Essa tese versa sobre um procedimento inovador para o controle e garantia da qualidade em processos de manufatura onde existem vários atributos e variáveis a serem monitorados, como acontece, por exemplo, na indústria cerâmica. O procedimento proposto, chamado de controle integrado da qualidade, começa com a definição de alguns postos de controle na linha de produção, onde o monitoramento será feito usando uma única carta de controle que agrega vários atributos e variáveis. Esse procedimento é complementado usando gráficos de Pareto, que hierarquizam as características de qualidade que contribuem mais fortemente para o percentual de defeituosos do posto, e as cartas de controle tradicionais, que são utilizadas seletivamente, apenas conforme a indicação dos gráficos de Pareto. A utilização conjunta dessas ferramentas auxilia no diagnóstico e solução dos problemas de qualidade, permitindo que as ações de melhoria aconteçam no tempo e local certo. Entre as vantagens do procedimento proposto, cita-se: (a) a simplicidade do controle integrado, que foca prioritariamente umas poucas cartas de controle; (b) a abordagem estatística, que fornece uma base sólida para a tomada de decisões; e (c) o forte apelo gerencial fornecido pelo controle integrado, que disponibiliza a cada amostra uma visão geral da condição de qualidade nos postos de controle e na linha de produção como um todo. A tese apresenta também uma abordagem sistemática para auxiliar na implantação do controle integrado da qualidade. Essa abordagem contempla 5 etapas principais: definição do projeto, planejamento da implantação, treinamento, implantação efetiva e acompanhamento e consolidação. Por fim, é apresentado e discutido um estudo de caso em uma indústria cerâmica, que contribui para a validação do método proposto. Com base nesse estudo de caso e no referencial teórico, conclui-se sobre o assunto. / This dissertation presents a new procedure, Integrated Quality Control, for quality control and assurance in situations where severa! variables and attributes are to be monitored. That is the case in the ceramics manufacturing industry, for example. The procedure is implemented by defining control (or work) stations in the production line; variables and attributes in each station are monitored simultaneously using an aggregated control chart for the percentages of non-conformities. When a signal occurs, a Pareto graph sorts out variables and attributes identifying those contributing the most to its occurrence: such variables and attributes are then examined using traditional control charts. The joint use of aggregate charts, Pareto graphs and univariate control charts turns monitoring and improvement of processes into a non-complex task. A systematic approach for the implementation of Integrated Quality Control is also presented. The approach comprises five main steps: project definition, implementation planning, training of personnel, effective implementation, and follow up. Key advantages of the procedure suggested here are: (i) ability to handle variables and attributes in a single integrated chart, (ii) the statistical approach which provides a solid basis for the decision-making, and (iii) strong managerial appeal provided by the integrated charts. The procedure is illustrated with a case study conducted in a ceramic plant. Some conclusions are drawn based on the results of this case study and in the theoretical references.

Thesis (Sc.D.)--Massachusetts Institute of Technology, Dept. of Chemical Engineering, 1980. / MICROFICHE COPY AVAILABLE IN ARCHIVES AND SCIENCE. / Bibliography: leaves 412-417. / by Siri Jirapongphan. / Sc.D.

