Measuring the thrust of cold gas micro-thrusters, due to the small magnitude of the generated force, is a challenging task, and to obtain satisfying quality of results usually careful adjustments of the setup has to be performed. The work presented in this paper aims at improving the quality of such measurements at the GomSpace Sweden company. The work consists of two parts: In the first part the author focuses on improving current thrust measurements setup implemented in the company, which consists of a vacuum compatible laboratory scale. It was observed that the results obtained from this setup were approximately 25‒35% lower than expected. A test campaign performed within the work found that this was associated mainly with backflow of plume expansion gases. Covering the scale and the tested propulsion unit with a box having openings only for thrusters was proved to be sufficient setup upgrade to mitigate the issue. In the second part of the project an alternative setup utilising a hanging pendulum-based solution with strain measurement-based force sensor is proposed and its design is thoroughly described. Results of initial tests of a 3D printed prototype of this setup are presented, proving its potential to produce results with better quality of output than the laboratory scale. / Att mäta dragkraften från mikrodysor baserade på kallgas är på grund av den låga genererade kraften en utmanande uppgift. För att få tillfredsställande kvalitet på resultatet måste vanligtvis noggranna justeringar av inställningarna utföras. Arbetet som presenteras i denna rapport syftar till att förbättra kvaliteten på sådana mätningar vid GomSpace Sweden AB. Arbetet består av två delar: I den första delen fokuserar författaren på att förbättra den nuvarande uppställningen för mätning av dragkraft vid GomSpace, bestående av en vakuumkompatibel laboratorievåg. Det observerades att resultaten från denna mätupställning var cirka 25‒35% lägre än förväntat. En testkampanj som utfördes inom ramen för detta arbete visade att detta främst var förknippat med ett tillbakaflöde av gaser från den expanderande bränsleplymen. Att täcka över vågen och den testade framdrivningsenheten med en låda med öppningar endast för dysan visade sig vara tillräckligt för att minska problemet. I den andra delen föreslås ett alternativt upplägg baserat på en hängande pendel med en töjningsmätningsbaserad kraftsensor, dess utformming beskrivs utförligt. Resultaten av de första testerna av en 3D-utskriven prototyp av denna uppställning presenteras, vilket visar på dess potential att producera mätningar med bättre kvalitet än laboratorievågen.

No description available.

In this paper the two reinforcement learning algorithmsQ-learning and deep Q-learning (DQN) are used tobalance an inverted pendulum. In order to compare the two, bothalgorithms are optimized to some extent, by evaluating differentvalues for some parameters of the algorithms. Since the differencebetween Q-learning and DQN is a deep neural network (DNN),some benefits of a DNN are then discussed.The conclusion is that this particular problem is simple enoughfor the Q-learning algorithm to work well and is preferable,even though the DQN algorithm solves the problem in fewerepisodes. This is due to the stability of the Q-learning algorithmand because more time is required to find a suitable DNN andevaluate appropriate parameters for the DQN algorithm, than tofind the proper parameters for the Q-learning algorithm. / I denna rapport används två algoritmer inom förstärkningsinlärning och djup Q-inlärning (DQN), för att balancera en omvänd pendel. För att jämföra dem så optimeras algoritmerna i viss utsträckning genom att testa olika värden för vissa av deras parametrar. Eftersom att skillnaden mellan Q-inlärning och DQN är ett djupt neuralt nätverk (DNN) så diskuterades fördelen med ett DNN. Slutstatsen är att för ett så pass enkelt problem så fungerar Q-inlärningsalgoritmen bra och är att föredra, trots att DQNalgoritmen löser problemet på färre episoder. Detta är pågrund av Q-inlärningsalgoritmens stabilitet och att mer tid krävs för att hitta ett passande DNN och hitta lämpliga parametrar för DQN-algoritmen än vad det krävs för att hitta bra parametrar för Q-inlärningsalgoritmen. / Kandidatexjobb i elektroteknik 2021, KTH, Stockholm

