Detecting twitch in conveyor belt steel wire rope joints using eddy current testing faces challenges such as insufficient training signals, low classification accuracy, and slow recognition. To address these issues, this paper designs a differential probe and proposes a MixUp-CS-CNN method. The finite element simulation model is validated by comparing simulated and experimental signals under different twitch distances. The MixUp method augments simulation samples, constructing a dataset of 600 groups covering six twitch distances. Fault features are extracted from time, frequency, and time-frequency domains, and the Chi-Squared test reduces dimensionality to obtain an optimal feature subset. This subset is then input into a CNN classifier. Results show a diagnostic accuracy of 98% for joint twitch. The proposed method overcomes the problems of insufficient sample data and poor generalization, while balancing detection accuracy, efficiency, and on-site adaptability through the optimized differential probe and feature reduction.
This paper investigates nondestructive evaluation (NDE) for assessing low-temperature thermal embrittlement (also known as the 475 degrees C embrittlement) in duplex stainless steel (DSS) alloys. Current assessments rely primarily on destructive methods such as the Charpy test and double-loop electrochemical potentiokinetic reactivation (DL-EPR). To date, no systematic comparative investigation has been conducted to identify reliable NDE methods for evaluating the 475 degrees C thermal embrittlement in DSS alloys. This paper reviews and appraises several candidate NDE techniques, including thermoelectric power (TEP), alternating current potential drop (ACPD), and nonlinear ultrasound (NLU), as well as ultrasonic velocity and attenuation, with respect to their ability to characterize the mechanical and corrosion property degradation of a DSS alloy. Benchmark comparisons are conducted using a set of controlled UNS S32750 super DSS samples. The advantages and drawbacks of the investigated NDE techniques are discussed in terms of measurement sensitivity, correlation with impact energy loss, and suitability for field implementation. Among the techniques examined, TEP emerges as the most viable technique, not only because of its high sensitivity (up to a 130% change) but also because of its relatively simple experimental setup. ACPD technique shows promise for continuous monitoring with a reasonable level of sensitivity (up to a 6.25% change) to Cr content. While NLU techniques also exhibit high sensitivity (up to 44% and 25% changes), it is currently uncertain to which microstructural mechanism they are sensitive, lacking the ability to isolate the thermal embrittlement effects from other concurrent microstructural changes.
In response to the need for identifying planetary gear damage in industrial robot gearboxes, this study proposes a damage feature extraction method based on acoustic emission (AE) signal analysis. Firstly, an experimental platform was established, and AE signals were collected under the condition of both normal and planetary gear damage in the gearbox. Next, the signals were decomposed into node signals at different frequency bands using a three-level wavelet packet decomposition(WPD), and the node signals containing the most significant damage-related information were selected based on kurtosis values and spectral entropy. The selected node signal was then used to reconstruct the AE signal. Subsequently, the Hilbert envelope demodulation technique was applied to extract the modulated components from the reconstructed signal, and the envelope spectrum was analyzed to reveal sideband features associated with gear damage. Finally, the relative sideband strength (RSS) was employed for quantitative analysis of the sidebands, further validating the correlation between the damage state and spectral energy distribution. Experimental results demonstrate that the proposed method shows strong accuracy and robustness in identifying planetary gear damage in industrial robot gearboxes, providing effective technical support for fault diagnosis and health monitoring of equipment.
