903 resultados para FUZZY-LOGIC SYSTEMS


Relevância:

80.00% 80.00%

Publicador:

Resumo:

Este trabalho apresenta a estruturação de um controle difuso, para a automação de reatores seqüenciais em batelada (RSB), no processo de remoção biológica de matéria orgânica e nitrogênio em águas residuárias domésticas, utilizando parâmetros inferenciais, pH, ORP e OD, em que as variáveis controladas foram as durações da reação aeróbia e anóxica. O experimento, em escala de bancada, foi composto por dois reatores seqüenciais em batelada, com volume útil de 10 L, no qual 6 L foram alimentados com esgoto sintético, com características de águas residuárias domésticas. O sistema de automação foi composto pela aquisição dos parâmetros eletroquímicos (pH, ORP e OD), pelos dispositivos atuadores (motor-bomba, aerador e misturador) e pelo controle predeterminado ou difuso. O programa computacional CONRSB foi implementado de forma a integrar o sistema de automação. O controle difuso, implementado, foi constituído pelos procedimentos de: normalização, nebulização, inferência, desnebulização e desnormalização. As variáveis de entrada para o controlador difuso, durante o período: aeróbio foram dpH/dt, dpH/d(t-1) e o pH ; anóxico foram dORP/dt, dORP/d(t-1) e o OD. As normalizações das variáveis crisps estiveram no universo de [0,1], utilizando os valores extremos do ciclo 1 ao 70. Nas nebulizações foram aplicadas as funções triangulares, as quais representaram, satisfatoriamente, as indeterminações dos parâmetros. A inferência nebulosa foi por meio da base heurística (regras), com amparo do especialista, em que a implicação de Mamdani foi aplicada Nessas implicações foram utilizadas dezoito expressões simbólicas para cada período, aeróbio e anóxico. O método de desnebulização foi pelo centro de áreas, que se mostrou eficaz em termos de tempo de processamento. Para a sintonia do controlador difuso empregou-se o programa computacional MATLAB, juntamente com as rotinas Fuzzy logic toolbox e o Simulink. O intervalo entre as atuações do controlador difuso, ficou estabelecido em 5,0 minutos, sendo obtido por meio de tentativas. A operação do RSB 1, durante os 85 ciclos, apresentou a relação média DBO/NTK de 4,67 mg DBO/mg N, sendo classificado como processo combinado de oxidação de carbono e nitrificação. A relação média alimento/microrganismo foi de 0,11 kg DBO/kg sólido suspenso volátil no licor misto.dia, enquadrando nos sistemas com aeração prolongada, em que a idade do lodo correspondeu aos 29 dias. O índice volumétrico do lodo médio foi de 117,5 mL/g, indicando uma sedimentação com características médias. As eficiências médias no processo de remoção de carbono e nitrogênio foram de 90,8% (como DQO) e 49,8%, respectivamente. As taxas específicas médias diárias, no processo de nitrificação e desnitrificação, foram de 24,2g N/kg SSVLM.dia e 15,5 g N/kg SSVLM.dia, respectivamente. O monitoramento, em tempo real, do pH, ORP e OD, mostrou ter um grande potencial no controle dos processos biológicos, em que o pH foi mais representativo no período aeróbio, sendo o ORP e o OD mais representativos no período anóxico. A operação do RSB com o controlador difuso, apresentou do ciclo 71 ao 85, as eficiências médias no processo de remoção de carbono e nitrogênio de 96,4% (como DQO) e 76,4%, respectivamente. A duração média do período aeróbio foi de 162,1 minutos, que tomando como referência o período máximo de 200,0 minutos, reduziu em 19,0% esses períodos. A duração média do período anóxico foi de 164,4 minutos, que tomando como referência o período máximo de 290,0 minutos, apresentou uma redução de 43,3%, mostrando a atuação robusta do controlador difuso. O estudo do perfil temporal, no ciclo 85, mostrou a atuação efetiva do controlador difuso, associada aos pontos de controle nos processos biológicos do RSB. Nesse ciclo, as taxas máximas específicas de nitrificação e desnitrificação observadas, foram de 32,7 g NO3 --N/kg sólido suspenso volátil no licor misto.dia e 43,2g NO3 --N/kg sólido suspenso volátil no licor misto.dia, respectivamente.