No Brasil, a fermentação alcoólica é o processo predominante para produção de etanol a partir de cana-de-açúcar. A utilização de mosto esterilizado associada ao uso de linhagens de leveduras selecionadas apresenta-se como uma das estratégias com potencial para proporcionar melhorias no processo de fermentação alcoólica para produção de etanol e para aumentar a produtividade da indústria sucroenergética. Embora existam diversos métodos aplicáveis na esterilização de compostos líquidos, o tratamento térmico baseado no uso de vapor superaquecido ou saturado é interessante na indústria sucroenergética devido a disponibilidade de redes de distribuição de vapor. Essa pesquisa teve como objetivo desenvolver um trocador de calor de contato direto equipado com sistema eletrônico microcontrolado para controle dinâmico da temperatura de misturas de fluidos a diferentes temperaturas. A tecnologia poderá ser aplicada ao tratamento térmico do mosto utilizado em processos de fermentação alcoólica na indústria sucroenergética. O sistema de mistura utilizado foi o aparato montado para misturar dois fluidos distintos. O aparato era composto por um trocador de calor de contato direto, uma válvula proporcional, um sensor de temperatura e duas válvulas de retenção. O trocador de calor foi montado em forma de um \"tê\" como dispositivo para mistura de fluidos. Foram realizados experimentos em laboratório; A partir dos dados gerados foi determinado a função de transferência característica do sistema seguida pela sintonização de um controlador Proporcional-Integral. O controlador foi sintonizado com a finalidade de obter temperatura alvo na câmara de mistura do sistema, através de ajustes na seção de escoamento da tubulação de vapor. O desempenho do equipamento foi avaliado com base no tempo de subida, sobressinal, tempo de acomodação, IAE e ITAE, evidenciando-se adequado desempenho para a aplicação em questão. As atividades foram limitadas ao desenvolvimento do equipamento, havendo maior enfoque em aspectos relacionados a controle. Não foram realizadas avaliações da eficácia do tratamento térmico em aplicações industriais ou laboratoriais. Trabalhos futuros poderão utilizar a tecnologia desenvolvida nesta pesquisa para estudos da eficácia do tratamento térmico conduzido sob diferentes temperaturas na redução de microrganismos ou esterilização do mosto. / In Brazil, alcoholic fermentation is the predominant process for the production of ethanol from sugarcane. The use of sterilized must associated with the use of selected yeast strains is one of the strategies with potential to provide improvements in the alcoholic fermentation process for ethanol production and to increase the productivity of the sugarcane industry. Although there are several methods applicable to the sterilization of liquid compounds, heat treatment based on the use of superheated or saturated steam is interesting in the sugarcane industry due to the availability of steam distribution networks. This research aimed to develop a direct contact heat exchanger equipped with a microcontrolled electronic system for dynamic temperature control of fluid mixtures at different temperatures. The technology can be applied to the heat treatment of must used in alcoholic fermentation processes in the sugar-energy industry. The mixing system was the apparatus which was assembled to mix two distinct fluids. The apparatus consisted of a direct contact heat exchanger, a proportional valve, a temperature sensor and two check valves. The heat exchanger was assembled in the form of a \"t\" as a fluid mixing device. Experiments were performed in the laboratory; the characteristic transfer function of the system was determined followed by the tuning of a Proportional-Integral controller. The controller was tuned for the purpose of obtaining target temperature in the system mixing chamber by adjusting the flow section of the steam pipe. The performance of the equipment was evaluated based on rise time, settling time, percent overshoot, IAE e ITAE, showing adequate performance for the application in question. The activities were limited to the development of technology, with a greater focus on aspects related to control. No evaluation of the effectiveness of heat treatment in industrial or laboratory applications has been made. Future works may use the technology developed in this research to study the efficacy of heat treatment conducted at different temperatures in the reduction of microorganisms or sterilization of the must.

