Feedback control of the acoustic pressure in ultrasonic wave propagation

TL;DR

This paper develops a feedback control method for the acoustic pressure in ultrasound wave propagation modeled by the Moore-Gibson-Thompson equation, addressing challenges of singular controls and non-analytic semigroups.

Contribution

It introduces a novel control framework for the linearized PDE model using $L^2$ controls, and establishes well-posedness of the associated Riccati equations.

Findings

Successfully synthesizes feedback control for the PDE model.

Addresses singular control problems with non-standard Riccati equations.

Provides mathematical analysis of control stability and well-posedness.

Abstract

Classical models for the propagation of ultrasound waves are the Westervelt equation, the Kuznetsov and the Khokhlov-Zabolotskaya-Kuznetsov equations. The Jordan-Moore-Gibson-Thompson equation is a prominent example of a Partial Differential Equation (PDE) model which describes the acoustic velocity potential in ultrasound wave propagation, where the paradox of infinite speed of propagation of thermal signals is eliminated; the use of the constitutive Cattaneo law for the heat flux, in place of the Fourier law, accounts for its being of third order in time. Aiming at the understanding of the fully quasilinear PDE, a great deal of attention has been recently devoted to its linearization -- referred to in the literature as the Moore-Gibson-Thompson equation -- whose mathematical analysis is also of independent interest, posing already several questions and challenges. In this work we consider and solve a quadratic control problem associated with the linear equation, formulated consistently with the goal of keeping the acoustic pressure close to a reference pressure during ultrasound excitation, as required in medical and industrial applications. While optimal control problems with smooth controls have been considered in the recent literature, we aim at relying on controls which are just $L^2$ in time; this leads to a singular control problem and to non-standard Riccati equations. In spite of the unfavourable combination of the semigroup describing the free dynamics that is not analytic, with the challenging pattern displayed by the dynamics subject to boundary control, a feedback synthesis of the optimal control as well as well-posedness of operator Riccati equations are established.