Research Topics

My research focuses on pattern recognition (machine and deep learning), digital signal processing, embedded systems, and instrumentation. The applications include different areas: electrical power systems, renewable energies, biomedical systems, industry, images, and videos. Here, you will find some of the most relevant aspects (codes, datasets, resources, news, etc.) of the most relevant projects I have worked on.

Electrical Power Systems

Lightning Overvoltage Monitoring:
- Description: measurement and analysis of overvoltages in a power distribution network for detecting lightning-induced events.
- Relevant Paper: link.
Nonintrusive Load Monitoring:
- Description: process for analyzing changes in the voltage and current going into a house and deducing what appliances are used in the house as well as their individual energy consumption.
- Relevant Paper: link.
- Codes: link.
- Datasets: LIT-Dataset (link).
Harmonics and Interharmonics Estimation:
- Description: estimation of harmonics and interharmonics of a signal mainly under noise and frequency deviations.
- Relevant Paper: link.
- Codes: link.
Waveform Analysis:
- Description: detection and classification of events in power distribution systems using waveform analysis.
- Relevant Paper: link.
Simulation in Power Systems:
- Description: automatically generate and control massive simulations of electrical systems using ATP (Alternative Transient Program).
- Relevant Paper: link.
- Codes: link.
High Impedance Fault Detection:
- Description: a high impedance ground fault results when a primary conductor makes unwanted electrical contact with a road surface, sidewalk, sod, tree limb, or with some other surface, or object which restricts the flow of fault current to a level below that reliably detectable by conventional overcurrent devices.
- Relevant Paper: link.
- Codes: link.
- Datasets: link.

Renewable Energies

Ground Resistance Measurements in Onshore Wind Farms:
- Description: solution to minimize methodological errors resulting from applying the clamp-on for grounding resistance measurements in wind farms.
- Relevant Paper: link.
- Codes: link.
PV Fault Classification:
- Description: monitoring System to measure the electrical and environmental variables of a PV plant to detect faults in the system.
- Relevant Paper: link.
- Datasets: link.
Nonintrusive Load Monitoring and PV Identification:
- Description: This work addresses the identification and classification of Distributed Generation (DG) connected to the secondary distribution network based on the Non-Intrusive Load Monitoring framework.
- Relevant Paper: link (TBA).
- Codes: link (TBA).
- Datasets: link (TBA).

Biomedical Systems

Fall Detection:
- Description: it presents the development and evaluation of a wrist-worn fall detection solution.
- Relevant Paper: link.
- Datasets: link.
sEMG Analysis:
- Description: Gesture recognition by surface electromyography (sEMG) signals.
- Relevant Paper: link.
EEG-FES-Force-MMG Closed-loop Control:
- Description: isometric neuromuscular control system of the quadriceps muscle activated by EEG. Additionally, the detection of neuromuscular fatigue through the mechanomyography (MMG) technique is proposed, which is used to shut-off the system.
- Relevant Paper: link.
- Codes: link.

Computer Vision

Anomaly Detection:
- Description: it uses a Convolutional Autoencoder in the anomaly detection context, by applying the reconstruction error of each frame as an anomaly score.
- Relevant Paper: link.
Open World Recognition:
- Description: Open-World Recognition is an important field of research that strives to develop machine learning models capable of identifying and learning new classes as they appear.
- Relevant Paper: link.
- Codes: link.
Semantic Segmentation:
- Description: Semantic segmentation is a deep learning algorithm that associates a label or category with every pixel in an image. It is used to recognize a collection of pixels that form distinct categories.
- Relevant Paper: link (TBA).
- Codes: link (TBA).
- Datasets: link (TBA).

Embedded Systems

NOP in FPGAs:
- Description: It proposes evaluating the NOP-DH (a new inference concept based on precise notifying collaborative minimal entities) applied to speed-up the well-known Random Forest algorithm.
- Relevant Paper: link.
- Codes: link.
freePMU:
- Description: The FreePMU project delivers an open source Phasor Measurement Unit (PMU) for power system analysis based on the STM32F769 Discovery kit and an instrumentation PCB. Its characteristics are low cost, open architecture and open source code. FreePMU measures fundamental and harmonic synchrophasors as well as the signal frequency for a three-phase distribution circuit.
- Relevant Paper: link.
- Codes: link.
Smart Transformer:
- Description: transformer for the electrical energy distribution network capable of automate network reconnection processes and include communication and remote monitoring functionalities.
- Relevant information (in portuguese): link.
Safety Helmet Sensor:
- Description: The device was designed to be attached to safety helmets and used by workers during maintenance on electrical networks. It is capable of immediately alerting the professional, through an audible warning, as soon as the minimum safety distance that must be maintained in energized regions is exceeded.
- Relevant information (in portuguese): link.

Other

High-Precision RFID Indoor Location:
- Description: Object location in indoor environments is challenging when there is no physical contact, a field of view, reflective materials, and an excess of obstacles. Several works propose using Radio Frequency Identification technology (RFID) and machine learning methods to develop location systems in those situations. However, using an object as a target class slows learning and prediction in large-scale environments. To circumvent such problems, we proposed a location system that uses hierarchical classification.
- Relevant Paper: link.
- Codes: link.
Two-phase Flow Pattern Classification:
- Description: Two-phase flow is a complex phenomenon present in several industrial applications such as chemical reactors, power generation, and in the exploration, production, and transport of oil and natural gas. The classification of the flow pattern is a fundamental step in such applications, as it influences several derived parameters and sub-processes such as flow rate, void fraction, and pressure drop estimation.
- Relevant Paper: link.
- Codes: link.