Determining dissipation inequalities from input-output samples

Knowledge of a dissipation inequality can disclose valuable system properties relevant for controller design. Therefore, we consider the problem of determining a dissipation inequality for an input-output system. In practice, yet, the core issue is that one would only know finitely many input-output samples while the input-output relationship would be undisclosed analytically. For such scenarios, we provide an approach for computing dissipation inequalities based upon knowledge of large amounts of input-output samples, as they arise in applications nowadays gathered under the term "big data". Our approach will, under certain conditions, provide dissipation inequalities which remain satisfied for all input-output pairs that the system can produce, though only having been derived from finitely many of them. (C) 2017, IFAC (International Federation of Automatic Control) Hosting by Elsevier Ltd. All rights reserved.