This dissertation presents a series of new results of iterative learning control (ILC) that progresses from model-based ILC algorithms to data-driven ILC algorithms. ILC is a type of trial-and-error algorithm to learn by repetitions in practice to follow a pre-defined finite-time maneuver with high tracking accuracy. Mathematically ILC constructs a contraction mapping between the tracking errors of successive iterations, and aims to converge to a tracking accuracy approaching the reproducibility level of the hardware. It produces feedforward commands based on measurements from previous iterations to eliminates tracking errors from the bandwidth limitation of these feedback controllers, transient responses, model inaccuracies, unknown repeating disturbance, etc. Generally, ILC uses an a priori model to form the contraction mapping that guarantees monotonic decay of the tracking error. However, un-modeled high frequency dynamics may destabilize the control system. The existing infinite impulse response filtering techniques to stop the learning at such frequencies, have initial condition issues that can cause an otherwise stable ILC law to become unstable. A circulant form of zero-phase filtering for finite-time trajectories is proposed here to avoid such issues. This work addresses the problem of possible lack of stability robustness when ILC uses an imperfect a prior model. Besides the computation of feedforward commands, measurements from previous iterations can also be used to update the dynamic model. In other words, as the learning progresses, an iterative data-driven model development is made. This leads to adaptive ILC methods. An indirect adaptive linear ILC method to speed up the desired maneuver is presented here. The updates of the system model are realized by embedding an observer in ILC to estimate the system Markov parameters. This method can be used to increase the productivity or to produce high tracking accuracy when the desired trajectory is too fast for feedback control to be effective. When it comes to nonlinear ILC, data is used to update a progression of models along a homotopy, i.e., the ILC method presented in this thesis uses data to repeatedly create bilinear models in a homotopy approaching the desired trajectory. The improvement here makes use of Carleman bilinearized models to capture more nonlinear dynamics, with the potential for faster convergence when compared to existing methods based on linearized models. The last work presented here finally uses model-free reinforcement learning (RL) to eliminate the need for an a priori model. It is analogous to direct adaptive control using data to directly produce the gains in the ILC law without use of a model. An off-policy RL method is first developed by extending a model-free model predictive control method and then applied in the trial domain for ILC. Adjustments of the ILC learning law and the RL recursion equation for state-value function updates allow the collection of enough data while improving the tracking accuracy without much safety concerns. This algorithm can be seen as the first step to bridge ILC and RL aiming to address nonlinear systems.

Sistemas de geração de referências para as malhas de controle de laminadores a frio têm proporcionado elevados níveis de produtividade e de qualidade do material processado. No caso de falha da unidade de processamento responsável pela execução do sistema de geração de referências, a operação do laminador fica comprometida e necessita-se de um modo de operação emergencial. Nesse trabalho é desenvolvido um sistema alternativo para geração de referências que visa substituir o modo de operação de emergência de consulta a tabelas de referências pré-calculadas da planta em questão. O sistema de geração de referências desenvolvido, assim como o sistema usado em condições normais de operação, é composto por uma função custo que avalia o nível de produtividade e de qualidade que um conjunto de referências pode proporcionar. A função custo é minimizada pelo método simplex de Nelder e Mead e as variáveis do processo envolvidas na função custo são avaliadas por um modelo de laminação a frio composto somente por equações algébricas, enquanto que o sistema usado em condições normais de operação usa um modelo clássico, composto por equações algébricas e integrais. A comparação entre o sistema usado em condições normais de operação e o sistema desenvolvido mostrou que apesar desse sistema ser mais simples, menos preciso e robusto, as referências por ele calculadas são aceitáveis e podem proporcionar resultados superiores que o modo de operação de emergência baseado na consulta a tabelas. / Setup generation systems for cold mill control loops have provided rolling mill operation with high levels of quality and productivity. In case of malfunctioning in the process unit responsible to execute the setup generation system, the normal procedure for the operation of the mill is interrupted, being necessary the use of an emergency operation mode. This work develops an alternative system for references generation which aims to replace the current emergency operation mode based on pre-calculated, not continuous, setup tables for the rolling mill. The developed references generation system, like the normal system used during mill operation, is composed by a cost function which evaluated the mill quality and productivity operation level which a set of references may propose. The cost function is minimized using the Nelder and Mead simplex method and the process variables involved in the cost function are evaluated by a cold rolling mill model composed only by algebraic equations, while the normal operation system uses a classical model, made up by integrals and algebraic equations. Comparing the existing main setup system to the developed reference system it was observed that, despite its simplicity, less accuracy and less robustness, references calculated using this system were accurate enough to achieve better results than the one emergency table mode has provided.

