Orientador: Jose Ricardo Pelaquim Mendes / Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecanica, Instituto de Geociencias / Made available in DSpace on 2018-08-12T14:15:08Z (GMT). No. of bitstreams: 1 Melo_AndreVerasde_M.pdf: 5529810 bytes, checksum: 4537025b6d5ddedffce982368776dbbd (MD5) Previous issue date: 2008 / Resumo: O cenario de producao em campos de petroleo tem apresentado nos anos recentes condicoes cada vez mais severas e complexas. Muitas descobertas em regioes de maior profundidade e com formacoes geologicas mais diversas tem demandado solucoes tecnológicas que beiram a fronteira do conhecimento, de forma que a producao tem andado lado a lado com o desenvolvimento de novas tecnologias. Assim, a necessidade do aumento do fator de recuperação para justificar os investimentos nestas condicoes sao desafios presentes nos novos projetos de pocos, de elevacao artificial e de recuperacao secundaria, o que vem levando ao desenvolvimento de solucoes submarinas e integradas, como o caso do separador submarino VASPS. No entanto, o controle de processos continuos, como a vazao de producao, a temperatura e a pressao, ainda e tratado da mesma forma que era feito quando do comeco da producao offshore em campos de oleo leve. Este trabalho visa permitir a melhoria de desempenho operacional do sistema VASPS, mensurada pelos custos totais gerados pelo sistema e pelos proventos gerados na producao de oleo e gas, bem como propor novas metodologias de sistemas de controle para sistemas similares, utilizando tecnologias recentes de controle inteligente. Vale ressaltar que o problema de controle e atualmente tratado em campo com solucoes de automacao, sem o estudo devido, causando danos a equipamentos que exigem intervencoes custosas. Os controladores propostos neste trabalho visam minimizar os esforcos aos quais estes atuadores estao expostos, provendo uma maior vida util as tecnologias de elevacao artificial de alta vazão e podendo ser aplicado de forma modular a outros processos relacionados. / Abstract: The oil production scenario has been becoming more and more complex and severe in the recent years. Plenty discoveries in deeper and more diversified formations have been requiring cutting edge technological solutions., such that the production has been deployed side by side with the technological development. Thus, the need for the enhancement of the recuperation factor in order to justify the investment in these new conditions are actual challenges within the recent design of wells, artificial lift systems and secondary recuperation projects, which has been leading to the development towards subsea integrated solutions, such as the VASPS subsea separator. However, the control of continuous processes, i.e. flow, temperature and pressure, is still dealt the same way it was done when the offshore production in light oil fields began. This work envisions the enhancement of the operational performance of the VASPS system, measured as a function of the total costs implied by the technology, altogether with the receipts generated in the gas and oil production, as well as the delivering of new methodologies for the design of controllers using late technologies of intelligent control. Also, it is worth recalling that nowadays the control task is dealt with automation solutions, without the proper handling, causing failures in equipments that require expensive rig intervention for workover. The controllers proposed in this work are designed to minimize the control effort, to which these actuators are exposed, providing extended lifetime to the high level production artificial lift technologies, being also able to be applied to other related processes in a modular form. / Mestrado / Explotação / Mestre em Ciências e Engenharia de Petróleo

M. Ing. / This work describes the establishment of an intelligent control system on board an AGV for automatic adaptation to the motions and actions of industrial robots. The movements can represent an assembly sequence or material loading sequence. A relationship is established between the workspaces of the industrial robot and the AGV. The coordinate systems are integrated for the AGV to respond to the movements of the industrial robot in the correct way. The integration of these two different coordinate systems leads to the creation of a common workspace. Within this common workspace, the AGV interacts with the robot using the same reference points. The mathematical analysis and practical implementation of this transient workspace is described. The adaptive control presented allows for an intelligent decision making process to be performed on line with the use of an expert system.

