Purified nonlinear guided waves through a metamaterial filter for inspection of material microstructural changes

Nonlinear guided waves (NGWs) exhibit extraordinary sensitivity to microstructural changes in materials which are often considered as damage precursors in many structural health monitoring (SHM) applications. However, these changes are usually very small at the microstructural level, thus generating weak nonlinear wave components which are prone to or even overwhelmed by other deceptive and non-damage-related nonlinear sources present in a monitoring system. This may jeopardize the practical implementation of the NGW-based SHM methodology. To tackle this problem, this paper proposes a metamaterial filter (meta-filter (MF)), in the form of a patch to be surface-mounted over the structure under inspection. The MF is designed to purify the probing waves during propagation before reaching the inspection area through tactical elimination of non-damage-related nonlinear components from the actuation. Considering different bonding conditions and the lattice symmetry of the MF, two types of bandgaps, corresponding to Bragg scattering and local resonances, alongside their respective band properties are analyzed. Time-domain finite element simulations are carried out to assess the efficacy of the MF in terms of wave purification in the context of SHM, supported by experimental validations. The designed MF is deployed in an aluminum strip to detect the material microstructural changes induced by a thermal ageing treatment. Results demonstrate the designed MF significantly enhances the detection ability of NGW-based SHM system. Both locally resonant and Bragg bandgaps can, in principle, be used for wave purification. While the former requires precise control of the bonding quality of the MF, the latter relies on symmetrical installation of the MF to ensure a reasonably wide wave filtering range. Considering variability factors in the implementation of the MF and better immunity to the nonlinearity of the bonding layers, preference is given to Bragg bandgap for NGW-based SHM applications.