Влияние амплитуды акустических сигналов на вероятность выявления источников акустической эмиссии

Abstract

The article discusses the results of using a standard algorithm for linearly locating the sources of acoustic emission signals generated by a broadband acoustic emission (AE) transducer mounted on the surface of a steel plate with sizes 1000×650×7 mm. To generate AE impulses with an amplitude of um = 55–100 dB, the electronic simulator’s difference of potentials was varied in the range of 10–300 V. As a result of laboratory experiments, the reduced error γ of the standard linear location algorithm was calculated. The maximum error equal to γ = 16.3% was recorded at the coordinate X = 100 mm in locating the source of acoustic signals with an amplitude of less than 60 dB and the antenna array basic size B = 800 mm. The minimum error equal to γ = 2.69% was recorded with the electronic simulator installed at the coordinate X = 400 mm. It is shown that the maximum error of the standard algorithm is observed in locating the sources of low-amplitude AE signals situated near the antenna array receiving transducers. The AE source detection probability as a function of recorded impulse amplitude is quantified. For determining the AE source detection probability p, the flow of recorded signals was divided into three amplitude ranges: 40–60 dB, 60–75 dB, and 75–100 dB. For the sources of acoustic signals with an amplitude of less than 60 dB and located at the coordinates X = 100, 200, 600, and 700 mm, the parameter p value tends to zero. It has been revealed in processing the experimental study results that the AE source detection probability increases with a growth in the maximum amplitude of the recorded signals. For AE impulses with an amplitude above 75 dB, the parameter p value approaches unity regardless of the source location. It has been determined that the error of the standard linear location algorithm depends on the distance between the AE source and the antenna array receiving transducers. The dependence p(X, um) has been demonstrated as a numerical assessment of the way in which the above-mentioned factors influence the obtained location picture results.