Repetitive control (RC) was proven to enable high performance for systems that are subject to periodically repeating signals by enhancing an existing feedback control system so that it produces zero tracking error to a periodic command, or zero tracking error in the presence of a periodic disturbance of known period. Periodic signals are very common in many applications like robotics, disk drive systems, power converters, photolithography, jitter or vibration elimination in spacecraft and many more. Due to the growth in micro-processor and micro-controller technologies, most of the controllers are implemented in digital domain. Digital RC is typically designed by assuming a known constant period of command/disturbance signal, which then leads to the selection of a fixed sampling period that keeps it synchronized with the command/disturbance signal. However, in practice, the period for these signals might not be accurately known or might vary with time. In order to overcome this problem, higher order RC (HORC) was proposed as one method to make RC less sensitive to period error or period fluctuations. This dissertation investigates HORC, specifically second and third order RC designs (SORC and TORC), to identify the limitations, gaps, and design tradeoffs that a control system designer faces. New designs and methods are developed to address such gaps including stability, designer tradeoffs, robustness and other related performance characteristics. This dissertation has three major parts: SORC designs and stability, SORC design tradeoffs, and TORC designs and stability.

Growing congestion problems of many metropolitan areas which cause excessive traffic delay, instability of travel time generated the need for the development of an Intelligent Vehicle Highway System (IVHS) program that is capable of making significant improvements in mobility, highway safety, and productivity on highways and urban streets. The success of such real time control system highly depends on the new methods that address dynamic traffic assignment. Thus there is an urgent need for an effective dynamic assignment model. The main objective of this research is to emphasize the importance of dynamic traffic assignment for Advanced Driver Information Systems (ADIS) which is one of the major components of IVHS and to present a practical traffic assignment model that is capable of running in real time and which can accurately predict link travel times, queue build up, and network performance. DYNTRAS (DYNamic Traffic Simulation Assignment), a simulation /assignment model is developed. The model uses an incremental loop that assigns a portion of the Origin-Destination matrix, and simulates the movement of the vehicles. Then, it updates travel times and assigns an additional portion of the O-D matrix. In contrast to traditional traffic assignment models like "capacity restraint" and “incremental assignment" techniques that do not consider time dimension, DYNTRAS incorporates time as a third dimension by keeping track of the vehicle movements in time. As a result, it is capable of predicting time-dynamic impacts of congestion and effects of diverted traffic on traffic flow more realistically. The model is applied to a test network. “Several experimental factors are varied to test the sensitivity of the model. The results obtained are presented and general conclusions are derived. The differences between dynamic and static traffic assignment results are also discussed by considering results obtained from both methods. The model needs to be calibrated using real traffic data. According to the results obtained, it needs to be validated. In addition, its long computation time should be reduced to be able to use it for real time applications. / Master of Science

This thesis is concerned with the applicability of model reference adaptive control to the control of robot manipulators under a wide range of configuration and payload changes, and a comparison of the performance of this technique with that of the non-adaptive schemes. The dynamic equations of robot manipulators are highly nonlinear and are difficult to determine precisely. For these reasons there is an interest in applying adaptive control techniques to robot manipulators. In this work, the detailed performance of three adaptive controllers are studied and compared with that of a non-adaptive controller, namely, the computed torque control scheme. Computer simulation results show that the use of adaptive control improves the performance of the manipulator despite changes in the payload or in the manipulator configuration. Making use of these results, a rule-based controller is developed by dividing a given manipulation task into portions where a particular adaptive control scheme, based on a specific linearized subsystem model, performs best. This strategy of selecting the proper controller during each portion of the overall task yields a performance having the least deviation from the desired trajectory during the entire length of the task.

