Companies within the manufacturing industry are often aiming to increase productivity and simultaneously maintain the quality of their products. To achieve higher productivity and quality it is imperative to have tools with a high speed and accuracy. Atlas Copco BLM's STbench is a measurement system which enables manufacturers to validate their tools during different situations. The purpose of this case study was to improve the robustness of the STbench so it would operate well during situations with both low and high tool speed. To define and investigate how to improve the STbench, a modified DMAIC-approach was used. During the investigation it was found that the area with the largest improvement possibilities was the STbench's PID-controllers. Design of Experiments was used as the method to optimize the P- and I-element of the PID-controllers; hence, increase the robustness. The optimal settings could improve the robustness of the STbench with approximately 50%, but the result has not been verified. This case study presents results that can increase the robustness of the STbench; thus answering the purpose. Furthermore, this master thesis presents several revelations regarding using experimental plans while optimizing control systems, an area that has not been extensively investigated in previous literature.

In this thesis, the design of a self balancing prototype bike with a reaction wheel as balancing mechanism will be investigated. Bikes are inherently unstable systems, that dynamically resemble an inverted pendulum at a stand still. Previous work has mostly been focused on keeping the balance through manipulating the steering input, and using state space modelling in order to balance the bike. In this project, a PID controller was implemented as a mean of control for the reaction wheel which was driven by a brushless DC motor. Different variations of Ziegler Nichols tuning methods were compared, and an alternative heuristic tuning method for the PID controller was also tested. The standard Ziegler Nichols and the alternative method yielded the best results. / I detta kandidatexamensarbete kommer designen av en självbalanserande prototyp cykel med ett reaktionshjul som balanseringsmekanism att undersökas. Cyklar är instabila system, som dynamiskt liknar en inverterad pendel när de är stationära. Tidigare arbeten har huvudsakligen fokuserat på att hålla balansen genom att elektroniskt manipulera styrinmatningen och använda modern regelteknik för att balansera cykeln. I det här projektet implementerades en PID-regulator som ett medel för att styra reaktionshjulet, driven av en borstlös likströmsmotor. Olika varianter av Ziegler Nichols-metoden och den alternativa metoden gav de bästa resultaten.

Autonomous mobile robots are on the rise and are to be expected on the market in about 5-10 years. Several challenges need to be solved for this to happen, and the most crucial ones are to develop versatile and safe robots. The Get a Grip robot is a dynamic force adjustment gripper using inputs from two different sensory systems. The construction of the robot consists of two parallel gripper plates moved by a rack and pinion gear attached to a direct current (DC) motor. Embedded into one of the plates is a Force Sensitive Resistor (FSR) for input of the gripper’s exerted force. Mounted to the other plate is a self constructed Slip sensor used for measuring the occurrence of slip and slip rate. A surrounding crane for mounting of the gripper and lifting was also constructed. The idea of this bachelor’s thesis project is to enable lifting of objects with unknown weight without the gripper exerting more force than necessary. This is something that will be useful in both industrial applications and in household robots in the future. In order to realize the concept two different methods for calculating the gripper’s applied force were tested, one using motor current and the other using a FSR sensor. Through testing it was concluded that the FSR sensor was the method giving better accuracy and consistency. Proportional–Integral–Derivative (PID) controllers were then tested for both setting force references for the gripper using the Slip sensor as input, and controlling the exerted force in the gripper using the FSR as input. The results led to two PID controllers thought to be sufficient as starting points for further testing of the complete system. / Mobila autonoma robotar förväntas vara på marknaden inom de närmaste 5-10 åren. För att det här ska ske är det många utmaningar som behöver lösas och de mest kritiska är att utveckla mångsidiga och säkra robotar. Get a Grip-roboten är en dynamisk kraftanpassande robotklo som tar insignaler från två olika sensorsystem. Konstruktionen består av två parallella plattor som förflyttas av kuggstänger och kugghjul drivna av en DC motor. Inbyggt i en av kloplattorna finns en tryckkänslig kraftsensor (FSR) monterad för att registrera kraften som klon genererar. På den andra kloplattan sitter en egenkonstruerad glidsensor som registrerar om glidning sker och själva glidhastighet. En kran för att montera klon och lyfta den konstruerades även. Idén bakom detta kandidatexamens projektet är att klon ska kunna lyfta ett objekt med okänd vikt utan att använda mer kraft än nödvändigt. Det är något som kommer vara användbart både vid industriella tillämpningar och hos husållsrobotar i framtiden. För att realisera konceptet testades två olika metoder för att estimera kraften klon genererar, den första genom motorströmmen och den andra genom en FSR sensor. Tester genomfördes för båda metoderna och slutsatsen blev att FSR sensorn gav bäst noggrannhet och var mest konsekvent. PID-regulatorn, för bestämning av kraftreferens, med insignal från glidsensorn och PID-regulatorn, för genererad klokraft, med insignal från FSR:n testades separat. Resultatet blev två PID-regulatorer som ansågs tillräckliga för fortsätta tester med båda regulatorerna tillsammans.