Relevância:

80.00% 80.00%

Publicador:

Resumo:

As pesquisas sobre governança no sistema financeiro contribuem significativamente para a análise dos diversos elementos que influenciam a performance nesse setor. No entanto, estudos empíricos nessa área ainda são limitados. Um dos motivos é a complexidade inerente à noção de governança na área pública. Da mesma forma que os sistemas complexos, a governança pode ser descrita como um sistema que abrange um grande número de entidades interdependentes entre si, com diferentes graus de relacionamentos. Neste trabalho de pesquisa, o significado de governança regulamentar do SFN se insere nesse escopo de entendimento, isto é, a governança como um fenômeno que resulta das diversas interações existentes entre os atores que influenciam ou são influenciados pelas atividades de regulação do setor financeiro. Em função das especificidades dos sistemas complexos, desenvolve-se e implementa-se um modelo baseado em agentes para a análise da governança regulamentar do SFN mediante experimentos de simulação. Os modelos baseados em agentes possibilitam explicitar aspectos relativos às interações e comportamentos dos agentes (nível micro), ou seja, os comportamentos não-lineares do sistema, que são difíceis de serem capturados com outros formalismos matemáticos. O modelo baseado em agentes é integrado a um modelo econométrico que tem como função caracterizar o ambiente macro-econômico. O ambiente micro é modelado por intermédio de agentes computacionais, com o uso da arquitetura BDI (do inglês, beliefs-desires-intentions). Esses agentes interagem entre si e com o ambiente, possuem crenças sobre o meio onde atuam e desejos que querem satisfazer, levando-os a formar intenções para agir. O comportamento dos agentes foi modelado utilizando-se lógica difusa (fuzzy logic), com o uso de regras construídas por intermédio de pesquisa de análise de conteúdo, a partir de informações coletadas em notícias de jornais, e entrevistas semiestruturadas com especialistasda área financeira. Os resultados dos experimentos demonstram o potencial da simulação dos modelos baseados em agentes para a realização de estudos de ambientes complexos de governança regulamentar.

Relevância:

80.00% 80.00%

Publicador:

Resumo:

The Behavioral Finance develop as it is perceived anomalies in these markets efficient. This fields of study can be grouped into three major groups: heuristic bias, tying the shape and inefficient markets. The present study focuses on issues concerning the heuristics of representativeness and anchoring. This study aimed to identify the then under-reaction and over-reaction, as well as the existence of symmetry in the active first and second line of the Brazilian stock market. For this, it will be use the Fuzzy Logic and the indicators that classify groups studied from the Discriminant Analysis. The highest present, indicator in the period studied, was the Liabilities / Equity, demonstrating the importance of the moment to discriminate the assets to be considered "winners" and "losers." Note that in the MLCX biases over-reaction is concentrated in the period of financial crisis, and in the remaining periods of statistically significant biases, are obtained by sub-reactions. The latter would be in times of moderate levels of uncertainty. In the Small Caps the behavioral responses in 2005 and 2007 occur in reverse to those observed in the Mid-Large Cap. Now in times of crisis would have a marked conservatism while near the end of trading on the Bovespa speaker, accompanied by an increase of negotiations, there is an overreaction by investors. The other heuristics in SMLL occurred at the end of the period studied, this being a under-reaction and the other a over-reaction and the second occurring in a period of financial-economic more positive than the first. As regards the under / over-reactivity in both types, there is detected a predominance of either, which probably be different in the context in MLCX without crisis. For the period in which such phenomena occur in a statistically significant to note that, in most cases, such phenomena occur during the periods for MLCX while in SMLL not only biases are less present as there is no concentration of these at any time . Given the above, it is believed that while detecting the presence of bias behavior at certain times, these do not tend to appear to a specific type or heuristics and while there were some indications of a seasonal pattern in Mid- Large Caps, the same behavior does not seem to be repeated in Small Caps. The tests would then suggest that momentary failures in the Efficient Market Hypothesis when tested in semistrong form as stated by Behavioral Finance. This result confirms the theory by stating that not only rationality, but also human irrationality, is limited because it would act rationally in many circumstances

