Parameter identification strategy for online detection of faults in smart structures for energy harvesting and sensing

Official URL: https://dx.doi.org/10.1016/j.prostr.2020.11.036

In this work, we propose a simple computational method to detect faults in smart piezoelectric structures based on a synchronization strategy. The flexible smart structures are in general described as distributed systems governed by partial differential equations. Numerical discetization is employed to derive a reduced order model such as his dynamic response is simulated solving only ordinary differential equations. Then, the parameter identification strategy is formalized as a dynamic optimization and evolution problem through a further proper set of ordinary differential equations. Lyapunov' theorems are employed to derive an integral type identification algorithm and to ensure the convergence of the procedure. The method is suitable to assess and model nonlinearities in the response of a flexible piezoelectric smart device due to material degradation or local failure. These features are very important to detect faults in the structure and to assess the system reconfiguration properties in real time.