Diffused acoustic fields exist during the operation in wind turbine blades because of consistent and stochastic loads and the aerodynamic noise generated from turbulent pressure fields. The Green’s function can be extracted from an ensemble average of the cross-correlations of signals recorded from two passive sensors within the diffused acoustic field. Here the wave fields reconstructed are multimodal and dispersive, which makes their extraction challenging. The causal and anti-causal Green’s function is estimated by taking the derivative of the ensemble average of cross correlation.
This method was studied experimentally using a wind turbine test blade (CX-100) located at the UCSD’s Powell Structural Laboratories. A diffuse field was approximated by exciting the skin of the blade with a random signal at several locations using an electrodynamic shaker. The reconstructed Green’s function estimate is compared to experimentally measured impulse response functions. The use of various features of the reconstructed Green’s function for potential applications to structural health monitoring of wind turbine blades is discussed.