Nondestructive evaluation of metal components using ultrasound is a widely adopted technique. Nevertheless, in practical applications, various factors may influence the outcomes of the testing. This study thoroughly examines the influencing factors of the ultrasonic testing process and employs deep learning techniques to explore how these factors affect the ultrasonic signals and the final testing results. Subsequently, a numerical simulation case was developed in COMSOL 6.2 to investigate how workpiece surface roughness, coupling materials, probe pressure affects the maximum amplitude of the ultrasonic signal, as well as their influence on the final testing results, thereby validating the effectiveness of our approach.
The NVH (Noise, Vibration, Harshness) of electric vehicles greatly affects driving comfort, and the NVH has become a key indicator for verifying the quality of their products, strengthening the research on noise testing technology can greatly promote the high-quality development of reducers. Given the problem that the noise data acquisition environment of the noise test stand is highly disturbed and difficult to eliminate, this paper proposes a whale optimization algorithm (WOA) with envelope entropy as the fitness function to carry out the improved variational modal decomposition (VMD) to decompose the noise signal of the speed reducer to obtain the IMF component, and to use the signal inter-correlation analysis of the IMF background noise to identify and filter out the noise reduction, and to complete the separation of the noise from the background noise of the speed reducer. The method is based on the iterative optimization of the number of decomposition layers and penalty factor of the VMD by WOA to find out the optimal combination of decomposition parameters and identify and reduce the background noise components based on the signal correlation analysis. Firstly, it is verified that the method has more effective decomposition performance than the traditional VMD algorithm by simulation signal test. Then the method is used for the background noise separation of electric vehicle gearbox, and the results show that the WOA-VMD proposed in this paper correctly separates the background noise signals in the mixed signals, and can effectively carry out the noise separation, and the decomposition filters out the background noise with an average reduction of the sound pressure level of 2.99 dB(A), and the average error of the sound pressure level with the baseline condition is 0.45 dB(A). Also, the noise reduction method using WOA-VMD reduces the sound pressure level by an average of 1.47 dB(A) than that obtained by the VMD algorithm, which has a better noise separation effect. The above conclusion proves that the present method improves the accuracy of noise testing of electric vehicle reducers.
Bearings are widely used in industrial production lines and rotating machinery equipment, and bearing failure may prevent the operation of the entire system, so its reliability is very important. It is particularly important to repair and maintain bearings before failure, so we need to accurately know the residual life of bearings and assess the degradation state of bearings. Therefore, this paper puts forward a bearing life prediction method based on EMD(empirical mode decomposition) and PF (particle filter). First, aiming at the non-linear and non-stationary features of vibration signals of bearings, EMD is conducted for signals, which mainly eliminates the noise mixed in the original signals effectively through effective component reconstruction, analyzes the intrinsic mode component obtained and extract its energy entropy, and take the energy entropy feature as the degradation characteristics indicating the heath state of bearings. Second, the performance degradation model of bearings is built according to the trend characteristics, and PFis used to update the parameters of the performance degradation model of bearings for predicting the remaining life. In order to verify its effectiveness, the method proposed is verified with two groups of whole-life data of bearings and compared with the traditional ARIMA model. The experimental results show that the RMSE value calculated by the proposed EMD-PF prediction model is significantly smaller than that of the ARIMA model, and the prediction error decreases gradually with the degradation process. The prediction model proposed has stable convergence and can effectively predict the degradation process of bearings.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.