Yes / This paper presents a new control strategy for stochastic distribution shape tracking regarding non-Gaussian stochastic non-linear systems. The objective can be summarised as adjusting the probability density function (PDF) of the system output to any given desired distribution. In order to achieve this objective, the system output PDF has first been formulated analytically, which is time-variant. Then, the PDF vectorisation has been implemented to simplify the model description. Using the vector-based representation, the system identification and control design have been performed to achieve the PDF tracking. In practice, the PDF evolution is difficult to implement in real-time, thus a data-driven extension has also been discussed in this paper, where the vector-based model can be obtained using kernel density estimation (KDE) with the real-time data. Furthermore, the stability of the presented control design has been analysed, which is validated by a numerical example. As an extension, the multi-output stochastic systems have also been discussed for joint PDF tracking using the proposed algorithm, and the perspectives of advanced controller have been discussed. The main contribution of this paper is to propose: (1) a new sampling-based PDF transformation to reduce the modelling complexity, (2) a data-driven approach for online implementation without model pre-training, and (3) a feasible framework to integrate the existing control methods. / This paper is partly supported by National Science Foundation of China under Grants (61603262 and 62073226), Liaoning Province Natural Science Joint Foundation in Key Areas (2019- KF-03-08), Natural Science Foundation of Liaoning Province (20180550418), Liaoning BaiQianWan Talents Program, i5 Intelligent Manufacturing Institute Fund of Shenyang Institute of Technology (i5201701), Central Government Guides Local Science and Technology Development Funds of Liaoning Province (2021JH6/10500137).

Šio darbo tikslas – suprojektuoti ir išanalizuoti skysčio lygio valdymo sistemas su PI reguliatoriumi bei neraiškiosios logikos reguliatoriumi, atlikti imitacijas ir palyginti gautus rezultatus. Teorinėje dalyje apžvelgiamos skysčio lygio valdymo sistemos, jų pagrindiniai elementai. Taip pat išnagrinėjami sistemų valdymo ypatumai, naudojant neraiškiosios logikos ir PI reguliatorius. Tiriamojoje dalyje sudaryta: sistemos funkcinė schema, sistemos matematinis ir kompiuterinis modeliai. Atlikus procesų imitaciją nustatyta, kad sistemoje su PI reguliatoriumi pakeitus nuostatą dėl perreguliavimo gaunamos nepageidaujamos slėgio aukščio ir debito vertės. To išvengiama, naudojant neraiškiosios logikos reguliatorių. Darbo pabaigoje pateikiamos išvados. Darbą sudaro 7 dalys: įvadas, literatūros apie skysčio lygio valdymo sistemas analizė, tyrimo tikslas ir uždaviniai, teorinė dalis, tiriamoji dalis, išvados, literatūra. Darbo apimtis - 71 puslapiai be grafinės dalies, 53 iliustracijos, 6 lentelės, 40 bibliografinių šaltinių ir 4 internetinės prieigos. / The main goal of the thesis is to design and analyze liquid level control systems with a PI controller and a fuzzy logic controller, perform simulations and to compare the results. In the analytical part, liquid level control systems and main elements of liquid level control systems are overviewed. Also system control using PI and Fuzzy logic regulators is overviewed. In research part liquid level control system is projected practically. The research chapter consists of a functional scheme, mathematical and computer models design. The simulation of the system shows that changing the step size in the system with a PI controller results in unwanted total head and flow rate values because of an overshoot. This is avoided in the system with a fuzzy controller. At the end of the thesis there is a conclusions part. The work consists of seven parts: introduction, analysis of liquid level control system literature, main goal of the research, theory part, design part, conclusions, references. Work volume – 71 pages of text without appendixes, 53 illustrations, 6 tables, 40 bibliographical and 4 internet entries.

