RESUMO

This paper presents a novel method for load torque estimation in three-phase induction motors using air gap flux measurement and the conversion of this type of time-domain signal into grayscale images for further processing as inputs for an inception-type convolutional neural network. The magnetic flux was measured employing a Hall effect sensor installed inside the machine, near the stator slots, and above the stator windings. In this case, the sensor was able to measure a resultant magnetic flux density, having both rotor and stator magnetic flux contributions. The present methodology does not require motor parameters for torque prediction. The proposed approach successfully estimated load torque using three optimizers across almost the entire motor load operational range, spanning from 1.5% to 93.9% of the rated load. Four model configurations achieved a mean absolute percentage error (MAPE) less than or equal to 3.7%. Specifically, two models for a 40 × 50 pixel image achieved MAPE of 3.7% and 3%, one model for a 40 × 25 pixel image achieved a MAPE of 3.5%, and one model for a 50 × 80 pixel image achieved a MAPE of 3.3%. This research has been experimentally validated with a 7.5 kW squirrel cage induction machine.

RESUMO

RESUMO Este artigo apresenta um modelo de rede neural para controlar a dosagem de cloro na água tratada, a fim de obter melhores resultados na estabilidade do cloro residual livre, quando comparados ao controle Proporcional-Integral-Derivativo (PID) empregado atualmente e na consequente redução do consumo de cloro no processo de tratamento de água. Entre os diferentes desinfetantes, o cloro é amplamente utilizado em estações de tratamento de água de grande porte no mundo. No Brasil, especialmente nas grandes cidades, esse produto químico é também empregado como um agente desinfetante. Nesse contexto, um modelo que utiliza redes neurais artificiais foi implementado e simulado no software Matlab, levando-se em conta as características operacionais de uma estação de tratamento localizada na região metropolitana de São Paulo.

ABSTRACT This paper presents an approach using a neural network model to control chlorine dosage in treated water, so as to obtain better results in the stability of the free residual chlorine, when compared to the control currently used, based on the Proportional-Integral-Derivative (PID) model and the consequent reduction in chlorine consumption in the water treatment process. Among the different disinfectants, chlorine is widely used in large water treatment plants in the world. In Brazil, especially in large cities, this chemical product is also used as a disinfectant agent. In this context, a neural model was implemented and simulated in MatLab software, taking into account the operational characteristics of a treatment plant, located in the São Paulo Metropolitan Region.

RESUMO

Brazil has been presenting in the last years a scientific production well-recognized in the international scenario, in several areas of knowledge, according to the impact of their publications in important events and especially in indexed journals of wide circulation. On the other hand, the country does not seem to be in the same direction regarding to the technological production and wealth creation from the established scientific development, and particularly from the applied research. The present paper covers such issue and discloses the main similarities and differences between a scientific paper and a patent application, in order to contribute to a better understanding of both types of documents and help the researchers to chose and select the results with technological potential, decide what is appropriated for industrial protection, as well as foster new business opportunities for each technology which has been created.

Assuntos

RESUMO

O Brasil vem apresentando, nos últimos anos, uma produção científica bastante reconhecida no cenário internacional, nas mais diversas áreas do conhecimento, tomando por base o impacto de suas publicações em eventos de grande monta e, especialmente, em revistas indexadas de grande circulação. De outra forma, o país parece não caminhar na mesma direção quando se trata da produção tecnológica e da geração de riqueza a partir do desenvolvimento científico estabelecido e, particularmente, a partir da pesquisa científica aplicada. O presente trabalho abordou tal questão e discorreu acerca das principais semelhanças e diferenças entre um artigo científico e o texto de um pedido de patente, a fim de contribuir para um melhor entendimento dos dois tipos de documento, auxiliar os pesquisadores na escolha e seleção dos resultados com potencial tecnológico, decidir o que é adequado para fins de proteção industrial, bem como alavancar novas oportunidades de negócio para cada tecnologia criada.