Thesis (MEng)--University of Stellenbosch, 2000. / ENGLISH ABSTRACT: The growth in intelligent control has primarily been a reaction to the realisation that nonlinear control theory has been unable to provide practical solutions to present day control challenges. Consequently the chemical industry may be cited for numerous instances of overdesign, which result as an attempt to avoiding operation near or within complex (often more economically viable) operating regimes. Within these complex operating regimes robust control system performance may prove difficult to achieve using conventional (algorithmic) control methodologies. Biological neuronal control mechanisms demonstrate a remarkable ability to make accurate generalisations from sparse environmental information. Neural networks, with their ability to learn and their inherent massive parallel processing ability, introduce numerous opportunities for developing superior control structures for complex nonlinear systems. To facilitate neural network learning, reinforcement learning techniques provide a framework which allows for learning from direct interactions with a dynamic environment. lts promise as a means of automating the knowledge acquisition process is beguiling, as it provides a means of developing control strategies from cause and effect (reward and punishment) interaction information, without needing to specify how the goal is to be achieved. This study aims to establish evolutionary reinforcement learning as a powerful tool for developing robust neurocontrollers for application in highly nonlinear process systems. A novel evolutionary algorithm; Symbiotic, Adaptive Neuro-Evolution (SANE), is utilised to facilitate neurocontroller development. This study also aims to introduce SANE as a means of integrating the process design and process control development functions, to obtain a single comprehensive calculation step for maximum economic benefit. This approach thus provides a tool with which to limit the occurrence of overdesign in the process industry. To investigate the feasibility of evolutionary reinforcement learning in achieving these aims, the SANE algorithm is implemented in an event-driven software environment (developed in Delphi 4.0), which may be applied for both simulation and real world control problems. Four highly nonlinear reactor arrangements are considered in simulation studies. As a real world application, a novel batch distillation pilot plant, a Multi-Effect Batch Distillation (MEBAD) column, was constructed and commissioned. The neurocontrollers developed using SANE in the complex simulation studies, were found to exhibit excellent robustness and generalisation capabilities. In comparison with model predictive control implementations, the neurocontrollers proved far less sensitive to model parameter uncertainties, removing the need for model mismatch compensation to eliminate steady state off-set. The SANE algorithm also proved highly effective in discovering the operating region of greatest economic return, while simultaneously developing a neurocontroller for this optimal operating point. SANE, however, demonstrated limited success in learning an effective control policy for the MEBAD pilot plant (poor generalisation), possibly due to limiting the algorithm's search to a too small region of the state space and the disruptive effects of sensor noise on the evaluation process. For industrial applications, starting the evolutionary process from a random initial genetic algorithm population may prove too costly in terms of time and financial considerations. Pretraining the genetic algorithm population on approximate simulation models of the real process, may result in an acceptable search duration for the optimal control policy. The application of this neurocontrol development approach from a plantwide perspective should also