Nos últimos anos, tem havido grande interesse em estudar manipuladores paralelos, aplicados principalmente em simuladores de voo, com seis graus de liberdade. O interesse em estruturas cinemáticas paralelas é motivado por sua alta rigidez e excelente capacidade de posicionamento em relação às estruturas cinemáticas seriais. Além disso, como os atuadores são posicionados em uma base, eles podem ser aplicados em cargas pesadas e ainda apresentam baixo consumo de energia, tendo em vista que vários atuadores atuam simultaneamente no mesmo corpo. A presente tese apresenta o projeto de três controladores, sendo eles, o controlador H infinito com realimentação de saída, o controlador PID e o controlador Fuzzy, com isto, esta metodologia poderá ser empregada na construção de um futuro simulador de voo. O modelo dos atuadores foi obtido através de uma entrada degrau de tensão nos motores, medindo os seus deslocamentos através dos encoders acoplados, individualmente, a cada um dos respectivos eixos dos motores. Sabendo-se a relação de transmissão do mecanismo de movimento entre o motor e cada haste dos atuadores obtém-se o deslocamento de cada haste a partir da rotação de cada motor medida pelo correspondente encoder e com isso obtém-se o modelo matemático de cada atuador em conjunto com seu sistema de transmissão. Entretanto, na prática, cada atuador é ligeiramente diferente dos outros, o que leva a comportamento e desempenho diferentes entre si. Isso afeta o comportamento da plataforma fazendo com que a trajetória final desejada não possa ser seguida adequadamente, algo que é extremamente necessário em simuladores de voo. Assim, uma das contribuições importantes deste trabalho é, em primeiro lugar, apresentar uma metodologia de padronização das respostas dos atuadores de modo a que todos eles tenham no final, um comportamento igual o mais próximo possível, particularmente em termos de velocidade e de posicionamento. Com os dados da cinemática e da dinâmica da plataforma compondo o modelo completo do sistema foram realizadas várias simulações que aplicadas na plataforma de Stewart real validaram o modelo e mostraram a eficiência das técnicas de controle aplicadas no controle de posição e orientação da plataforma. Para validar o projeto da Plataforma de Stewart como uma possível base de movimento de um simulador de voo, foi implementada a dinâmica longitudinal e lateral de um Boeing 747-100, e com o auxilio de um sensor inercial Xsens® MTi-G, foram realizadas as medições dos ângulos de Euler da Plataforma. Os resultados obtidos pelos três controladores foram satisfatórios e ilustram o desempenho e a robustez da metodologia proposta. / In recent years there has been great interest in studying parallel manipulators, mainly applied in flight simulators, with six degrees of freedom. The interest in parallel kinematic structures is motivated by its high stiffness and excellent positioning capability in relation to serial kinematic structures. Furthermore, since the actuators are positioned on a base, they can handle heavy loads and also have low power consumption, considering that several actuators act on the same platform. This thesis presents the design of three controllers, which are, H-infinity controller with output feedback, PID controller and Fuzzy controller, so that this methodology can be employed in building a future flight simulator. The actuators models were obtained by a step voltage input to the engines and measuring their displacements by the encoders that are coupled to each of the respective axes of the motors. Knowing the relation from the motion transmission mechanism between the motor and the spindle of each actuator, the displacement of each spindle is obtained from the rotation of each motor measured by the corresponding encoder and thus we obtain the mathematical model of each actuator together with its transmission system. However, in practice, each actuator is slightly different from others, which leads to different behavior and performance of each. This affects the behavior of the platform making the final desired trajectory cannot be properly followed something that is extremely necessary in flight simulators. Thus, one of the important contributions of this work is first to present a methodology to standardize the actuators responses so that they all have in the end a behavior equal a close as possible, particularly in terms of velocity and positioning. With the kinematics data and platform dynamics composing the complete system model, several simulations applied to the real Stewart Platform validate the model and show the effectiveness of control techniques applied to control the position and orientation of the platform. In order to validate the Stewart Platform design as a possible base for a motion flight simulator, the longitudinal and lateral dynamics of a Boeing 747-100 model were implemented, and with the aid of an inertial sensor Xsens® MTi-G, measurements of the Euler angles of the platform were performed. The results obtained by the three controllers were satisfactory and illustrate the performance and robustness of the proposed methodology.

