Lab of Complex Systems Modeling and Control seminar: "Emergence, Mitigation and Prediction of Extreme Events"

Laboratory of Complex Systems Modeling and Control will have the next online seminar on Friday, December 8, 13:00.
Zoom link

Speaker: Dr. S. Sudharsan, National Post-Doctoral Fellow, Physics and Applied Mathematics Unit, Indian Statistical Institute, Kolkata

Extreme events are rare and recurrent events occurring in nature that leads to disastrous aftermath. Examples of these events include rogue waves, drought, epileptic seizures, share market crash and so on. In this talk, I will confine my presentation on the following three main aspects in the study of extreme events, namely (i) Emergence and Mechanism, (ii) Mitigation and (iii) Prediction, in a parametrically driven non-polynomial mechanical system, which describes the motion of a particle in a rotating parabola. Under such consideration, without the influence of external forcing, we found that extreme events emerge differing from that of the previous case. Here extreme events emerge at the regions where chaotic attractor is found to expand and contract alternatively. From this, we have shown that extreme events occur whenever the system experiences a dip in the velocity.In the second part, I will discuss about the mitigation strategies to control extreme events.  To obtain a complete suppression of extreme events, we have effectively utilized the non-feedback methods, namely (i) first external forcing, (ii) second external forcing and (iii) constant bias to completely suppress extreme events. In the final part, I will discuss about the prediction of these extreme events using learning based approach. We have carried out two major types of ML studies, namely (i) Regression and (ii) Classification.

References:

- Sudharsan, S., Venkatesan, A., Muruganandam, P and Senthilvelan, M, Emergence and mitigation of extreme events in a parametrically driven system with velocity-dependent potential, Eur. Phys. J. Plus 136, 129 (2021).
- Meiyazhagan, J., Sudharsan, S., & Senthilvelan, M, Model-free prediction of emergence of extreme events in a parametrically driven nonlinear dynamical system by deep learning, Eur. Phys. J. B 94, 156 (2021)
- Meiyazhagan, J., Sudharsan, S., Venkatesan, A. and Senthilvelan, M., Prediction of Occurrence of Extreme Events using Machine Learning, Eur. Phys. J. Plus 137, 16 (2022).