Projects
Beamforming usando ondas ultrasónicas en guías mecánicas
Beamforming is an algorithmic method for combining information from an array of sensors to identify the orientation and distance of an acoustic source of interest. The source can be active or passive. It is active, for example, when the source emits or sends energy into the observation space, such as the sound generated by an aircraft or the noise generated by a turbine at some point in the structure. A passive source is when the source does not generate energy; instead, it reflects energy produced by an additional source, such as objects illuminated by acoustic waves. For example, a robot searching for an object in low-light environments or, in the case of interest here, locating discontinuities (illuminated by acoustic actuators) in plates, pipes, or engineering structures. A recent publication on this topic is by Fernandez-Ramirez, Baltazar & Kim (2020).
Ultrasonic assessment of damage in mechanical guides such as structures in the form of pipes or plates.
Propagation of acoustic waves in pipes is a topic that has fascinated the scientific and engineering communities. The former are intrigued by its complexity as a physical phenomenon, while the latter are interested in its potential application in solving the problem of damage detection. This has economic implications and, in many cases, implications for human lives. For instance, in hydrocarbon transmission lines, where failures due to wear, fatigue, etc., could have serious consequences. A recent publication on this topic is by Guerra-Bravo & Baltazar (2023).
Evaluación por ultrasonidos de daño en guías mecánicas tales como estructuras en forma de tuberías o placas
Non-stationary signals are signals that vary over time and represent various phenomena in the real world, such as biomedical signals, vibratory machinery, and acoustic signals, among others. Their accurate classification is crucial for improving diagnostic capabilities and predicting critical events. A widely used method for their characterization is based on time-frequency mapping techniques. However, this approach introduces redundant information and complicates the extraction of relevant features for accurate classification.
We are studying the development of robust classification methods for non-stationary signals using time-frequency maps. Some of the techniques developed are based on Singular Value Decomposition (SVD) and Principal Component Analysis (PCA) for dimensionality reduction, clustering, and pattern recognition. An example of the studied methods was presented at the recent QNDE2023 Congress (YouTube video by Esteban Guerra, a doctoral student).
​
We are always looking for motivated graduate and undergraduate students who enjoy developing experimental prototypes and numerical analysis to join our laboratory. If you are interested, please send me an email or stop by my office. Graduate students may be eligible for a scholarship through the National Council of Science and Technology, and tuition fees may be waived. Both national and international students are invited to participate. Contact us