Sliding isolation bearings are used widely around the world to minimize damage to structures and their contents during earthquakes. Past studies have typically focused on the behavior of sliding isolation bearing under design conditions; however, as the performance-based earthquake engineering advances, it is necessary and critical to understand the ultimate or even failure behavior, of structural systems under extreme conditions. Using a double friction pendulum bearing with non-articulated slider as an example, this thesis comprehensively investigates the extreme behavior of the sliding bearing components as well as steel frame buildings isolated using these bearings. This thesis is comprised of two major parts. The first includes numerical and experimental studies of double friction pendulum bearings at the component-level. Finite element investigation shows that depending on the superstructure mass there are two major failure modes for the double friction pendulum bearings. When the superstructure mass is sufficiently large, the failure mode is dominated by the restraining rim yielding; however, when the mass is relatively small, its failure mode shifts to bearing uplift. A simplified analytical model which can directly simulate the impact and uplift behavior of double friction pendulum bearing is also implemented, comparing well to the finite element analysis. Then, to validate the ability of the models to predict extreme behavior as well as to investigate the effect of the restraining rim design, which varies around the world, an experimental study was carried out. Uplift behavior and significant rim yielding were observed during the shake table tests. Moreover, other response parameters, including uplift and shear forces, are evaluated and compared among different rim designs. It is found the restraining rim design has a substantial influence on the bearing’s extreme behavior. The second part of the thesis investigates the system-level behavior of steel frame buildings isolated with double friction pendulum bearings. It is found that the stiffness of the superstructure largely dictates the system-level failure modes and collapse probability. Initially, bearings with rigid restraining rims are investigated. For flexible moment-resisting frames, the system-level failure modes are mixed: both the bearing uplift and superstructure yielding contribute; also, using current code-minimum design results in acceptably low probability of collapse. However, for stiff concentrically-braced frames, the impact force can impose large ductility demands on the superstructure regardless of its strength. As a result, the system-level failure comes exclusively from superstructure yielding, and only by increasing bearing’s displacement capacity beyond the minimum code allowed can the design meet as acceptably low collapse probability. When flat rims are used instead for the bearing design, the failure modes for both building types are exclusively bearing failure. Furthermore, while it is more apparent for concentrically-braced frames, using flat rims for the bearings can reduce the collapse probability compared to using rigid rims. / Thesis / Doctor of Philosophy (PhD)

Aim of this thesis is building of a simulator model of a rotary (Furuta) pendulum and design of appropriate regulators. This paper describes assembly of a nonlinear simulator model, using Matlab–Simulink and its library Simscape–Simmechanics. Furthermore the paper discuss linear discrete model obtained from the system response, using least squares method. This linear model serves as aproximation of the system for designing of two linear discrete state space regulators with sumator. These regulators are supported by a simple swing–up regulator and logics managing cooperation.

Galilée (1564-1642) a introduit dans son Il Saggiatore (1623) une nouvelle conception du corps. L’ontologie et la théorie de la connaissance des corps de Spinoza peuvent être conçues comme une réponse à Galilée. En effet, le philosophe hollandais répète de nombreuses fois que les qualités sensibles n’appartiennent pas aux corps en soi. En outre, il précise que les idées des affections sont des idées inadéquates qui représentent plutôt le corps affecté que les corps affectants. Néanmoins, Spinoza (1632-1677) donne une interprétation tout à fait particulière de cette conception galiléenne. Comme il le précise dans sa définition de l’Abrégé de physique, un corps en soi est un ensemble de parties qui sont unies par un rapport mutuel de mouvement et de repos. Ce ratio est à concevoir comme une nature, une proportion ou une loi physique d’un corps. Par sa nouvelle conception du corps, Galilée a dépassé la distinction ontologique entre les corps artificiels et les corps naturels, ouvrant la voie à l’application des modèles et des analogies. A première vue, Spinoza n’applique pas le modèle du pendule. Néanmoins, une étude plus détaillée dévoile l’importance de la physique de l’horloge pendulaire (inventée par Galilée et perfectionnée par Chr. Huygens) dans la conception spinoziste du corps. Dans sa définition du corps de l’E1, Spinoza conçoit le corps dans sa relation avec l’essence de Dieu. Néanmoins, dans sa CM, il a introduit le conatus ou l’essence d’une chose en donnant le paradigme d’un corps en mouvement. Comme Galilée l’avait démontré, le mouvement est essentiellement une force. Spinoza a généralisé cette idée de force, comme il l’a fait avec l’idée de loi de la nature. / In Il Saggiatore (1623), Galileo (1564-1642) introduced a novel conception of the body. Spinoza’s ontology and epistemology can be viewed as original responses to this. Indeed, the author of the Ethics writes repeatedly that sensible qualities do not belong to the body as such. Moreover, he clearly states that the ideas of affections are inadequate, representing much less the intrinsic properties of external bodies than the affected body itself. However, Spinoza (1632-1677) gives a very particular interpretation of the Galilean conception. As he makes clear in his Physical Interlude, a body consists in a group of parts united by a mutual relation of motion and rest. Furthermore, this relation is to conceived as a nature, proportion or physical law of the body. By means of his new conception, Galileo radically upended the ontological distinction between artificial and natural bodies, which opened the door to the application of models and analogies for the explanation of natural phenomena. At a first glance, Spinoza does not apply the model of the pendulum clock, which was a leading model of the 17th century. However, a closer look reveals the importance of the physics of the pendulum clock (invented by Galileo and perfected by Christian Huygens) for Spinoza’s conception of the body. In his definition in E1, Spinoza conceives of the body in its relation to the essence of God. In his CM, however, he had introduced the conatus, or the essence of a thing, in terms of the paradigm of the body in motion. As Galileo had shown, motion is essentially a force. Spinoza generalized this notion of force, just as he generalized the idea of the law of nature.