Brazil has been presenting in the last years a scientific production well-recognized in the international scenario, in several areas of knowledge, according to the impact of their publications in important events and especially in indexed journals of wide circulation. On the other hand, the country does not seem to be in the same direction regarding to the technological production and wealth creation from the established scientific development, and particularly from the applied research. The present paper covers such issue and discloses the main similarities and differences between a scientific paper and a patent application, in order to contribute to a better understanding of both types of documents and help the researchers to chose and select the results with technological potential, decide what is appropriated for industrial protection, as well as foster new business opportunities for each technology which has been created.

Assuntos

RESUMO

OBJECTIVE: To propose a measure (index) of expected risks to evaluate and follow up the performance analysis of research projects involving financial and adequate structure parameters for its development. METHODS: A ranking of acceptable results regarding research projects with complex variables was used as an index to gauge a project performance. In order to implement this method the ulcer index as the basic model to accommodate the following variables was applied: costs, high impact publication, fund raising, and patent registry. The proposed structured analysis, named here as RoSI (Return on Scientific Investment) comprises a pipeline of analysis to characterize the risk based on a modeling tool that comprises multiple variables interacting in semi-quantitatively environments. RESULTS: This method was tested with data from three different projects in our Institution (projects A, B and C). Different curves reflected the ulcer indexes identifying the project that may have a minor risk (project C) related to development and expected results according to initial or full investment. CONCLUSION: The results showed that this model contributes significantly to the analysis of risk and planning as well as to the definition of necessary investments that consider contingency actions with benefits to the different stakeholders: the investor or donor, the project manager and the researchers.

Assuntos

RESUMO

Objective: To propose a measure (index) of expected risks to evaluate and follow up the performance analysis of research projects involving financial and adequate structure parameters for its development. Methods: A ranking of acceptable results regarding research projects with complex variables was used as an index to gauge a project performance. In order to implement this method the ulcer index as the basic model to accommodate the following variables was applied: costs, high impact publication, fund raising, and patent registry. The proposed structured analysis, named here as RoSI (Return on Scientific Investment) comprises a pipeline of analysis to characterize the risk based on a modeling tool that comprises multiple variables interacting in semi-quantitatively environments. Results: This method was tested with data from three different projects in our Institution (projects A, B and C). Different curves reflected the ulcer indexes identifying the project that may have a minor risk (project C) related to development and expected results according to initial or full investment. Conclusion: The results showed that this model contributes significantly to the analysis of risk and planning as well as to the definition of necessary investments that consider contingency actions with benefits to the different stakeholders: the investor or donor, the project manager and the researchers.

Objetivos: Propor um novo índice adaptado de riscos esperados para avaliar, planejar e acompanhar projetos de pesquisa que dependem de financiamento e estrutura adequada para o seu desenvolvimento. Métodos: Uma grade de resultados aceitáveis em relação aos projetos de pesquisa com variáveis complexas foi aplicada como índice para mensurar o desempenho dos projetos. Para sua implementação foram utilizados o índice de ulcer como medida básica para acomodar as seguintes variáveis: custos, fator de impacto de publicação, levantamento de fundos e registro de patentes. Obteve-se uma ferramenta de modelagem aqui denominada "Retorno sobre Investimento Científico" (RoSI) que permitiu mensurar e acompanhar projetos de modo analítico, levando-se em conta o conjunto dos resultados esperados, as respectivas variáveis e suas interações analisadas num ambiente virtual semi-quantitativo. Resultados: O modelo foi testado para avaliação de três projetos da Instituição (projetos A, B, e C). Diferentes curvas refletem os índices de ulcer identificando o projeto de menor risco (projeto C) relacionado ao desenvolvimento e resultados esperados de acordo com o investimento inicial ou total considerado. Conclusão: O modelo contribuiu tanto para a quantificação dos riscos e planejamento, quanto para a definição de investimentos necessários que contemplem ações de contingência, beneficiando os diferentes stakeholders envolvidos: o financiador (investidor ou doador), o gestor e os pesquisadores.

Assuntos