Dr Felipe Campelo

Lecturer in Computer Science

Email: f.campelo@aston.ac.uk

Phone: 0121 204 3845

Room: MB214M

 

Profile

I joined Aston’s School of Engineering and Applied Science as a Lecturer in Data Mining in February 2019. Prior to that I spent 8 years as an Assistant (and then Associate) Professor in Optimisation and Computational Intelligence for the Department of Electrical Engineering at Universidade Federal de Minas Gerais (UFMG) in Belo Horizonte, Brazil, where I served as Deputy Head of Department between 2013 and 2017.

During my time at UFMG I had the opportunity to participate in a number of university-industry partnerships, particularly related to the development of Smart Grid applications for the largest power distribution utility in South America, CEMIG-D; as well as for the power systems conglomerate Eletrobras. These applications leveraged the power of data science, computational intelligence, and optimisation to improve efficiency in the response to fault scenarios, as well as in the strategic asset management of electric power equipment.

I have previously taught a variety of modules, including basic Electrical Engineering (DC and AC Circuit Analysis, Measurement Systems, etc.), Linear and Nonlinear Optimisation, Systems Reliability, and Statistical Design and Analysis of Experiments.

My current research focuses on the development of integrated solution frameworks for data analytics, seamlessly connecting data mining, statistical modelling, optimisation and multi-criteria decision making. I also focus on the development of methodologically and statistically sound protocols for the experimental comparison of algorithms; and on the development of algorithms for robust optimisation, multi/many-objective optimisation, and heavily constrained optimisation.

 

  • BSc in Electrical Engineering, Universidade Federal de Minas Gerais, Brazil, 2003
  • MSc in Information Science and Technology, Hokkaido University, Japan 2006
  • PhD in Systems Science and Informatics, Hokkaido University, Japan, 2009
  • 2019 - date: Lecturer in Data Mining, Department of Computer Science, School of Engineering and Applied Science, Aston University. 
  • 2018 - 2019: Associate Professor of Optimisation and Computational Intelligence, Department of Electrical Engineering, Universidade Federal de Minas Gerais, Brazil.
  • 2010 - 2017: Assistant Professor of Optimisation and Computational Intelligence, Department of Electrical Engineering, Universidade Federal de Minas Gerais, Brazil. 
  • Statistical Design and Analysis of Experiments: 60 hours course offered 2x/year between 2011 and 2018 for the Graduate Program in Electrical Engineering, Universidade Federal de Minas Gerais, Brazil. All course materials (lecture slides, example code, demos etc. – all in English) are freely available online: https://git.io/v3Kh8
  • Data Mining: my current teaching responsibilities (2019 - date) include the Data Mining modules for both the undergraduate and M.Sc. Computer Science students at EAS. These modules explore both the theoretical and applied aspects of data mining, including data pre-processing, visualisation and modelling (e.g., for classification, regression or clustering) – in short, the process to extract meaningful information from existing data sets.

My current research focuses on the development of integrated solution frameworks for data analytics, seamlessly connecting data mining, statistical modelling, optimisation and multi-criteria decision making. I also focus on the development of methodologically and statistically sound protocols for the experimental comparison of algorithms; and on the development of algorithms for robust optimisation, multi/many-objective optimisation, and heavily constrained optimisation.

Some of my research, particularly the development of integrated solution frameworks, has a strong focus on applications. Past examples include partnerships with the power distribution sector to develop solutions related to the Smart Grid, such as:

  • Strategic Asset Management of High-Power Transformer Fleets, performed in collaboration with the largest power distribution company in Latin America, CEMIG-D. This work involved the development of solution pipelines integrating data retrieval and consolidation, statistical modelling of reliability indicators, multi-objective optimisation, and multi-criteria decision making, to generate optimised policies to simultaneously minimise maintenance budgets and the expected costs due to transformer failures.
  • Integrated Fault Management System, performed in collaboration with Brazilian power systems conglomerate Eletrobras. This work involved the development of automated tools for addressing scenarios with multiple simultaneous faults in the power distribution system (e.g., in case of severe storms). The resulting system seamlessly integrated fault identification and isolation, network reconfiguration (coordinating remotely and manually activated switches), and real-time maintenance team routing, to minimize off-time for customers and costs for the utility.
It is important to highlight that the possible applications of my research are not limited to the Smart Grid, but can be easily generalised to different services and industries – essentially any application in which data-driven optimisation and decision making is desired.
IEEE, Association for Computing Machinery (ACM)

Some selected recent publications include:

  • F. Campelo, F. Takahashi: “Sample size estimation for power and accuracy in the experimental comparison of algorithms”. Journal of Heuristics 25(2):305-338, 2019.
  • A.L. Maravilha, E.G. Carrano, F. Campelo: "A recombination-based matheuristic for mixed integer programming problems with binary variables". International Transactions in Operational Research, 2019 (accepted)
  • F. Goulart, A.L. Maravilha, E.G. Carrano, F. Campelo: “Permutation-based optimization for the load restoration problem with improved time estimation of maneuvers”, International Journal of Electrical Power & Energy Systems 101:339-355, 2018.
  • A.L. Maravilha, F. Goulart , E.G. Carrano, F. Campelo: "Scheduling maneuvers for the restoration of electric power distribution networks: Formulation and heuristics". Electric Power Systems Research 163(A):301-309, 2018
  • F. Goulart, S.T. Borges, F. Takahashi, F. Campelo: "Robust multiobjective optimization using regression models and linear subproblems". Genetic and Evolutionary Computation Conference (GECCO '17), pp. 569-576, 2017
  • F. Campelo, L.S. Batista, R.H.C. Takahashi, H.E.P. Diniz, E.G. Carrano: "Multicriteria transformer asset management with maintenance and planning perspectives". IET Generation, Transmission & Distribution 10(9):2087 - 2097, 2016
  • M.A.M. Teixeira, F. Goulart, F. Campelo: "Evolutionary Multiobjective Optimization of Winglets". Genetic and Evolutionary Computation Conference (GECCO '16), pp. 1021-1028, 2016.
  • F. Goulart, F. Campelo: "Preference-guided evolutionary algorithms for many-objective optimization". Information Sciences 329:236-255, 2016