Nonstationary multivariate structural damage risk evaluation method for green energy harvesters

Thorough experimental research is necessary for the development of green Energy Harvesters (EH), in addition to appropriate safety and reliability studies. The dynamic behavior of a particular galloping EH was studied through a number of wind tunnel experiments, carried out at various realistic wind speeds. Utilizing recently developed multivariate statistical approach, an excessive multivariate dynamic response of an experimental galloping EH was investigated. Majority of existing piezoelectric wind EHs, utilize a cantilever beam support structure. The effective mass, damping coefficient, and cantilever beam stiffness were determined by free vibration tests. Stationary aerodynamic loads were generated through wind tunnel experiments. This case study benchmarks recently developed design concept for the structural reliability analysis of multi-dimensional, nonstationary nonlinear dynamic systems. Presented case study demonstrates efficiency of combining recently developed multivariate Gaidai reliability method with an accurate deconvolution extrapolation scheme, allowing for accurate estimations of the likelihood of structural damage and dynamic system’s failure. The latter is a distinct practical benefit of the proposed multivariate methodology, since existing structural reliability methodologies often struggle to cope with dynamic system's high dimensionality, along with nonlinear cross-correlations across system parts/dimensions. The aim of this investigation was to establish a baseline for the proposed multivariate structural reliability concept, facilitating the effective retrieval of pertinent statistical information from the underlying measured dataset. System performance or limit state function depends on multiple random variables (e.g., load, resistance, environmental factors). Engineering relevance : assessing the reliability of high-dimensional structures where failure probabilities P F are quite low (e.g., less than 10 − 6 ). Presented study benchmarked novel multivariate structural reliability concept that is more conservative than existing univariate or bivariate reliability schemes. Proposed deconvolution extrapolation scheme enabled assessment of design valued with P F four decimal orders of magnitude lower that actually measured data permits, thus enabling device design with over 1 year service lifetime.