have significant benefits, as individual controller interactions are so doing implicitly eliminated. / AFRIKAANSE OPSOMMING: The huidige groei in intelligente beheerstelsels is primêr 'n reaksie op die besef dat nie-liniêre beheerstelsel teorie nie instaat is daartoe om praktiese oplossings te bied vir huidige beheer kwelkwessies nie. Gevolglik kan talle insidente van oorontwerp in die chemiese nywerhede aangevoer word, wat voortvloei uit 'n poging om bedryf in of naby komplekse bedryfsgebiede (dikwels meer ekonomies vatbaar) te vermy. Die ontwikkeling van robuuste beheerstelsels, met konvensionele (algoritmiese ) beheertegnieke, in die komplekse bedryfsgebiede mag problematies wees. Biologiese neurobeheer megamsmes vertoon 'n merkwaardige vermoë om te veralgemeen vanaf yl omgewingsdata. Neurale netwerke, met hulle vermoë om te leer en hulle inherente paralleie verwerkingsvermoë, bied talle geleenthede vir die ontwikkeling van meer doeltreffende beheerstelsels vir gebruik in komplekse nieliniêre sisteme. Versterkingsleer bied a raamwerk waarbinne 'n neurale netwerk leer deur direkte interaksie met 'n dinamiese omgewing. Versterkingsleer hou belofte in vir die inwin van kennis, deur die ontwikkeling van beheerstrategieë vanaf aksie en reaksie (loon en straf) interaksies - sonder om te spesifiseer hoe die taak voltooi moet word. Hierdie studie beaam om evolutionêre versterkingsleer as 'n kragtige strategie vir die ontwikkeling van robuuste neurobeheerders in nie-liniêre prosesomgewings, te vestig. 'n Nuwe evolutionêre algoritme; Simbiotiese, Aanpasbare, Neuro-Evolusie (SANE), word aangewend vir die onwikkeling van die neurobeheerders. Hierdie studie beoog ook die daarstelling van SANE as 'n weg om prosesontwerp en prosesbeheer ontwikkeling vir maksimale ekonomiese uitkering, te integreer. Hierdie benadering bied dus 'n strategie waardeur die insidente van oorontwerp beperk kan word. Om die haalbaarheid van hierdie doelwitte, deur die gebruik van evolusionêre versterkingsleer te ondersoek, is die SANE algoritme aangewend in 'n Windows omgewing (ontwikkel in Delphi 4.0). Die Delphi programmatuur geniet toepassing in beide die simulasie en werklike beheer probleme. Vier nie-liniêre reaktore ontwerpe is oorweeg in die simulasie studies. As 'n werklike beheer toepassing, is 'n nuwe enkelladingsdistillasie kolom, 'n Multi-Effek Enkelladingskolom (MEBAD) gebou en in bedryf gestel. Die neurobeheerders vir die komplekse simulasie studies, wat deur SANE ontwikkel is, het uitstekende robuustheid en veralgemeningsvermoë ten toon gestel. In vergelyking met model voorspellingsbeheer implementasies, is gevind dat die neurobeheerders heelwat minder sensitief is vir model parameter onsekerheid. Die noodsaak na modelonsekerheid kompensasie om gestadigde toestand afset te elimineer, word gevolglik verwyder. The SANE algoritme is ook hoogs effektief vir die soek na die mees ekonomies bedryfstoestand, terwyl 'n effektiewe neurobeheerder gelyktydig vir hierdie ekonomies optimumgebied ontwikkel word. SANE het egter beperkte sukses in die leer van 'n effektiewe beheerstrategie vanaf die MEBAD toetsaanleg getoon (swak veralgemening). Die swak veralgemening kan toegeskryf word aan 'n te klein bedryfsgebied waarin die algoritme moes soek en die negatiewe effek van sensor geraas op die evaluasie proses. Vir industriële applikasies blyk dit dat die uitvoer van die evolutionêre proses vanaf 'n wisselkeurige begintoestand nie koste effektief is in terme van tyd en finansies nie. Deur die genetiese algoritme populasie vooraf op 'n benaderde modelop te lei, kan die soek tydperk na 'n optimale beheerstrategie aansienlik verkort word. Die aanwending van die neurobeheer ontwikkelingstrategie vanuit 'n aanlegwye oogpunt mag aanleiding gee tot aansienlike voordele, aaangesien individuele beheerder interaksies sodoende implisiet uitgeskakel word.