This work describes analysis, design, and implementation of a novel switching robust adaptive control (SRAC) method for nonlinear systems. The proposed method takes advantage of both adaptive control (AC) and robust control (RC) methods. SRAC employs one of the methods when this method is advantageous and switches to the other method when the other one becomes the preferred choice. To this end, RC is used to deal with transient effects caused by uncertainties and disturbances. The system switches over to AC for good steady state performance when certain switching criteria are satisfied. If external disturbances become dominant or new uncertainties are introduced while AC is active, the system will switch back to RC. In this manner, the switching process between AC and RC will continue to take place guaranteeing improved performance, robustness, and accuracy for the entire operation of the system. The novel idea behind the proposed method is a smart novel mechanism of bi-directional switching between RC and AC. In this mechanism, the involvement of estimators and switching rules play a decisive part in guaranteeing the smooth switching and the stability of the system. The implementation and design issues of the novel method were first evaluated by simulation on a mass spring system and then on a robot manipulator system. To control these systems with satisfactory performance, nonlinearities and uncertainties have been properly analysed and embedded into models and control algorithms. Simulation results showed the superior performance of the proposed method compared with other control methods. The experimental validation of the proposed method was conducted on a Puma 560 robot manipulator system which was established by joints 2 and 3 of the robot. Extensive comparative experimental results have validated the efficacy and superior performance of the proposed SRAC method over other control methods in the face of uncertainties and disturbances. As part of this work, a comprehensive dynamic model of robotic manipulator in the presence of joint motors, gravitational forces, friction forces and payload has been developed using MAPLE. A systematic design framework for the SRAC method has also been developed.

The recent downturn in world sugar prices has placed even greater demands upon the Australian sugar industry to reduce the costs of sugar manufacture and increase the consistency of producing high quality sugar. One of the proposed approaches in increasing the consistency of very high quality sugar production and leveraging further avenues for cost saving is in the development of a computer based advisory system. This system is able to provide expert knowledge in the area of pan stage management and best practices in the absence of human experts. This thesis explores the design, key features and outcomes of a knowledge based supervisory support system (KBSSS) framework proposed specifically for providing cooperative decision support in the area of pan stage operations within a sugar mill. To demonstrate the viability of the proposed KBSSS framework a prototype system was developed in accordance with the proposed framework. The KBSSS utilises three core innovative system technologies that form the core components of the proposed KBSSS framework. These technologies are: 1) Dynamic industrial pan stage process models for identifying the dynamic relationships between sections of pan stage operations to allow for future forecasting of pan stage operating conditions, 2) Integration techniques for the merging of the developed pan stage process models into the hybrid fuzzy logic expert system rule base to provide localisation adjustment to match with local real world factory operational conditions, and 3) Explanatory capabilities to provide justification and support of system advice and recommendations. As a result of research and development carried out in this thesis, the KBSSS's test results demonstrated in the thesis indicate the viability of the proposed KBSSS framework and highlight the forecasting capabilities of the developed system resulting in favourable outcomes compared to data from pan stage operations. As a result of the research undertaken in the thesis a prototype KBSSS, for pan stage operations, based upon the three core supporting intelligent system technologies reported in the thesis has been developed.

The operation of a chemical process plant has become more complex with the addition of process integration and intensification. A greater emphasis on producing goods with the lowest product variability in the safest manner possible and stringent environmental regulation limiting the quantity of effluent release have all put more constraints on the physical and economic performance of the chemical plant. The performance of a plant is quantified by the ability of the process system to achieve its objectives, which is governed by its process design and control. The conventional approach to process design and control selection starts sequentially by proposing a process flowsheet for the plant. The selection criteria for a flowsheet are normally based only on its environmental impact and economic merits. It is after a process flowsheet is deemed financially suitable that process control development commences. However, a more integrated approach to process design and control stage may thus lead to a plant that has better achievable performance. The aim of this project is to provide a new approach to quantitative dynamic controllability analysis for integration of process design and control by using the concept of passivity and passive systems. Passivity is an input/output property of processes. Passive processes are stable and minimum phase and therefore very easy to control. For a given process, its shortage of passivity, which reflects destabilizing effects of factors such as time delays and Right-Half Plane (RHP) zeros, can be used to indicate its controllability. The project focuses in developing the proposed controllability analysis by combining the idea of passivity and IMC invertibility, which is then formulated into an optimization problem that can be solved by either using Semi-Definite Programming or Non-Linear Optimization. The achievable performance of the plant is quantified in terms of the sensitivity function of the open-loop process. The selection of a process from four different heat-integrated distillation column schemes was used as a case study and the result had clearly shown that the passivity-based controllability analysis was able to select a process based on the plant achievable performance under the constraint of passivity and design parameters.

