Gustavo Melo

Mr. Gustavo Melo

EDGE, Computing Institute, A. C. Simões Campus, Federal University of Alagoas—UFAL,, Maceió, Alagoas, Brazil
orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
10 May 2021 in Sensors
Verified
Journal Article
Open access
PDF file
GMGustavo Melo
Gustavo Melo
Igor TorresIgor Torres
Igor Torres
ÍAÍcaro Bezerra Queiroz de Araújo
Ícaro Bezerra Queiroz de Araújo
DBDavi Brito
Davi Brito
EBErick de Andrade Barboza
Erick de Andrade Barboza
Monitoring and data acquisition are essential to recognize the renewable resources available on-site, evaluate electrical conversion efficiency, detect failures, and optimize electrical production. Commercial monitoring systems for the photovoltaic system are generally expensive and closed for modifications. This work proposes a low-cost real-time internet of things system for micro and mini photovoltaic generation systems that can monitor continuous voltage, continuous current, alternating power, and seven meteorological variables. The proposed system measures all relevant meteorological variables and directly acquires photovoltaic generation data from the plant (not from the inverter). The system is implemented using open software, connects to the internet without cables, stores data locally and in the cloud, and uses the network time protocol to synchronize the devices’ clocks. To the best of our knowledge, no work reported in the literature presents these features altogether. Furthermore, experiments carried out with the proposed system showed good effectiveness and reliability. This system enables fog and cloud computing in a photovoltaic system, creating a time series measurements data set, enabling the future use of machine learning to create smart photovoltaic systems.

25 January 2023 in Sensors
Verified
Journal Article
Open access
PDF file
CACarlos Torres Andrade
Carlos Torres Andrade
GMGustavo Melo
Gustavo Melo
TVTiago Vieira
Tiago Vieira
ÍAÍcaro Bezerra Queiroz de Araújo
Ícaro Bezerra Queiroz de Araújo
AMallan martins
allan martins
ASAlana Kelly Xavier Santos
Alana Kelly Xavier Santos
The use of models capable of forecasting the production of photovoltaic (PV) energy is essential to guarantee the best possible integration of this energy source into traditional distribution grids. Long Short-Term Memory networks (LSTMs) are commonly used for this purpose, but their use may not be the better option due to their great computational complexity and slower inference and training time. Thus, in this work, we seek to evaluate the use of neural networks MLPs (Multilayer Perceptron), Recurrent Neural Networks (RNNs), and LSTMs, for the forecast of 5 min of photovoltaic energy production. Each iteration of the predictions uses the last 120 min of data collected from the PV system (power, irradiation, and PV cell temperature), measured from 2019 to mid-2022 in Maceió (Brazil). In addition, Bayesian hyperparameters optimization was used to obtain the best of each model and compare them on an equal footing. Results showed that the MLP performs satisfactorily, requiring much less time to train and forecast, indicating that they can be a better option when dealing with a very short-term forecast in specific contexts, for example, in systems with little computational resources.

10 April 2024 in Sensors
Verified
Journal Article
Open access
PDF file
WTWinnie de Lima Torres
Winnie de Lima Torres
LSLilian Giselly Pereira Santos
Lilian Giselly Pereira Santos
GMGustavo Melo
Gustavo Melo
AOAndressa Martins Oliveira
Andressa Martins Oliveira
PNPedro Henrique de Brito Nascimento
Pedro Henrique de Brito Nascimento
ÍAÍcaro Bezerra Queiroz de Araújo
Ícaro Bezerra Queiroz de Araújo
In recent decades, technological advancements have transformed the industry, highlighting the efficiency of automation and safety. The integration of augmented reality (AR) and gesture recognition has emerged as an innovative approach to create interactive environments for industrial equipment. Gesture recognition enhances AR applications by allowing intuitive interactions. This study presents a web-based architecture for the integration of AR and gesture recognition, designed to interact with industrial equipment. Emphasizing hardware-agnostic compatibility, the proposed structure offers an intuitive interaction with equipment control systems through natural gestures. Experimental validation, conducted using Google Glass, demonstrated the practical viability and potential of this approach in industrial operations. The development focused on optimizing the system’s software and implementing techniques such as normalization, clamping, conversion, and filtering to achieve accurate and reliable gesture recognition under different usage conditions. The proposed approach promotes safer and more efficient industrial operations, contributing to research in AR and gesture recognition. Future work will include improving the gesture recognition accuracy, exploring alternative gestures, and expanding the platform integration to improve the user experience.