Nesta dissertação, um tipo diferente de pêndulo invertido controlado por rodas de reação é apresentado. Sua principal diferença está em seu ponto de articulação, que é constituído por uma junta esférica que permite com que o pêndulo gire em torno de seus três eixos. Além disso, três rodas de reação são utilizadas para seu controle e estabilização. Primeiramente, um modelo do sistema é obtido a partir das equação de Euler-Lagrange, das leis de Newton e das leis de Kirchhoff. Em seguida, uma lei de controle que assegura a estabilização assintótica do sistema em um grande domínio é proposta. Por fim, simulações são realizadas para validar o controlador projetado. Esse sistema possui diversas características interessantes, tanto do ponto de vista teórico como do ponto de vista de pesquisa. Do ponto de vista teórico, o sistema é nãolinear e suas entradas são fortemente acopladas, o que torna particularmente adequado para o processo de projeto e implementação de diversas técnicas de estabilização. Do ponto de vista de pesquisa, são consideradas duas técnicas de controle não linear: linearização padrão e linearização exata. Para que o sistema seja robusto e não desperdice energia, essas duas leis de controle diferentes são comutadas para a obtencão de um número suficiente de domínio de estabilidade. / In this dissertation, a different kind of the reaction wheel pendulum is presented. The main difference is that its articulation point consists of a ball joint that allows the pendulum to rotate around its three axes. Furthermore, three reaction wheels are used for its control and stabilization. First, a model of the system is obtained from Euler-Lagrange equations, Newton laws and Kirchhoff laws. After that, a control law that assure asymptotic stabilization of the system in a large domain is proposed. Finally, simulations are performed to validate the designed controller. This system has several interesting features, both from a theoretical standpoint as from a research standpoint. From a theoretical standpoint, the system is nonlinear and its inputs are tightly coupled, making it particularly suitable for the design and implementation process of various stabilization techniques. From a research standpoint, two non-linear control techniques are considered: standard linearization and exact linearization. For the system to be robust and do not waste energy, these two different control laws are switched for obtaining a sufficiently large domain of stability.

O estudo da estabilidade e desempenho em sistemas de controle é um tópico relevante na teoria de sistemas. Quando são assumidas incertezas no modelo da planta, existe uma maior dificuldade para garantir um nível de desempenho adequado do sistema dinâmico e a estabilidade pode ser comprometida. Neste trabalho são utilizados um regulador linear quadrático robusto e um filtro de Kalman robusto combinados em uma única formulação para tratar de sistemas dinâmicos incertos em tempo real. O caso de estudo selecionado é o pêndulo invertido. Seus principais desafios de controle encontrados na literatura: estabilização, seguimento e levantamento-captura, serão considerados. Os algoritmos utilizados são motivados pelo fato de que problemas estocásticos podem ser resolvidos por meio de argumentos determinísticos, baseados nos conceitos de função penalidade e mínimos quadrados regularizados. Desta forma, é possível a obtenção do melhor desempenho em contrapartida à máxima influência de incerteza admissível. A análise de desempenho do controlador robusto é realizada por meio de ensaios práticos incluindo incertezas na planta, ruído nos sensores e distúrbios no sinal de controle do pêndulo. / The study of stability and performance in control systems is a relevant topic in systems theory. When uncertainties are considered in the model of the plant, there is a greater difficulty in ensuring an appropriate performance level of the dynamic system, plus, the stability could be compromised as well. In this dissertation a robust linear quadratic regulator and a robust Kalman filter are used in a unified manner to deal with uncertain dynamic systems in real time. The selected case study is the inverted pendulum. Its main control challenges found in the literature will be considered: stabilization, tracking and catching swing-up. The used algorithms are motivated by the fact that stochastic problems can be solved through deterministic arguments based on the concepts of penalty function and regularized least-squares. Thus, it is possible to obtain an optimal performance for the maximum acceptable uncertainty. The performance analysis of the robust control is carried out by practical experiments including uncertainties in the plant, noise in the sensors and disturbance in the pendulum control signal.

