Control-aware echo state networks (Ca-ESN) for the suppression of extreme events

TL;DR

This paper introduces the control-aware echo state network (Ca-ESN), a novel neural network approach that combines ESNs with control strategies to effectively suppress extreme events in chaotic systems, significantly outperforming traditional methods.

Contribution

The paper presents a new Ca-ESN framework that integrates ESNs with control techniques for the suppression of extreme events in nonlinear chaotic systems.

Findings

Reduced extreme event occurrence by two orders of magnitude

Effective suppression of chaotic-turbulent flow extreme events

Demonstrated superiority over traditional control methods

Abstract

Extreme event are sudden large-amplitude changes in the state or observables of chaotic nonlinear systems, which characterize many scientific phenomena. Because of their violent nature, extreme events typically have adverse consequences, which call for methods to prevent the events from happening. In this work, we introduce the control-aware echo state network (Ca-ESN) to seamlessly combine ESNs and control strategies, such as proportional-integral-derivative and model predictive control, to suppress extreme events. The methodology is showcased on a chaotic-turbulent flow, in which we reduce the occurrence of extreme events with respect to traditional methods by two orders of magnitude. This works opens up new possibilities for the efficient control of nonlinear systems with neural networks.