Issue 78

D. S. Lobanov et alii, Fracture and Structural Integrity, 78 (2026) 207-219; DOI: 10.3221/IGF-ESIS.78.12

digitization of the signal waveform without loss of information. These settings are consistent with the results and recommendations presented in [17].

Figure 3: Fracture surface of a 4 mm thick fiberglass specimen following the introduction of a simulated defect using spherical indenters until failure A discrimination threshold of 40 dB was set to suppress mechanical and electrical noise from the test equipment and the surrounding environment. To verify the selected settings before the main series of experiments, setup tests were conducted, including AE source simulation tests (ASTM E976) and idle loading without a specimen. Analysis of the recorded data confirmed that at the set threshold, background noise is reliably rejected, while signals from microfractures in the material, the amplitude of which exceeded the threshold by 10–15 dB, are recorded without loss. The methodology of selecting the discrimination threshold based on a preliminary noise level analysis is generally accepted in the practice of acoustic emission testing of composites [6,17]. Previously, in [20], the authors developed a methodology for identifying the relationship between acoustic emission signals and fracture processes. In this study, a similar methodology was applied, and an analysis was carried out on the following signal characteristics: energy parameter, peak frequencies, and the RA-AF parameter. A scheme for distributing signals across sectors was used (Fig. 4). Sector I – signals indicating delamination in the material; sector III – signals associated with brittle fracture. Near the lower boundary of sectors I and II – signals that appear during matrix fracture, following the transition from sector I to sector II, and from sector III to sector IV – signals indicating fiber cracking.

Figure 4: Signal distribution diagram in RA-AF coordinates. The dataset obtained during acoustic emission monitoring was first processed using Spearman’s method to identify correlations between signal parameters. Spearman’s rank correlation coefficient allows the strength of the relationship

210