27 November 2020 in Anais Congresso Brasileiro de Energia Solar - CBENS
Verified
Conference Paper
Open access
PDF file
IBIcaro Gabriel Paiva Bastos
Icaro Gabriel Paiva Bastos
GMGustavo Melo
Gustavo Melo
ÍAÍcaro Bezerra Queiroz de Araújo
Ícaro Bezerra Queiroz de Araújo
EBErick de Andrade Barboza
Erick de Andrade Barboza
TVTiago Vieira
Tiago Vieira
DBDavi Brito
Davi Brito
A natureza estocástica das condições climáticas, tais como radiação solar, temperatura ambiente, velocidade do vento, influencia diretamente a geração de energia fotovoltaica. Em decorrência, estimar a quantidade de energia gerada é de suma importância para a qualidade da produção, de maneira a evitar problemas na rede elétrica. Este trabalho apresenta uma aplicação de redes LSTM (Long Short-Term Memory) para previsão de dados de geração (tensão e corrente) em painéis fotovoltaicos a partir da utilização de dados meteorológicos. Os dados foram coletados a partir de uma estação solarimétrica de baixo custo próxima a usina de produção. As informações obtidas pela estação incluem: radiação solar, temperatura dos painéis, temperatura ambiente, umidade, velocidade do vento e quantidade de chuva. Estas informações são utilizadas para alimentar uma rede neural do tipo LSTM cuja função é a predição dos dados de corrente e tensão produzidos pelos painéis solares. Os resultados mostraram que a rede LSTM apresenta bons valores de estimação, tendo um erro médio absoluto menor do que 1%, tanto para a corrente quanto para a tensão.

10 June 2026 in IEEE Access
Verified
Journal Article
Open access
GMGustavo Melo
Gustavo Melo
PNPedro Henrique de Brito Nascimento
Pedro Henrique de Brito Nascimento
ÍAÍcaro Bezerra Queiroz de Araújo
Ícaro Bezerra Queiroz de Araújo
AMallan martins
allan martins
This paper presents a proof-of-concept implementation of a Python virtual machine executed directly on an FPGA using a dedicated, synthesizable microarchitecture described in Verilog. Unlike high-level synthesis (HLS) flows or hardware-oriented domain-specific languages, the proposed approach consumes standard CPython-generated bytecode, preserving Python execution semantics without requiring hardware description language expertise from the developer. The virtual machine follows a stack-based execution model and implements a subset of CPython opcodes, including arithmetic operations, variable storage, and function invocation through built-in calls. A Python-based host tool is used to transmit bytecode instructions to the FPGA and to collect execution results. The proof of concept was implemented and evaluated on a Digilent Basys 3 development board featuring an AMD/Xilinx Artix-7 FPGA. Experimental results demonstrate functional correctness for the supported opcode subset and confirm the feasibility of executing Python bytecode natively in reconfigurable hardware, with performance, resource utilization, and energy consumption metrics establishing a quantitative baseline for the proposed approach.

27 November 2020 in Anais Congresso Brasileiro de Energia Solar - CBENS
Verified
Conference Paper
Open access
PDF file
GMGustavo Melo
Gustavo Melo
IBIcaro Gabriel Paiva Bastos
Icaro Gabriel Paiva Bastos
EBErick de Andrade Barboza
Erick de Andrade Barboza
ÍAÍcaro Bezerra Queiroz de Araújo
Ícaro Bezerra Queiroz de Araújo
DBDavi Brito
Davi Brito
Um dos maiores desafios de um sistema fotovoltaico é o monitoramento confiável e em tempo real da sua geração de energia e das variáveis climáticas em que está submetido. O monitoramento permite estimar com precisão o desempenho do sistema fotovoltaico e detectar falhas que podem causar danos ao próprio sistema e a rede de distribuição elétrica ao qual está conectado. Os sistemas de monitoramento comerciais geralmente são caros e não medem todos os parâmetros necessários para um monitoramento completo dos sistema fotovoltaico. Este trabalho propõe um sistema de aquisição de dados de baixo custo, com alta taxa de amostragem e com capacidade de monitorar sete variáveis meteorológicas, além da tensão e corrente gerada por sistemas de micro e minigeração fotovoltaica. Uma comparação do sistema proposto com uma estação meteorológica comercial mostrou uma alta similaridade nas medições. Contudo, o sistema proposto apresenta uma maior frequência de amostragem e menor custo, ao se considerar o mesmo conjunto de variáveis meteorológicas.