Projects

24/7 Carbon-Free Electrified Campus Bus Fleet

• Stanford has completed the transition to 100 percent renewable electricity in March 2022, with on- and off-campus renewable electricity generation exceeding campus consumption on an annual basis. However, the campus is still plugged into the grid which carries carbon-based electricity. To completely eliminate emissions, Stanford’s next challenge is to match its electricity consumption with carbon-free resources at all hours of the year. To begin, this project will demonstrate a 24/7 Carbon-Free Electrified Stanford Campus Fleet by developing a scalable platform that intelligently coordinates solar, storage, electric bus route assignments, and bus charging. Additionally, the platform also optimizes for minimizing energy costs, providing system resiliency, and uniformly distributing vehicle utilization. This platform also serves as the first transferable model for bus fleets in school districts, corporate campuses, and municipalities to become 24/7 carbon-free.

  2022 - Current
  Learn More
  Project outcomes: 1 conference paper (ICAE 2023)
  

Through-Tubing Logging Tool for Cement Quality Evaluation in Multistring Wells

• Plug & abandonment operations are an essential step in the life of an oil well. Focusing on these operations, this project aimed to propose a tool and/or method to assess cement quality through oil wells composed of multiple casing strings. The project used different approaches, such as exploring experimental laboratory benches, investigating numerical simulations in different relevant finite element method software, and using analytical solutions jointly with machine learning adoption.

  2019 - 2022
  Project outcomes: 2 journal papers (Geoenergy Science and Engineering), 1 conference paper (Rio Oil & Gas Expo 2020) and 3 pending patents (BR1020220146888, BR1020220146861, and WO2023039653A1)
  

Structural Health Monitoring based on guided waves and machine learning

• Ultrasonic guided waves are one of the most common inputs of Structural Health Monitoring (SHM) systems. When coupled with machine learning, the mentioned input can produce satisfactory results when diagnosing faults in different entities. This project could use such approaches in many materials and various feature extraction strategies.

  2019 - 2022
  Project outcomes: 2 journal papers (IEEE Access 2021 and Measurement 2022) and 1 conference paper (SSCI 2020)
  

Machine Learning applied to sports applications

• This study aimed to associate different biomechanical characteristics with 15 m turning performance while performing different backstroke-to-breaststroke turning techniques. Only tree-based models were addressed, and the importance of each of the features was assessed.

  2020 - 2021
  Project outcomes: 1 journal paper (Sports Biomechanics 2021)
  

Detection and Classification of Faults in Aeronautical Gas Turbine Engines

• Using different types of fuzzy classifiers the main goal of this project was to detect and classify faults in simulated aeronautical gas turbine engines states. Data were obtained from the software Propulsion Diagnostic Method Evaluation Strategy (ProDiMES) developed by the National Aeronautics and Space Administration (NASA).

  2017 - 2018
  Project outcomes: 1 journal paper (Journal of the Brazilian Society of Mechanical Sciences and Engineering 2019) and 1 conference paper (WCCI 2018)