A ideia de usar redes neurais biologicamente inspiradas na computação tem sido amplamente utilizada nas últimas décadas. O fato essencial neste paradigma é que um neurônio pode integrar e processar informações, e esta informação pode ser revelada por sua atividade de pulsos. Ao descrever a dinâmica de um único neurônio usando um modelo matemático, uma rede pode ser implementada utilizando um conjunto desses neurônios, onde a atividade pulsante de cada neurônio irá conter contribuições, ou informações, da atividade pulsante da rede em que está inserido. Neste trabalho é apresentado um controlador de posição no eixo Z utilizando fusão de sensores baseado no paradigma de Redes Neurais Recorrentes. O sistema proposto utiliza uma Máquina de Estado Líquido (LSM) para controlar o robô colaborativo BAXTER. O framework foi projetado para trabalhar em paralelo com as LSMs que executam trajetórias em formas fechadas de duas dimensões, com o objetivo de manter uma caneta de feltro em contato com a superfície de desenho, dados de sensores de força e distância são alimentados ao controlador. O sistema foi treinado utilizando dados de um controlador Proporcional Integral Derivativo (PID), fundindo dados de ambos sensores. Resultados mostram que a LSM foi capaz de aprender o comportamento do controlador PID em diferentes situações. / The idea of employing biologically inspired neural networks to perform computation has been widely used over the last decades. The essential fact in this paradigm is that a neuron can integrate and process information, and this information can be revealed by its spiking activity. By describing the dynamics of a single neuron using a mathematical model, a network in which the spiking activity of every single neuron will get contributions, or information, from the spiking activity of the embedded network. A positioning controller based on Spiking Neural Networks for sensor fusion suitable to run on a neuromorphic computer is presented in this work. The proposed framework uses the paradigm of reservoir computing to control the collaborative robot BAXTER. The system was designed to work in parallel with Liquid State Machines that performs trajectories in 2D closed shapes. In order to keep a felt pen touching a drawing surface, data from sensors of force and distance are fed to the controller. The system was trained using data from a Proportional Integral Derivative controller, merging the data from both sensors. The results show that the LSM can learn the behavior of a PID controller on di erent situations.

High blood pressure, also called hypertension is one of the most common worldwide diseases afflicting humans and is a major risk factor for stroke, myocardial infarction, vascular disease, and chronic kidney disease. If blood pressure is controlled and oscillations in the hemodynamic variables are reduced, patients experience fewer complications after surgery. In clinical practice, this is usually achieved using manual drug delivery. Given that different patients have different sensitivity and reaction time to drugs, determining manually the right drug infusion rates may be difficult. This is a problem where automatic drug delivery can provide a solution, especially if it is designed to adapt to variations in the patient’s conditions. This research work presents an investigation into the development of abnormal blood pressure (hypertension) controllers for postoperative patients. Control of the drugs infusion rates is used to simultaneously regulate the hemodynamic variables such as the Mean Arterial Pressure (MAP) and the Cardiac Output (CO) at the desired level. The implementation of optimal control system is very essential to improve the quality of patient care and also to reduce the workload of healthcare staff and costs. Many researchers have conducted studies earlier on modelling and/or control of abnormal blood pressure for postoperative patients. However, there are still many concerns about smooth transition of blood pressure without any side effect. The blood pressure is classified in two categories: high blood pressure (Hypertension) and low blood pressure (Hypotension). The hypertension often occurred after cardiac surgery, and the hypotension occurred during cardiac surgery. To achieve the optimal control solution for these abnormal blood pressures, many methods are proposed, one of the common methods is infusing the drug related to blood pressure to maintain it at the desired level. There are several kinds of vasodilating drugs such as Sodium Nitroprusside (SNP), Dopamine (DPM), Nitro-glycerine (NTG), and so on, which can be used to treat postoperative patients, also used for hypertensive emergencies to keep the blood pressure at safety level. A comparative performance of two types of algorithms has been presented in chapter four. These include the Internal Model Control (IMC), and Proportional-Integral-Derivative (PID) controller. The resulting controllers are implemented, tested and