In the era of big data, data-driven technologies are increasingly leveraged by industry to facilitate autonomous learning and intelligent decision-making.
The hermetic refrigeration compressor is the core component of the refrigeration system, failure of which will cause energy waste and reduce service life. Fault diagnosis based on vibration signal is a research hotspot. However, it is challenging to extract features of nonlinear and nonstationary vibration signals, which severely restricts the development of this method. This paper proposes a dual time-frequency images fusion method to obtain more effective features for diagnosing compressor manufacturing defects. Firstly, two time-frequency images are obtained by implementing continuous wavelet transform and Hilbert-Huang transform of the same vibration signal sample. Then, a convolutional neural network is used for image feature extraction and fusion, where the features extracted from two time-frequency images have complementarity. A data set containing six categories of typical manufacturing defects is used to verify the proposed method. The results show that the average diagnostic accuracy of the proposed method reaches 95.9%, and the proposed method has a better performance than other methods.
In this study, in order to improve and restore the performance of the polishing pads and reduce the cost of chemical mechanical polishing, three types of material polishing pads, namely, polyurethane, damping cloth, and non-woven fabric, were selected for the experiment. Accordingly, each polishing pad was set up with diamond conditioner and high-pressure micro-jet (HPMJ) conditioning control experiments. Subsequently, the fluctuation ranges of the material removal rate on the three polishing pads were 2.73–3.75 μm/h, 1.38–1.99 μm/h, and 2.36–4.32 μm/h, respectively under the HPMJ conditioning method, while the fluctuation ranges of the material removal rate on the three polishing pads were 1.80–4.14 μm/h, 1.02–2.09 μm/h, and 1.78–5.88 μm/h under the diamond conditioning method. Comparing the polishing pad morphologies under SEM, we observed that the surface of the polishing pad after HPMJ conditioning was relatively clean, and the hole structure was not blocked. Contrastingly, there remained numerous abrasive particles on the surface after the conventional diamond conditioning and the hole structure was blocked. Thus, the HPMJ conditioning technology is better than the traditional diamond conditioning technology. Subsequently, the polishing pad after HPMJ conditioning has a longer service life and a more stable material removal rate than that after traditional diamond conditioning.