Selective source reduction to identify masked sources using time reversal acoustics

Abstract

The presence of strong sources of elastic waves often makes it impossible to localize weaker ones, which are sometimes the most meaningful, e.g. in the characterization of complexity of active Earth faults or of microdamage in a composite structural material. To address this problem, a selective source reduction method is proposed here which, applied in conjunction with time reversal acoustics (TRA), provides the means to selectively reduce the contribution of strong sources allowing full illumination of the weak ones. The method is complementary to other methods based on TRA which aim at the selective illumination of scatterers in the propagation medium. In this paper, a description of the method is given along with presentation of a few numerical results to demonstrate its usefulness for localization of sources. Validation and some experimental results are also presented.

Experimentally identifying masked sources applying time reversal with the selective source reduction method

“…This paper has provided important experimental verifications of the SSR method using TR, which was proposed by Scalerandi et al 9 It has been shown that a weaker secondary source, which is masked by a stronger dominant source, may be illuminated, by employing the SSR method, based on subtraction of the dominant source direct arrival and coda. SSR has application in the detection of complex microcrack distributions in mechanical parts ͑NDE of solid materials͒ and possibly in the spatial/temporal characterization of complex earthquake sources.…”

“…Note that this remnant of focusing at S 1 also occurs in modeling experiments ͓see Figs. 6͑a͒, 7͑a͒, 9͑a͒, and 10͑a͒ in the paper by Scalerandi et al 9 ͔. The fact that the remnant is also manifested in modeling results, where the transducers were not included in the model, suggests that an imperfect TR reconstruction can also result in the focusing remnant at S 1 .…”

“…In the case of a strong source whose TR reconstruction masks the TR reconstruction of a smaller source, no evidence of focusing at S 2 is apparent and the reconstruction w 2 ͑t͒ is not possible. Then, as Scalerandi et al 9 proposed in Sec. VA of their paper, the selected focal signal is time reversed and windowed such that only the focal waveform remains, w 1 Ј͑t͒.…”

“…9 This method can be successfully accomplished with a single receiving transducer, although additional receivers should improve the results. In the standard implementation of TR, a source or sources emit forward signals, which propagate through the medium and are detected, time reversed, and sent back to the source͑s͒ from a set of receivers/transducers placed in the specimen, referred to as the TR mirror ͑TRM͒.…”

“…In this manner the weaker source can be characterized. The theory of SSR has been presented by Scalerandi et al, 9 including numerical simulations and a preliminary controlled experimental verification to illustrate the procedure. However, in that work, knowledge of the two source locations was available a priori.…”