verified for three sensitivity patient response. SNP is used for all three patients’ situation in order to reduce the pressure smoothly and maintain it at the desire level. A Genetic Algorithms (GAs) optimization technique has been implemented to optimise the controllers’ parameters. A set of experiments are presented to demonstrate the merits and capabilities of the control algorithms. The simulation results in chapter four have demonstrated that the performance criteria are satisfied with the IMC, and PID controllers. On the other hand, the settling time for the PID control of all three patients’ response is shorter than the settling time with IMC controller. Using multiple interacting drugs to control both the MAP and CO of patients with different sensitivity to drugs is a challenging task. A Multivariable Model Reference Adaptive Control (MMRAC) algorithm is developed using a two-input, two-output patient model. Because of the difference in patient’s sensitivity to the drug, and in order to cover the wide ranges of patients, Model Reference Adaptive Control (MRAC) has been implemented to obtain the optimal infusion rates of DPM and SNP. This is developed in chapters five and six. Computer simulations were carried out to investigate the performance of this controller. The results show that the proposed adaptive scheme is robust with respect to disturbances and variations in model parameters, the simulation results have demonstrated that this algorithm cannot cover the wide range of patient’s sensitivity to drugs, due to that shortcoming, a PID controller using a Neural Network that tunes the controller parameters was designed and implemented. The parameters of the PID controller were optimised offline using Matlab genetic algorithm. The proposed Neuro-PID controller has been tested and validated to demonstrate its merits and capabilities compared to the existing approaches to cover wide range of patients.

De manière générale, les turbomachines sont des machines tournantes permettant la conversion des différents types d’énergie. Ces dernières sont composées d’une partie mécanique en rotation, appelée rotor, interagissant avec un fluide. La rotation a donc un rôle clé pour ces machines et la liaison entre les parties fixes et les parties tournantes, appelée palier, est primordiale pour un fonctionnement fiable et optimal. Les turbomachines supportées par des paliers magnétiques actifs (PMAs) sont de plus en plus utilisées par les industriels notamment grâce à l’absence de contact direct entre parties fixes et parties tournantes, permettant un gain d’énergie et une réduction des émissions de CO2. La plupart du temps, ces machines sont « embarquées » et reposent sur des supports mobiles. Les mouvements générés par ces supports doivent être considérés dans la prévision du comportement dynamique des turbomachines afin d’améliorer les designs en conséquence. Cette thèse est une contribution à l’étude des turbomachines supportées par des PMAs sujettes à de fortes sollicitations extérieures. L’approche est numérique et expérimentale. L’utilisation d’un banc d’essais académique composé d’un système rotor-PMA, aux propriétés d’une turbomachine industrielle, a permis de tester les modèles développés pour des cas de sollicitations extérieures de type séisme et choc, générées grâce à l'excitateur 6-axes de l'equipex PHARE. Il est montré que le modèle permet la bonne prévision du comportement réel de la machine. Cet outil pourra donc être utilisé pour des designs de type industriel. / Turbomachines are rotating machines enabling the conversion of the different types of energy. The latter are composed of a rotating mechanical part, called rotor, interacting with a fluid. Therefore, rotation play a key role in these machines and the mechanical link between the fixed and the rotating parts, called bearing, is essential for reliable and optimal operations. Turbomachines supported by active magnetic bearings (AMBs) are increasingly used by industrial companies, especially thanks to the absence of direct contact between fixed and rotating parts, enabling energy savings and reduction of CO2 emissions. Most of the time, these machines are "on-board" and are fixed on mobile supports. The motions generated by these supports must be considered in the prediction of the dynamic behaviour of turbomachinery in order to improve the designs accordingly. This PhD is a contribution to the study of turbomachines supported by AMBs subjected to strong external base motions. The approach is numerical and experimental. The use of an academic scale test rig comprising a rotor-AMB system, with the properties of an industrial turbomachine, allowed to test the developed models for cases of external solicitations such as earthquake and shock, thanks to the 6-axis shaker of the equipex PHARE. It is shown that the model provides good predictions of the behaviour of the machine for the tested cases. This tool can therefore be used for industrial designs.

