Model identification of guided wave propagation in composites structures for SHM

• Grant number: 17/11216-5
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: September 01, 2017
• End date: February 28, 2018
• Field of knowledge: Engineering - Mechanical Engineering - Mechanics of Solids
• Principal Investigator: Samuel da Silva
• Grantee: Pedro Cicilini de Nadai
• Host Institution: Faculdade de Engenharia (FEIS). Universidade Estadual Paulista (UNESP). Campus de Ilha Solteira. Ilha Solteira , SP, Brazil

Abstract

The damage detection using non-destructive tests is motivated by economic and security reasons. In particular, the use of piezoelectric actuators and sensors in structural health monitoring (SHM) has allowed in the last 20 years a wide range of applications, in particular those associated with the possibility for exciting with guided wave a structure in order to interrogate its structural state locally for Identify low-intensity incipient damage well in advance of global methods based on vibration measurements. With the measurements of these input and output signals, and knowing the wave propagation path, one can verify the existence or not of discontinuities, being able to correlate it with damages associated with stiffness, inertia or damping changes of the structures. Several algorithms were and still are proposed as a way to optimize these applications within five levels: (1) To detect a structural change associated with damage; (2) To find where the change in structure is; (3) To characterize this change; (4) To quantify the extent of this structural change; And (5) To make a prognosis to verify how this damage may be propagated before a fault occurs. In general, the levels (1), (2) and (3) could be implemented based only on an analysis of the measured signals, and the levels (4) and (5) require mathematical models, for example a physical and mathematical model approximated by finite elements or a spectral model, or a black-box model, for example an experimentally identified model, such as the case of the models autoregressive (AR). In this sense, this work of IC aims to implement and bench test a new method that was recently proposed in a master's thesis supervised by the supervisor that aims to simultaneously perform the steps (1) Detection and part of the (5) Prognosis step. The proposed method is based on extrapolating the coefficients of AR models that are identified as deviations from the experimental measurements are considered statistically significant and may be related to discontinuities from possible damages. However, this method was tested with few experiments and using only metallic structures, requiring a deeper study with better statistical definition and experimental planning. Thus, the present work should carry out a more detailed study of this method using a composite material plate with commercial sensors of the type Smart Layer that already is mounted in the laboratory of Unesp - Ilha Solteira. All signal acquisition will be done with the Acellent program and the processing will be performed in own computational routines implemented in the Octave program. As a result, the project will contribute to verify the applicability and feasibility of this method, in addition to proposing possible modifications, adaptations and even improvements in the procedure. This proposal details the background and context of the project, besides the objectives, contributions and the work plan. (AU)