAE) vehicle competitions in the past few years. There is an urgent need to develop a design tool for improving the performance of the vehicle. In this thesis, multibody dynamics (MBD) models have been developed which allow the student team to improve their vehicle design, while reducing the required time and actual testing costs. Although there were some studies about MBD analyses for vehicles in literature, a detailed modeling study of key parameters is still missing. Specifically, the effect of suspension system on the vehicle performance is not well studied. </div><div>The objective of the thesis is to develop an MBD based model to improve the FSAE vehicle’s performance. Based on the objective and knowledge gap, the following research tasks are proposed: (1) MBD modeling of current suspension systems; (2) Modification of suspension systems, and (3) Evaluation of performance of modified suspension systems. </div><div>The models for the front suspension system, rear suspension system, and full assembly are created, and a series of MBD analyses are conducted. The parameters of the vehicle by conducting virtual tests on the suspension model and overall vehicle model are studied. In this work, two main virtual tests are performed. First, parallel wheel travel test on suspension system, in which the individual suspension system is subject to equal force on both sides. The test helps understand the variation in stability parameters, such as camber angle, toe angle, motion ratio, and roll center location. Second, skid-pad test on full assembly of the vehicle. The test assists in understanding the vehicle’s behavior in constant radius cornering and the tire side slip angle variation, as it is one of the important parameters controlling alignment of the vehicle in this test.</div><div>Based on the vehicle’s dynamics knowledge obtained from the existing vehicle, a modified version of the FSAE vehicle is proposed, which can provide a better cornering performance with minimum upgrades and cost possible. Based on the results from the parallel wheel travel test and skid-pad test, the lateral load transfer method is used to control the vehicle slip, by making changes to the geometry of the vehicle and obtaining appropriate roll center height for both front and rear suspension system. The results show that the stiffness in front suspension system and rear suspension system are controlled by manipulating roll center height. This study has provided insightful understanding of the parameters and forces involved in suspension system and their variations in different events influencing vehicle stability. Moreover, the MBD approach developed in this work can be readily extended to other commercial vehicles and sports vehicles.</div><div><br></div>

A mechanical model of an axial piston-type machine with a so-called wobble plate and Z-shaft mechanism is presented. The overall aim is to design and construct an oil-free piston expander demonstrator as a first step to realizing an advanced and compact small-scale steam engine system. The benefits of a small steam engine are negligible NOx emissions (due to continuous, low-temperature combustion), no gearbox needed, fuel flexibility (e.g., can run on biofuel and solar), high part-load efficiency, and low noise. Piston expanders, compared with turbines or clearance-sealed rotary displacement machines, have higher mechanical losses but lower leakage losses, much better part-load efficiency, and for many applications a more favourable (i.e., lower) speed. A piston expander is thus feasible for directly propelling small systems in the vehicular power range. An axial piston machine with minimized contact pressures and sliding velocities, and with properly selected construction materials for steam/water lubrication, should enable completely oil-free operation. An oil-free piston machine also has potential for other applications, for example, as a refrigerant (e.g., CO2) expander in a low-temperature Rankine cycle or as a refrigerant compressor.   An analytical rigid-body kinematics and inverse dynamics model of the machine is presented. The kinematical analysis generates the resulting motion of the integral parts of the machine, fully parameterized. Inverse dynamics is applied when the system motion is completely known, and the method yields required external and internal forces and torques. The analytical model made use of the “Sophia” plug-in developed by Lesser for the simple derivation of rotational matrices relating different coordinate systems and for vector differentiation. Numerical solutions were computed in MATLAB. The results indicate a large load bearing in the conical contact surface between the mechanism’s wobble plate and engine block. The lateral force between piston and cylinder is small compared with that of a comparable machine with a conventional crank mechanism.   This study aims to predict contact loads and sliding velocities in the component interfaces. Such data are needed for bearing and component dimensioning and for selecting materials and coatings. Predicted contact loads together with contact geometries can also be used as input for tribological rig testing. Results from the model have been used to dimension the integral parts, bearings and materials of a physical demonstrator of the super-critical steam expander application as well as in component design and concept studies.