This thesis introduces a new measure of balance for bipedal robotics called the foot placement estimator (FPE). To develop this measure, stability first is defined for a simple biped. A proof of the stability of a simple biped in a controls sense is shown to exist using classical methods for nonlinear systems. With the addition of a contact model, an analytical solution is provided to define the bounds of the region of stability. This provides the basis for the FPE which estimates where the biped must step in order to be stable. By using the FPE in combination with a state machine, complete gait cycles are created without any precalculated trajectories. This includes gait initiation and termination. The bipedal model is then advanced to include more realistic mechanical and environmental models and the FPE approach is verified in a dynamic simulation. From these results, a 5-link, point-foot robot is designed and constructed to provide the final validation that the FPE can be used to provide closed-loop gait control. In addition, this approach is shown to demonstrate significant robustness to external disturbances. Finally, the FPE is shown in experimental results to be an unprecedented estimate of where humans place their feet for walking and jumping, and for stepping in response to an external disturbance.

Current seismic design practice promotes inelastic response in order to reduce the design forces. By allowing the structure to yield while increasing the ductility of the structure, the global forces can be kept within the limited bounds dictated by the yield strength. However, during severe earthquakes, the structures undergo significant inelastic deformations leading to stiffness and strength degradation, increased interstory drifts, and damage with residual drift. The research presented in this thesis has three components that seek to address these challenges. To prevent the inelastic effects observed in yielding systems, a new concept “apparent weakening” is proposed and verified through shake table studies in this thesis. “Apparent weakening” is introduced in the structural system using a complementary “adaptive negative stiffness device” (NSD) that mimics "yielding” of the global system thus attracting it away from the main structural system. Unlike the concept of weakening and damping, where the main structural system strength is reduced, the new system does not alter the original structural system, but produces effects compatible with an early yielding. Response reduction using NSD is achieved in a two step sequence. First the NSD, which is capable of exhibiting nonlinear elastic stiffness, is developed based on the properties of the structure. This NSD is added to the structure resulting in reduction of the stiffness of the structure and NSD assembly or “apparent weakening”-thereby resulting in the reduction of the base shear of the assembly. Then a passive damper, designed for the assembly to reduce the displacements that are caused due to the “apparent weakening”, is added to the structure-thereby reducing the base shear, acceleration and displacement in a two step process. The primary focus of this thesis is to analyze and experimentally verify the response reduction attributes of NSD in (a) elastic structural systems (b) yielding systems and (3) multistory structures. Experimental studies on 1:3 scale three-story frame structure have confirmed that consistent reductions in displacements, accelerations and base shear can be achieved in an elastic structure and bilinear inelastic structure by adding the NSD and viscous fluid damper. It has also been demonstrated that the stiffening in NSD will prevent the structure from collapsing. Analogous to the inelastic design, the acceleration and base shear and deformation of the structure and NSD assembly can be reduced by more than 20% for moderate ground motions and the collapse of structure can be prevented for severe ground motions. Simulation studies have been carried on an inelastic multistoried shear building to demonstrate the effectiveness of placing NSDs and dampers at multiple locations along the height of the building; referred to as “distributed isolation”. The results reported in this study have demonstrated that by placing a NSD in a particular story the superstructure above that story can be isolated from the effects of ground motion. Since the NSDs in the bottom floors will undergo large deformations, a generalized scheme to incorporate NSDs with different force deformation behavior in each storey is proposed. The properties of NSD are varied to minimize the localized inter-story deformation and distribute it evenly along the height of the building. Additionally, two semi-active approaches have also been proposed to improve the performance of NSD in yielding structures and also adapt to varying structure properties in real time. The second component of this thesis deals with development of a novel device to control the response of structural system using adaptive length pendulum smart tuned mass damper (ALP-STMD). A mechanism to achieve the variable pendulum length is developed using shape memory alloy wire actuator. ALP-STMD acts as a vibration absorber and since the length is tuned to match the instantaneous frequency, using a STFT algorithm, all the vibrations pertaining to the dominant frequency are absorbed. ALP-STMD is capable of absorbing all the energy pertaining to the tuned-frequency of the system; the performance is experimentally verified for forced vibration (stationary and non-stationary) and free vibration. The third component of this thesis covers the development of an adaptive control algorithm to compensate hysteresis in hysteretic systems. Hysteretic system with variable stiffness hysteresis is represented as a quasi-linear parameter varying (LPV) system and a gain scheduled controller is designed for the quasi-LPV system using linear matrix inequalities approach. Designed controller is scheduled based on two parameters: linear time-varying stiffness (slow varying parameter) and the stiffness of friction hysteresis (fast varying parameter). The effectiveness of the proposed controller is demonstrated through numerical studies by comparing the proposed controller with fixed robust H∞ controller. Superior tracking performance of the LPV-GS over the robust H∞ controller in different displacement ranges and various stiffness switching cases is clearly evident from the results presented in this thesis. The LPV-GS controller is capable of adapting to the parameter changes and is effective over the entire range of parameter variations.