EST - Escola Superior de Tecnologia

URI permanente desta comunidadehttps://ri.uea.edu.br/handle/riuea/3366

Navegar

Resultados da Pesquisa

Agora exibindo 1 - 10 de 358
  • Imagem de Miniatura
    Item
    Protótipo de suspensão ativa para veículos de competição Off-Road
    (Universidade do Estado do Amazonas, 2024) SOUZA, Luiz Antonio Sobrinho de; Araújo, Rodrigo Farias; http://lattes.cnpq.br/2107906714409879; http://lattes.cnpq.br/7674140777746782; Bessa, Iury Valente de; Cordovil Júnior, Luiz Alberto Queiroz; Sicchar Vilchez, José Ruben
    The vehicle’s suspension system plays a crucial role in dealing with the contact between the vehicle and the ground. Its function encompasses keeping the tires firmly grounded on uneven terrain, ensuring the isolation of these chassis irregularities, reacting to longitudinal (acceleration and braking) and lateral (curving) forces, and even resisting chassis rolling. Therefore, it is evident that the suspension system describes a significant portion of the vehicle’s dynamics, directly influencing the drivability and comfort of the driver. This project aims to develop an active suspension prototype utilizing modern control techniques to evaluate the enhancement of an off-road vehicle’s performance in terms of rider comfort. To achieve this, a 1/4 car control plant will be built based on the reduced-scale geometry of a Baja SAE type off-road vehicle, incorporating the SAVGS actuation mechanisms concept. Electronics, DC motors, encoders, and accelerometers, will be used to control the entire system. System modeling will employ ARMAX models using system identification techniques. The robust controller to be used will be based on minimizing the H2 norm, considering D-Stability
  • Imagem de Miniatura
    Item
    Projeto e Controle de um Pêndulo Invertido Rotacional
    (Universidade do Estado do Amazonas, 2024) Casas, Vitoriano Medeiros; Araújo, Rodrigo Farias; http://lattes.cnpq.br/2107906714409879; http://lattes.cnpq.br/8832748559231137; Bessa, Iury Valente de; Condovil Júnior, Luiz Alberto Queiroz; Sicchar Vilchez, José Ruben
    The modeling and control of nonlinear systems depend on appropriate methods. A large portion of real systems are nonlinear and require methods capable of overcoming these nonlinearities so that they can be used in different applications. Linearization around the operating point is a widely used resource when you want to approximate the dynamics of the non-linear model and make use of linear controllers. The application of control to real systems also depends on the discretization of the model so that a discrete control law can be designed. This is due to the fact the controllers designed are implemented in computers which receive signals from real quantities in discrete values. This work includes the methodology for building a real dynamic system of the rotational inverted pendulum type, the description of the continuous model, the process of linearizing and discretizing the model and the design of the digital controller. The control law is found by the Lyapunov stability method using linear matrix inequalities.
  • Item
    Desenvolvimento de um sistema de simulação e comparação da resposta dinâmica em solo via Matlab de um Vant de asa fixa em voo subsônico a partir de dados coletados por um sistema embarcado
    (Universidade do Estado do Amazonas, 2023-09-23) Janzen, Renan Araújo; Oliveira, Jozias Parente de; http://lattes.cnpq.br/3571273776575544; Oliveira, Jozias Parente de
    In the present work, software algorithms were developed, associated with embedded hardware, and in Matlab with the objective of carrying out the simulation and comparative analysis of the dynamic response of an unmanned aerial vehicle based on the data acquired by the aircraft's embedded system. For this, an embedded system was used, on an aircraft in flight,which performs data acquisition, recording and transmission in real time to hardware connected to a computer on the ground. The data received by the computer is used to simulate the responses of transfer functions developed in Matlab based on linearized mathematical modeling that describes the dynamic behavior of an aircraft in straight, level flight. The complete modeling presents six degrees of freedom and describes a non-linear dynamic behavior, however, based on simplifications and assumptions that aim to reduce the computational cost and enable functional visibility, the model becomes linear and decoupled, allowing a simulation of movements longitudinal and lateral-directional of the aircraft separately. The inputs to the transfer functions consist only of the commands sent by the pilot, therefore external disturbances, such as wind gusts, are disregarded. The outputs of the transfer functions consist of the orientation of the aircraft around the x, y and z body axes, represented by the attitude, pitch and yaw angles. These same angles are acquired during the flight by an onboard inertial measurement unit, enabling a comparison between the outputs of the transfer functions and the acquired data. The results obtained were in line with the objective of developing the simulation and comparison system, however, an aircraft different from that modeled was used to carry out the flight tests. To compare the acquired and simulated data, a statistical analysis was carried out, which was only possible for the longitudinal movement, as the latero-directional transfer functions were unstable, which made the analysis unfeasible. It was visually verified that few flight stages showed similarities between the acquired and simulated data. In the statistical analysis, it was concluded that the modeling was not suitable for simulating all flight stages, which should be expected due to the conditions and assumptions related to straight and level flight that are applied during modeling. Finally, it became clear that it is necessary to limit the actual flight test condition by an indication via firmware associated with instructions given to the pilot to execute this condition.
