Flatness-based algebraic fault identification for a wave equation with dynamic boundary conditions

This paper presents a fault identification approach for a boundary controlled wave equation with dynamic boundary conditions. The faulty system is subject to an additive time-varying actuator fault and an unknown in-domain disturbance. These signals are assumed to be the solution of a finite-dimensi...

Full description

Saved in:
Bibliographic Details
Published inIFAC-PapersOnLine Vol. 53; no. 2; pp. 7647 - 7652
Main Authors Fischer, F., Deutscher, J.
Format Journal Article
LanguageEnglish
Published Elsevier Ltd 2020
Subjects
Distributed-parameter systems
fault diagnosis
flatness
modulating functions
motion planning
fault diagnosis
motion planning
flatness
Distributed-parameter systems
modulating functions
Online AccessGet full text

Cover

More Information
Summary:This paper presents a fault identification approach for a boundary controlled wave equation with dynamic boundary conditions. The faulty system is subject to an additive time-varying actuator fault and an unknown in-domain disturbance. These signals are assumed to be the solution of a finite-dimensional signal model so that polynomial and trigonometric faults as well as disturbances can be taken into account. By making use of integral transformations an algebraic expression is derived to obtain the fault from the known input and output in finite time. The kernels determining the integral transformations are obtained by solving the so-called kernel equations. This problem is traced back to the flatness-based realization of a setpoint change for an ODE-PDE casacade. From this, a condition for fault identification is derived. A simulation example demonstrates the proposed approach.
ISSN:2405-8963
2405-8963
DOI:10.1016/j.ifacol.2020.12.1366