Gå til hovedinnhold

Mulugeta Weldezgina Asres

Researcher

Research Fellow

 
Kontor:
A03205 ( Jon Lilletuns vei 9, Grimstad )
Kontortid:
10:00 - 13:00

I am a Ph.D. researcher on large-scale contextual time-aware anomaly detection and prediction systems for CERN CMS HCAL detector since 2020.

I received his B.Sc. and M.Sc. in Computer Engineering at EiT-M, Mekelle University, Ethiopia, with honors of Gold-Medal Award, in 2012 and 2016, respectively. Moreover, I conducted my Master's thesis and post-graduate research on machine learning for non-intrusive load monitoring (NILM) of complex systems at Midori Srl, an innovative start-up in energy efficiency incubated in I3P, Italy. 

Furthermore, I am a former academic member of Mekelle University (2012-2017) and Polytechnic of Turin, Italy, (2018-2020). I researched and developed various AI-powered industrial data science and monitoring applications for energy analysis, telecommunication, automotive, software computational energy consumption and natural language processing.

My research interest revolves around data science, machine learning, and deep leaning for data-driven modeling of time-series, anomaly detection, IoT, and Industry 4.0.

Forskning

Multivariate Time-Series, Large Datasets, Anomaly Detection, Data Science, Data-Driven Models, Deep Learning, IoT, and Industry 4.0

Arbeidserfaring

Graduate Research Fellow at Center of Artificial Intelligence Research (CAIR), Norway
2020 -

  • Development of a large scale contextual time-aware AD system for CERN detectors.

Research Fellow at Politecnico di Torino, Turin Area, Italy

2018 - 2020

  • R & D on smart-data automation by applying explorative, descriptive, and predictive data analytics using AI for Industry 4.0.
  • Data analytics and machine learning for telecom monitoring system for Telecom Italia (TIM);
  • Predictive energy modeling in vehicle manufacturing factory (FCA);
  • Data analytics in NILM: anomaly detection, recommender systems

NILM Researcher and Developer at Midori srl, Turin Area, Italy

2016 - 2018

  • Developed Algorithms and Machine Learning Systems for NED (Midori's Commercial Product)
  • Developed robust NILM algorithms for domestic and commercial buildings using unsupervised machine learning and signal processing tools.
  • Developed an efficient event-based load disaggregation system with a robust capability for diverse appliance switching transients especially long transients from ACs and overlapped events which are challenging to capture using conventional detectors;
  • Devised an effective mechanism that solves dynamic frequent switching event overlapping challenges from Washing Machines, Microwave, SASD, etc using unsupervised signal analysis;
  • Researched on Non-Intrusive Energy Disaggregation for complex scenarios such as commercial utilities and low voltage substations

Lecturer and Researcher at Ethiopian Institute of Technology - Mekelle, Ethiopia
2013 - 2017

  • Delivered lectures and supervised interns and undergraduate projects.
  • Assisted postgraduate lab courses and projects.
  • Manage and develop applied machine learning research and projects.
  • Freelance software, web, and embedded system developer.

Faglige interesser

R & D

Utvalgte publikasjoner

Asres, MW, Ardito, L., & Patti, E. (2021). Computational Cost Analysis and Data-Driven Predictive Modeling of Cloud-based Online NILM Algorithm. IEEE Transactions on Cloud Computing, (01), 1-1.

Asres, MW, Mengistu, MA, Castrogiovanni, P., Bottaccioli, L., Macii, E., Patti, E., & Acquaviva, A. (2020). Supporting Telecommunication Alarm Management System With Trouble Ticket Prediction. IEEE Transactions on Industrial Informatics17 (2), 1459-1469.

Asres, MW, Girmay, AA, Camarda, C., & Tesfamariam, GT (2019). Non-intrusive load composition estimation from aggregate ZIP load models using machine learning. International Journal of Electrical Power & Energy Systems105, 191-200.

Mali, D., Weldezgina, M., Birhanu, A., & Gebrehiwot, H. (2014). Simulation of GSM Based Home Security and Control System. Wireless Communication, 6 (1), 7-11.

Vitenskapelige publikasjoner

  • Asres, Mulugeta Weldezgina; Omlin, Christian Walter Peter; Dittmann, Jay; Parygin, Pavel; Hiltbrand, Joshua; Cooper, Seth I.; Cummings, Grace; Yu, David (2024). Lightweight Multi-System Multivariate Interconnection and Divergence Discovery. 2024 19th Annual System of Systems Engineering Conference (SoSE). ISBN: 979-8-3503-6591-7. IEEE conference proceedings. Chapter. s 36 - 43.
  • Asres, Mulugeta Weldezgina; Omlin, Christian Walter Peter; Wang, Long; Yu, David; Parygin, Pavel; Dittmann, Jay; Karapostoli, Georgia; Seidel, Markus; Venditti, Rosamaria; Lambrecht, Luka; Usai, Emanuele; Ahmad, Muhammad; Menendez, Javier Fernandez; Maeshima, Kaori (2023). Spatio-Temporal Anomaly Detection with Graph Networks for Data Quality Monitoring of the Hadron Calorimeter. Sensors. ISSN: 1424-8220. 23 (24). doi:10.3390/s23249679.
  • Asres, Mulugeta Weldezgina; Cummings, Grace; Khukhunaishvili, Aleko; Parygin, Pavel; Cooper, Seth I.; Yu, David; Dittmann, Jay; Omlin, Christian Walter Peter (2022). Long Horizon Anomaly Prediction in Multivariate Time Series with Causal Autoencoders. Proceedings of the European Conference of the Prognostics and Health Management Society (PHME). ISSN: 2325-016X. 7 (1). s 21 - 31. doi:10.36001/phme.2022.v7i1.3367.
  • Asres, Mulugeta Weldezgina; Cummings, Grace; Parygin, Pavel; Khukhunaishvili, Aleko; Toms, Maria; Campbell, Alan; Cooper, Seth I.; Yu, David; Dittmann, Jay; Omlin, Christian Walter Peter (2021). Unsupervised Deep Variational Model for Multivariate Sensor Anomaly Detection. 2021 IEEE International Conference on Progress in Informatics and Computing (PIC). ISBN: 978-1-6654-2655-8. IEEE conference proceedings. Chapter. s 364 - 371.
  • Asres, Mulugeta Weldezgina (2023). AI for Policing-Aware Utility-Preservation on CCTV Anonymization.
  • Asres, Mulugeta Weldezgina; Wang, Long; Yu, David; Omlin, Christian Walter Peter (2022). Spatio-Temporal Anomaly Detection for the DQM of the CMS Experiment via Graph Networks.
  • Asres, Mulugeta Weldezgina; Lei, Jiao (2022). Anonymization for CCTV footage: Challenges and Possible Solutions.

Sist endret: 27.01.2022 10:01