Echo State Networks: analysis, training and predictive control
For researchers and practitioners in nonlinear system identification and control, this work provides theoretical guarantees and practical algorithms for ESN-based predictive control, though it is incremental.
The paper theoretically analyzes Echo State Networks (ESNs), proposes a modified training algorithm for dimensionality reduction, and designs a model predictive controller for tracking. Numerical results on a pH neutralization process confirm the effectiveness of the proposed methods.
The goal of this paper is to investigate the theoretical properties, the training algorithm, and the predictive control applications of Echo State Networks (ESNs), a particular kind of Recurrent Neural Networks. First, a condition guaranteeing incremetal global asymptotic stability is devised. Then, a modified training algorithm allowing for dimensionality reduction of ESNs is presented. Eventually, a model predictive controller is designed to solve the tracking problem, relying on ESNs as the model of the system. Numerical results concerning the predictive control of a nonlinear process for pH neutralization confirm the effectiveness of the proposed algorithms for the identification, dimensionality reduction, and the control design for ESNs.