For the problem that the fault states of parallel shaft gearboxes are difficult to identify, a diagnostic method is proposed to optimize variational modal decomposition (VMD) and t-distributed stochastic neighbor embedding (t-SNE) using an improved dung beetle optimization algorithm I have checked and revised all. (IDBO). IDBO is obtained by amplifying dung beetle optimization (DBO) using strategies such as chaos mapping, Levy flight policy, and dynamic adaptive weighting. IDBO is employed to optimize VMD, extracting decomposed eigenvalues restructured into high-dimensional feature vectors. Subsequently, we employ the t-SNE algorithm for dimensionality reduction to eliminate redundancy, obtaining two-dimensional vectors. Finally, these vectors are input into a support vector machine (SVM) for fault diagnosis. We apply IDBO, grey wolf optimization (GWO), DBO, and the sparrow search algorithm (SSA) to both benchmark functions and VMD, conducting a performance comparison. The results demonstrate that IDBO exhibits superior convergence speed and global search capability, effectively suppressing modal aliasing issues in VMD, thereby enhancing the algorithm’s robustness. Through experimental fault diagnosis on a gear transmission system, we compare our proposed method with EMD + t-SNE and traditional VMD + t-SNE feature extraction approaches. The experimental results indicate that the fault diagnosis accuracy reaches 100% after processing the fault signals with IDBO-VMD + t-SNE. This method proves to be an effective fault diagnosis approach specifically tailored for parallel-axis gearboxes, providing a reliable means to enhance diagnostic accuracy.

With the rapid development of a high‐power scroll compressor (HSC), the dynamic performance of a scroll compressor has an important impact on the vibration and noise of the whole compressor. So, the dynamic characteristics and structural acoustic radiation characteristics of HSC under dynamic excitation are studied by simulation and experiment in this paper. Firstly, the dynamic simulation model of HSC is established to analyze the time‐domain and frequency‐domain characteristics of bearing excitation and upper support excitation. At the same time, the finite element models of housing, upper support, and lower support are constructed to analyze the transient dynamic response of housing under the excitation source. Based on the vibration displacement of the housing surface, the prediction model of the housing radiated noise is also established, and then the radiated noise of the housing surface and the noise spectrum of the standardized test points are analyzed. Finally, the accuracy of the prediction model is verified by acoustic experiments. The experimental results show that the vibration and radiation noise of the housing can be accurately predicted by reasonable modeling to calculate the gas force and approximate the actual gas load. The vibration of the housing coupling system is the main source of the vibration characteristics of the HSC. The conclusions of the study can provide a reference for the low noise design of the HSC.

In this study, a high-hardness and wear-resistant ceramic coating was prepared on the surface of 319S aluminum alloy using the micro-arc oxidation (MAO) technique. The effects of pulse width, negative voltage, and KOH concentration on the MAO coating were investigated, and the microhardness and surface roughness of the coating were measured. The morphology, elemental distribution, and phase composition of the coating were analyzed using SEM, EDS, XRD, and digital microscopy. The influence of the MAO coating on the wear of the 319S aluminum alloy was evaluated using a friction-wear tester. The results showed that in the sodium silicate solution system, with an increase in pulse width, the thickness of the coating gradually increased and the surface hardness initially increased and then decreased. With an increase in negative voltage, the density of the coating first increased and then decreased, the thickness of the dense layer initially increased and then decreased, and the surface hardness initially increased and then decreased. With an increase in the KOH concentration, the coating thickness increased and the roughness initially decreased and then increased. When the pulse width was 3000 ms, the negative voltage was 130 V, and the KOH concentration was 1 g/L, the coating exhibited the best density, with the highest surface hardness of 1426.8 HV and the thickest dense layer of 55 μm. The reduction in surface cracks and improvement in density indicated an enhancement in the hardness and wear resistance of the coating. The decrease in width and depth of the wear scars demonstrated the excellent wear resistance of the coating.

As a core component of rotating machinery, the condition of rolling bearings is directly related to the reliability and safety of equipment operation; therefore, the accurate and reliable monitoring of bearing operating status is critical. However, when dealing with non-stationary and noisy vibration signals, traditional fault diagnosis methods are often constrained by limited feature characterization from single time–frequency analysis and inadequate feature extraction capabilities. To address this issue, this study proposes a lightweight fault diagnosis model (WaveCAResNet) enhanced with multi-source time–frequency features. By fusing complementary time–frequency features derived from continuous wavelet transform, short-time Fourier transform, Hilbert–Huang transform, and Wigner–Ville distribution, the capability to characterize complex fault patterns is significantly improved. Meanwhile, an efficient and lightweight deep learning model (WaveCAResNet) is constructed based on residual networks by integrating multi-scale analysis via a wavelet convolutional layer (WTConv) with the dynamic feature optimization properties of channel-attention-weighted residuals (CAWRs) and the efficient temporal modeling capabilities of weighted residual efficient multi-scale attention (WREMA). Experimental validation indicates that the proposed method achieves higher diagnostic accuracy and robustness than existing mainstream models on typical bearing datasets, and the classification performance of the newly proposed model exceeds that of state-of-the-art bearing fault diagnostic models on the same dataset, even under noisy conditions.

Currently, research on the edge effect issue in the micro-arc oxidation process primarily focuses on investigating process conditions and enhancing additives. However, some scholars have utilized finite element analysis software to simulate the edge effect during the simulation process, overlooking the investigation of the morphology of the auxiliary cathode. This study analyzes the growth characteristics of the oxide film on aluminum alloy 2A12 during micro-arc oxidation. Additionally, the inter-electrode discharge electric field is simulated using the finite element analysis method. The auxiliary cathode is optimized to mitigate the influence of the edge effect on the film layer. The findings indicate that employing a cylindrical shape as the auxiliary cathode instead of a rectangular groove leads to an increased thickness of the micro-arc oxidation film. However, it also results in an augmented length of the film layer affected by the edge effect at both ends of the workpiece. Decreasing the distance between the auxiliary cathode and the workpiece surface leads to a higher thickness of the obtained micro-arc oxidation film. Decreasing the length of the auxiliary cathode results in a reduction in both the thickness of the film layer on the workpiece surface and the area affected by the edge effect. Increasing the eccentric cone ratio of the auxiliary cathode enhances the uniformity of the micro-arc oxidation film layer. In this study, we present a novel auxiliary cathode model that incorporates a smaller cylindrical shell at the center and eccentric cone shells on each side. This model has the potential to enhance the optimization rate of the micro-arc oxidation film layer on cylindrical workpieces by 17.77%.

In response to the high missed detection rates and low efficiency of traditional methods in detecting tiny defects on the machining surfaces of mechanical parts, this study proposes an efficient defect detection method based on deep learning. Initially, referencing the network architectures of Resnet and Yolo, an image detection network was designed featuring a shared encoder, a classification decoder, and a localization decoder. The shared encoder is used to extract a unified feature representation; the classification decoder accomplishes efficient data classification; and the localization decoder achieves precise defect localization. Furthermore, upon acquiring high-resolution images of the machining surfaces with dimensional features, this study introduces a real-time sliding window method to perform segmented detection and classification of these images, transforming most of the target detection tasks into image classification problems, thereby further enhancing the efficiency and accuracy of defect detection and target localization. Practical results demonstrate that this method outperforms traditional approaches in terms of missed detection rates and detection efficiency, effectively addressing the challenge of detecting complex machining surface defects, and providing a high-precision, high-efficiency defect detection solution for the mechanical part machining field.