  • Item
    Inteligência artificial para avaliação da qualidade da energia do sistema elétrico de um prédio da Universidade do Estado do Amazonas
    (Universidade do Estado do Amazonas, 2024-02-27) Almeida Junior, Vilmar Bolzan de; Río, Daniel Guzman del; http://lattes.cnpq.br/4451217067318841; Río, Daniel Guzman del; Torné, Israel Gondres; Cisneros, Edry Antônio Garcia
    This research presents a study on the application of a system based on artificial intelligence to evaluate the quality of electricity at the Escola Normal Superior. An artificial intelligence model consisting of two scripts was developed. The first script uses a fuzzy controller that employs rules in the if-then format, based on the Distribution Rules and Procedures (PRODIST) of the National Electric Energy Agency (ANEEL). The result of the first script is an assessment of the power quality as good or bad. This information is then inserted into a dataframe to provide the target for the next script. The second script is based on machine learning models. The available samples were divided into a portion of 70% for training and 30% for testing, in order to select the best classifier for future online use. The machine learning models that showed the best precision and accuracy rates were DecisionTree and RandomForest, achieving metrics of over 99%. The study was validated with system data collected by an electrical magnitude analyzer installed at the ENS. The data obtained indicates poor energy quality in the ENS building, but has enabled the creation of an artificial intelligence model that not only classifies the energy quality, but can also contribute to improving it in the future.
  • Item
    Identificação de lixo hospitalar por meio de imagens de smartphones com deep learning
    (Universidade do Estado do Amazonas, 2023-09-11) Pinheiro, Vinícius dos Santos; Figueiredo, Carlos Maurício Serodio; http://lattes.cnpq.br/0402098601975654; Figueiredo, Carlos Maurício Serodio; Pantoja, Antônio Luiz Alencar; Oliveira, Jozias Parente de
    The inadequate management of hospital waste poses a significant risk to urban environments, which has been further exacerbated by the COVID-19 pandemic. This study has created an application utilizing deep learning techniques to identify instances of improper waste disposal. A meticulously curated database was compiled, encompassing a wide variety of images from various sources. Segmentation masks were meticulously generated for each image, and data preprocessing and augmentation techniques were Applied to create a secondary dataset. Subsequently, we trained twelve distinct models using specialized techniques to enhance their performance, employing the YOLOv8 architecture for image segmentation. The final model of choice was YOLOv8x, which yielded the most promising results in key metrics such a recall rate of 0.62, an mAP50 of 0.67, and an mAP50-95 of 0.43, particularly when trained on the augmented dataset. This selected model was then deployed on a local server and seamlessly integrated with the application through asynchronous web requests. The application itself was developed using Swift, providing users with the capability to submit images captured by their câmeras for server-based analysis. The server draws masks to identify instances of improper waste disposal and returns the modified image to the user through the application.
  • Item
    Manual de instalação de sistemas fotovoltaicos conectados à rede com potência até 15 kwp
    (2024-02-19) Luz, Victor Hugo Viegas Da; Azevedo, Cláudio Gonçalves de ; Azevedo, Cláudio Gonçalves de ; Siqueira, José Gilson ; Silva, José Gomes da
    This work presents a model for an installation manual for photovoltaic systems connected to the grid with a power of up to 15 kWp, aiming to generate literature to help electrical engineers and other installers of photovoltaic systems to improve their installation process and guide them to basic information that must be followed to carry out safe and eficiente work. The manual will be prepared using as a bibliographical reference the standards that govern safety with electrical installations, installations with photovoltaic arrangements, work at heights and also manuals from manufacturers of photovoltaic components such as inverters, module fixing and support structures, photovoltaic modules and other components that can be part of a system of up to 15 kWp. Content and records obtained from both bibliographic research and photovoltaicinstallations that were carried out in the state of Amazonas will be used, considering all the main adversities found in these installations and the procedures adopted to carry out a safe and reliable installation.
  • Item
    Projeto e implantação de um sistema solar fotovoltaico e de um sistema de automação de iluminação em uma granja de pequeno porte.