The wide range of applications for single input single output controllers have encouraged interest in monitoring their performance. Over the past two decades researchers in the area have found many performance enhancement opportunities by applying these techniques. These are most evident in large operational plants with hundreds of controllers being monitored at the same time. Early performance measures were based on minimum variance control as a benchmark for controller performance. Many other procedures have since emerged that have improved the level of accuracy in these performance measures. In addition, these improvements made it easier to implement control loop performance monitoring in large industrial settings. This thesis looks at the performance measures in use for single input single output controllers. The work here looks at incorporating these different measures for a specific manufacturing plant. Ways of identifying the goals and objectives of controllers in a system are presented. Furthermore, measures are proposed that most accurately indicate if these goals and objectives are being met. The concept is demonstrated on a distillation system in a gas plant. It is shown how using these objective driven techniques can provide the user with sound results. These results do not require much user analysis to identify sources of problems and areas of improvement.

<p>The main objective of this thesis is to use available knowledge about a process and combine this with measurement data from the same process to extract more information about the process. The combination of knowledge and measurement data is referred to as Multi Sensor Data Fusion, MSDF. This added information is then used to control the process towards a specified goal.</p><p>The process studied in this thesis is the process of drilling wells in a petroleum reservoir, while the oil is flowing from the reservoir. In the petroleum industry, this is defined as underbalanced drilling (UBD), where the bottom hole pressure (BHP) in the well is below the pore pressure in the reservoir.</p><p>Detailed knowledge of the process is of paramount importance when using multi sensor data fusion. Due to this, various process modelling efforts are examined and evaluated, from simple relations between parameters to a finite-element approach of modelling the fluid flow in the well during drilling. Several sensors are used in the various cases, and existing sensors such as pressure sensors and flow sensors are the main data source in the analysis. Future scenario with sensors such as pressure arrays and non-intrusive multiphase flow meters are evaluated. In addition, new positions of existing sensor systems are discussed.</p><p>The methods available for fusing the knowledge of the process represented as models together with the available data is ranging from artificial intelligent methods such as neural networks, to methods incorporating statistical analysis such as various Kalman filters. History matching techniques using gradient techniques are also examined.</p><p>The migration of reservoir fluids into the well during UBD influences the BHP of the well. The results in the thesis show that this reservoir influx can be calculated by estimating some of the important reservoir parameters such as reservoir pore pressure or reservoir permeability. These reservoir parameters can be estimated most efficiently by performing an MSDF using a detailed nonlinear model of the well and reservoir dynamic behaviour together with real-time measurements of the fluid flow parameters such as fluid temperature, fluid pressure and fluid flow rates. The unscented Kalman filter shows the best performance when evaluating both estimation accuracy and computational requirements.</p><p>Regarding available instrumentation for use during UBD, the analysis shows that there is a major potential in introducing new sensors. As new data transmission methods are emerging and making data from sensors distributed along the drillstring available, this can generate a shift in paradigm regarding real-time analysis of reservoir properties during drilling.</p><p>Controlling the process is an important usage of the information gained from the MSDF analysis. Various control methods for controlling the most important process variables are examined and evaluated. The results show that acceptable pressure control can be obtained when using the choke valve opening as the primary control parameter. However, the choke valve operation has to be closely coordinated with drilling fluid flow rate adjustments. The choke valve opening control is able to compensate for pressure variations during the whole drilling operation.</p><p>A suggested nonlinear model predictive control algorithm gives best results when looking at the control accuracy, and can easily be expanded to handle multiple control inputs and system constraints. This control algorithm uses a detailed model of the well and reservoir dynamics. The Levenberg-Marquardt algorithm is used to calculate the optimal future control variables. The main drawback of the control algorithm is computational burden. A linear control algorithm, which also is evaluated, uses less computational resources, but has less control accuracy and is more difficult to expand into a multivariable control system.</p><p>Recommendations for further work are to expand the suggested model predictive control algorithm to handle more control inputs, while reducing the computational burden by incorporating low-order models for describing the future behaviour of the well.</p>