Relevância:

80.00% 80.00%

Publicador:

Resumo:

This work presents a hybrid approach for the supplier selection problem in Supply Chain Management. We joined decision-making philosophy by researchers from business school and researchers from engineering in order to deal with the problem more extensively. We utilized traditional multicriteria decision-making methods, like AHP and TOPSIS, in order to evaluate alternatives according decision maker s preferences. The both techiniques were modeled by using definitions from the Fuzzy Sets Theory to deal with imprecise data. Additionally, we proposed a multiobjetive GRASP algorithm to perform an order allocation procedure between all pre-selected alternatives. These alternatives must to be pre-qualified on the basis of the AHP and TOPSIS methods before entering the LCR. Our allocation procedure has presented low CPU times for five pseudorandom instances, containing up to 1000 alternatives, as well as good values for all considered objectives. This way, we consider the proposed model as appropriate to solve the supplier selection problem in the SCM context. It can be used to help decision makers in reducing lead times, cost and risks in their supply chain. The proposed model can also improve firm s efficiency in relation to business strategies, according decision makers, even when a large number of alternatives must be considered, differently from classical models in purchasing literature

Relevância:

80.00% 80.00%

Publicador:

Resumo:

In order to guarantee database consistency, a database system should synchronize operations of concurrent transactions. The database component responsible for such synchronization is the scheduler. A scheduler synchronizes operations belonging to different transactions by means of concurrency control protocols. Concurrency control protocols may present different behaviors: in general, a scheduler behavior can be classified as aggressive or conservative. This paper presents the Intelligent Transaction Scheduler (ITS), which has the ability to synchronize the execution of concurrent transactions in an adaptive manner. This scheduler adapts its behavior (aggressive or conservative), according to the characteristics of the computing environment in which it is inserted, using an expert system based on fuzzy logic. The ITS can implement different correctness criteria, such as conventional (syntactic) serializability and semantic serializability. In order to evaluate the performance of the ITS in relation to others schedulers with exclusively aggressive or conservative behavior, it was applied in a dynamic environment, such as a Mobile Database Community (MDBC). An MDBC simulator was developed and many sets of tests were run. The experimentation results, presented herein, prove the efficiency of the ITS in synchronizing transactions in a dynamic environment

Relevância:

80.00% 80.00%

Publicador:

Resumo:

The traditional processes for treatment of hazardous waste are questionable for it generates other wastes that adversely affect people s health. As an attempt to minimize these problems, it was developed a system for treatment of hazardous waste by thermal plasma, a more appropriate technology since it produces high temperatures, preventing the formation of toxic pollutants to human beings. The present work brings out a solution of automation for this plant. The system has local and remote monitoring resources to ensure the operators security as well as the process itself. A special attention was given to the control of the main reactor temperature of the plant as it is the place where the main processing occurs and because it presents a complex mathematical model. To this, it was employed cascaded controls based on Fuzzy logic. A process computer, with a particular man-machine interface (MMI), provides information and controls of the plant to the operator, including by Internet. A compact PLC module is in charge of the central element of management automation and plant control which receives information from sensors, and sends it to the MMI

Relevância:

80.00% 80.00%

Publicador:

Resumo:

The Oil Measurement Evaluation Laboratory (LAMP), located in the Federal University of Rio Grande do Norte (UFRN), has as main goal to evaluate flow and BS&W meters, where the simulation of a bigger number of operation variable in field, guarantees a less uncertain evaluation. The objective of this work is to purpose a heating system design and implementation, which will control the temperature safely and efficiently in order to evaluate and measure it. Temperature is one of the variables which influence the flow and BS&W accurate measurement, directly affecting the fluid viscosity and density in the experiment. To project the heating system it is of great importance to take the laboratory requirements, conditions and current restrictions into consideration. Three alternatives were evaluated: heat exchanger, internal resistance and external resistance. After the analyses are made in order to choose the best alternative for the heating system in the laboratory, control strategies were determined for it, PID control methods in combination with fuzzy logic were used. Results showed a better performance with fuzzy logic than with classic PID

Relevância:

80.00% 80.00%

Publicador:

Resumo:

Formalization of logical systems in natural deduction brings many metatheoretical advantages, which Normalization proof is always highlighted. Modal logic systems, until very recently, were not routinely formalized in natural deduction, though some formulations and Normalization proofs are known. This work is a presentation of some important known systems of modal logic in natural deduction, and some Normalization procedures for them, but it is also and mainly a presentation of a hierarchy of modal logic systems in natural deduction, from K until S5, together with an outline of a Normalization proof for the system K, which is a model for Normalization in other systems

Relevância:

80.00% 80.00%

Publicador:

Resumo:

A segurança ocupacional é imprescindível na indústria da construção civil e a análise e avaliação de riscos para a segurança ocupacional (AARSO) é o primeiro e fundamental passo para alcançá-la, baseado na definição e implementação de programas de prevenção. A AARSO é um processo complexo, que implica a consideração e análise de muitos parâmetros quantitativos e/ou qualitativos que são difíceis de quantificar. As metodologias AARSO utilizadas na indústria da construção civil são baseadas em informação sujeita a incerteza (sendo tratada por técnicas probabilísticas e/ou estatísticas), difusa, imprecisa e/ou incompleta. Isso implica algumas limitações, como, por exemplo, obrigar os analistas a estimar parâmetros ou efetuar comparações com outros canteiros de obras (o que afasta do sistema real em estudo). O objetivo inicial deste estudo foi efetuar a pré-validação de um método AARSO, o QRAM, em duas cidades brasileiras, de médio e grande porte.

Relevância:

80.00% 80.00%

Publicador:

Resumo:

This paper introduces a method for the supervision and control of devices in electric substations using fuzzy logic and artificial neural networks. An automatic knowledge acquisition process is included which allows the on-line processing of operator actions and the extraction of control rules to replace gradually the human operator. Some experimental results obtained by the application of the implemented software in a simulated environment with random signal generators are presented.

Relevância:

80.00% 80.00%

Publicador:

Resumo:

This work presents a procedure for electric load forecasting based on adaptive multilayer feedforward neural networks trained by the Backpropagation algorithm. The neural network architecture is formulated by two parameters, the scaling and translation of the postsynaptic functions at each node, and the use of the gradient-descendent method for the adjustment in an iterative way. Besides, the neural network also uses an adaptive process based on fuzzy logic to adjust the network training rate. This methodology provides an efficient modification of the neural network that results in faster convergence and more precise results, in comparison to the conventional formulation Backpropagation algorithm. The adapting of the training rate is effectuated using the information of the global error and global error variation. After finishing the training, the neural network is capable to forecast the electric load of 24 hours ahead. To illustrate the proposed methodology it is used data from a Brazilian Electric Company. © 2003 IEEE.

Relevância:

80.00% 80.00%

Publicador:

Resumo:

This paper proposes a fuzzy classification system for the risk of infestation by weeds in agricultural zones considering the variability of weeds. The inputs of the system are features of the infestation extracted from estimated maps by kriging for the weed seed production and weed coverage, and from the competitiveness, inferred from narrow and broad-leaved weeds. Furthermore, a Bayesian network classifier is used to extract rules from data which are compared to the fuzzy rule set obtained on the base of specialist knowledge. Results for the risk inference in a maize crop field are presented and evaluated by the estimated yield loss. © 2009 IEEE.