    (Universidade do Estado do Amazonas, 2023-08-31) Alves, Thiago Francisco Paes; Río, Daniel Guzmán del; Río, Daniel Guzmán del; Oliveira, Jozias Parente de; Macedo, Pierre
    Searching for an applicable alternative to solve problems in the development of the interiors of the state of Amazonas related to lack of energy and technological resources. This work aims to address the design and implementation of a photovoltaic solar generator in a small farm, as well as the automation of local lighting, describing the technical procedures, electrical design and technology used, to finally perform a financial analysis of the system through economic indicators such as rate of return investment and payback to measure the real gain with the implementation of the system, resulting in a clearance of the viability of the system in a real environment.
  • Item
    Desenvolvimento de um software para auxiliar o aprendizado de conceitos e aplicações abordados na disciplina de processamento digital de imagens.
    (Universidade do Estado do Amazonas, 2023-08-31) Campos, João Vitor Da Silva; Oliveira, Jozias Parente de; Oliveira, Jozias Parente de; Figueiredo, Ingrid Sammyne Gadelha; Pantoja, Antônio Luiz Alencar
    The work presents the development of a software for learning concepts of the discipline of digital image processing, bringing in assistant professors who teach these subjects and exemplifying in a practical way the techniques taught in the classroom, facilitating and contributing to the understanding by the student. . The techniques developed in the tool are the limits of where the image is binarized from a pixel intensity value, changing the brightness and contrast in the image with visualization of their respective histograms, applications of low pass filters with Gaussian and average filters that allow smoothing the image with noise like salt and pepper, applying high pass filters with edge detection by the Canny and Sobel operators, and finally the visualization of RGB and YCbCr color spaces and in grayscale. To validate the results and the functioning of the tool, a satisfaction survey was carried out with students who attended or attended the PDI subject using the average ranking calculation.
  • Imagem de Miniatura
    Item
    Classificação de cenas aéreas em sensoriamento remoto: Uma abordagem utilizando dados de imagem e som e self-supervised learning
    (Universidade do Estado do Amazonas, 2024-02-08) Ayres, Talissa Moura; Figueiredo, Carlos Maurício Serodio; Figueiredo, Carlos Maurício Serodio; Pantoja, Antônio Luiz Alencar; Cardoso, Fábio de Sousa
    Scene classification is an activity in computer vision where models can understand a context or environment without focusing solely on classifying a single object, as in image classification. Therefore, it is an area of extensive research currently, as it is used in important tasks such as content-based retrieval and smart content moderation. Additionally, when performed with remote sensing data, it is crucial for understanding the environment around us, being applied in tasks such as city monitoring and land use classification. Emphasizing the classification of aerial scenes, many of these studies are based on using convolutional neural networks for this activity, thus relying on a large number of annotations for images. Hence, the application of new training techniques such as self-supervised learning (SSL), where the model first learns to generate representations from pseudolabels before performing the main task, has been more widely applied in recent literature. Furthermore, the possibility of using multimodal data with geolocated images and sounds to improve model performance in this task has been demonstrated through the ADVANCE and SoundingEarth datasets. Therefore, this paper demonstrates the use of SSL and audiovisual remote sensing data in conjunction with the application of vision transformers, a new deep learning architecture based on attention mechanisms, for generating embeddings. Firstly, pre-training was conducted on SoundingEarth, using batch triplet loss to bring closer pairs of positive image and sound data and separate distinct pairs. Subsequently, these representations were applied to a logistic regression model to classify aerial scenes from ADVANCE. The results obtained showed precision, recall, and F1-Score above 80% for models trained with both image and sound embeddings. Considering only image embeddings, results were also above 80%, and considering only audio, results were above 40% for these metrics.
  • Imagem de Miniatura
    Item
    Desenvolvimento de um sistema de telemetria para monitoramento de qualidade de água em rios amazônicos
    (Universidade do Estado do Amazonas, 2023-09-01) Silva, Rubem Silas Dias; Fernandes, Rubens de Andrade; Fernandes, Rubens de Andrade; Torné, Israel Gondres; Oliveira, Jozias Parente de
    The present work presents the development of a telemetry system for water quality indicators. The purpose is to enable remote, periodic, and self-sustaining monitoring of chemical and biological parameters in Amazonian rivers through precise sensors over long distances. The implementation of the proposed research began with the modeling of a microcontrolled hardware device designed for low energy consumption, featuring energy harvesting, battery backup, LoRa connectivity, and support for reading laboratory water quality measurement sensors. In addition to the envisioned hardware system, the developed firmware was designed to efficiently coordinate sensor activation, peripheral devices, and LoRa radio communication, as well as establish communication via the LoRaWAN protocol with a web application for displaying water quality information. With the implementation of the proposed telemetry device, it is expected to assess the system’s autonomy in remote locations and compare the measurements taken by the device with those recorded by the research group at the LabRios laboratory, located at the Center for Higher Studies in Parintins. This will ultimately test the viability of the developed solution as a valid tool for monitoring river parameters.