Submitted by Rosivalda Pereira (mrs.pereira@ufma.br) on 2017-07-18T19:13:43Z No. of bitstreams: 1 HiltonSantos.pdf: 3137200 bytes, checksum: a7b77b12eeb29959ab49e7ef675229d9 (MD5) / Made available in DSpace on 2017-07-18T19:13:43Z (GMT). No. of bitstreams: 1 HiltonSantos.pdf: 3137200 bytes, checksum: a7b77b12eeb29959ab49e7ef675229d9 (MD5) Previous issue date: 2017-06-27 / The emergence of new industrial plants with great complexity and the need to improve the operation of existing plants has fostered the development of high performance control systems, these systems must not only meet the design specifications, such as merit figures, but also operate at minimal cost and impacts at environment. Motivated by this demand, it is presented in this dissertation the development of methods for on-line tuning of control system parameters, ie, a methodology is presented for the on-line tuning of adaptive and optimal PID controllers via Artificial Neural Networks(ANNs). The approach developed in this dissertation is based on three PID controllers parameters. [Artificial neural networks with radial base functions and Model Predictive Control (MPC). From the union of these approaches a general formulation of an Adaptive-optimal PID controller via artificial neural networks with on-line tuning was presented. The on-line tuning methodology for the ANN parameters is presented in the context of MPC, predicting plant output. For the PID controller, we proposed a modification of the standard structure in order to adapt the error function. The adjustment of the PID controller parameters and the prediction of the optimally plant output, are performed by the ANN-RBF weights adjustments. In addition, an indoor implementation of the control system were proposed for the positioning of a photovoltaic panel. The performance evaluations of the proposed system were obtained from computational experiments results that were based on mathematical models and hardware experiments, that were obtained from a reduced model of a photovoltaic panel. Finally, a comparison between the proposed methodology with the classical PID controller were performed and the proposed methodology presented to be more flexible to the insertion of new performance metrics and the results achieved from the ANN, were better than the ones obtained by the classical PID tuning, such as: Ziegler-Nichols or trial and error. / O surgimento de novas plantas industriais com grande complexidade e a necessidade de melhorar a operação das plantas já existentes tem fomentado o desenvolvimento de sistemas de controle de alto desempenho, estes sistemas devem atender não só as especificações de projeto, tal como: figuras de mérito, mas também devem operar com um custo mínimo e sem causar impactos desastrosos para o meio ambiente. Motivados por esta demanda, apresenta-se nesta dissertação o desenvolvimento de métodos para sintonia online dos parâmetros dos sistemas de controle, ie, apresenta-se uma metodologia para a sintonia online de controladores PID adaptativo e ótimo via Redes Neurais Artificiais (RNAs). A abordagem desenvolvida nesta dissertação tem base as ações dos controladores PID de três termos, redes neurais artificiais com funções de base radial e Controle preditivo baseado em modelo (MPC - Model Predictive Control), a partir da união destas abordagens elabora-se a formulação geral do controlador PID Adaptativo-Ótimo via redes neurais artificiais, com sintonia online. A metodologia de ajuste online dos parâmetros da RNA está no contexto do MPC para predição de saída da planta. Para o caso do controlador PID, tem-se a modificação da estrutura padrão com o objetivo de adaptação em função do erro. O ajuste dos termos do controlador PID e da predição da saída na planta, de forma ótima, é realizada pelo ajustes dos pesos da RNA-RBF. Além disso, apresenta-se a implementação indoor do sistema de controle desenvolvido para o posicionamento de um painel fotovoltaico. As avaliações de desempenho do sistema proposto são obtidos de resultados de experimentos computacionais que são baseados em modelos matemáticos e experimentos em hardware que são obtidos de um modelo reduzido de um painel fotovoltaico. Por fim, comparando o PID clássico com o controlador desenvolvido constatou-se que este último apresenta mais flexibilidade para inserir novas métricas de desempenho e os resultados atingidos são melhores do que os parâmetros obtidos por meio da sintonia do PID clássica, tais como: métodos de Ziegler-Nichols ou tentativa e erro