Dissertation (PhD)--University of Stellenbosch, 2000. / ENGLISH ABSTRACT: A standard fuzzy logic controller is not robust enough to guarantee consistent closed-loop performance for highly non-linear plants. A finely tuned closed-loop response loses relevance as the system dynamics change with operating conditions. The self-adaptive fuzzy logic controller can track changes in the system parameters and modify the controller parameters accordingly. In most cases, self-adaptive fuzzy logic controllers are complex and rely on some form of mathematical plant model. The multi-mode fuzzy logic controller extends the working range of a standard fuzzy logic controller by incorporating knowledge of the non-linear system dynamics into the control rule-base. The complexity of the controller and difficulty in finding control rules have limited the application of multi-mode fuzzy logic controllers. An automated design algorithm is proposed for the design of a multi-mode control rule-base using qualitative plant knowledge. The design algorithm is cost function-based. The closed-loop response, local to a domain of the non-linear state space, can be tuned by manipulation of the cost function weights. Global closed-loop response tuning can be done by manipulation of the controller input gains. Alternatively, a self-learning or self-adaptive algorithm can be used in a model reference adaptive control architecture to optimise the control rule-base. Control rules responsible for unacceptable closed-loop performance are identified and their consequences modified. The validity of the proposed design method is evaluated in five case studies. The case studies illustrate the advantages of the multi-mode fuzzy logic controller. The results indicate that the proposed self-adaptive algorithm can be used to optimise a rule-base given a required closed-loop specification. If the system does not conform to the model reference adaptive architecture then the intuitive nature of the cost function based design algorithm proves to be an effective method for rule-base tuning. / AFRIKAANSE OPSOMMING: Standaard wasige logika beheerders is nie noodwendig robuust genoeg om goeie geslote lus werkverrigting vir hoogs nie-liniere aanlegte te waarborg nie. In Perfek ge-optimeerde beheerder se geslote lus werkverrigting mag verswak indien die aanleg-parameters weens bedryfstoestande verander. Self-aanpassende beheerders kan die verandering in die aanleg-parameters volg en die beheerder dienooreenkomstig optimeer. As In reël is In self-aanpassende beheerder kompleks en afhanklik van In wiskundige model van die aanleg. Die multi-modus wasige logika beheerder vergroot die werksbereik van die standaard wasige logika beheerder deur kennis aangaande die stelsel se bedryfstoestand en stelselparameters in die reël-basis in te bou. Die aanwending van die multi-modus beheerder word tans beperk deur die struktuur kompleksiteit en moeilike optimering van die reël-basis. In Ge-outomatiseerde multi-modus reël-basis ontwerps-algoritme wat gebruik maak van kwalitatiewe kennis van die aanleg en In kostefunksie word in hierdie proefskrif voorgestel. Die geslote lus gedrag beperk tot In gebied in die toestands-ruimte kan ge-optimeer word deur die kostefunksie gewigte te manipuleer. Die globale werkverrigting kan ge-optimeer word met die beheerder intree aanwinste. In Self-aanpassende algoritme in In model-verwysings aanpassende argitektuur word as altematieftot reël-basis optimering voorgestel. Reëls verantwoordelik vir swak werkverrigting word ge-identifiseer en verbeter deur modifikasie van die reëls se gevolgtrekkings. Die voorgestelde ontwerps-metode word deur middel van vyf gevallestudies ondersoek. Die studies dui die voordele van die multi-modus struktuur aan. Die self-aanpassende argitektuur is In kragtige hulpbron om In reël-basis te optimeer vir In gegewe geslote lus spesifikasie. Hierdie proefskrif toon aan dat indien die stelsel nie aan die vereistes van In model verwysingstelsel voldoen nie, is die kostefunksie benadering tot reël-basis ontwerp In aantreklike en intuïtief verstaanbare opsie om die reël-basis te optimeer.

published_or_final_version / Mechanical Engineering / Doctoral / Doctor of Philosophy

published_or_final_version / abstract / Industrial and Manufacturing Systems Engineering / Doctoral / Doctor of Philosophy

The thesis investigates two domains, initially the kite domain and then part of a more demanding Rolls-Royce domain (jet engine design). Four main types of refinement rules that use the associated application conditions and domain ontology to support the maintenance of constraints are proposed. The refinement rules have been implemented in ConEditor and the extended system is known as ConEditor+. With the help of ConEditor+, the thesis demonstrates that an explicit representation of application conditions together with the corresponding constraints and the domain ontology can be used to detect inconsistencies, redundancy, subsumption and fusion, reduce the number of spurious inconsistencies and prevent the identification of inappropriate refinements of redundancy, subsumption and fusion between pairs of constraints.