Open Access

Articles

Article ID: 2252

The influence of metro operation vibration on single-layer cable net glass curtain wall structure based on frequency method

by Zhirong Shen, Yuze Yan, Yi Tao, Zhiwei Wang

Sound & Vibration, Vol.59, No.1, 2025;

Using the single-layer cable net glass curtain wall of Siemens Shanghai Center Building A as a case study, the influence of vibrations from two metro lines on the structure was investigated through combination of field measurements with vibration signal analysis, and numerical simulations using finite element analysis. Time-domain data from cable vibrations were converted to frequency-domain data using fast Fourier transform for spectral analysis. The frequencies of the cables under different constraints were calculated based on the linear theory for vibrations of a string. A comprehensive numerical model of the single-layer cable net curtain wall was established for modal analysis. Comparative analysis shows that cables with horizontal floor constraints are highly vibration-sensitive, with metro operations causing high-order modal vibrations and fundamental frequency vibrations in sections between adjacent constraints, leading to glass deformation and damage. Neither metro operation induces the fundamental frequency vibration of the entire structure. Comparative frequency analysis favors theoretical frequency calculations with horizontal constraints for multi-constraint cable frequency analysis.

Open Access

Articles

Article ID: 2022

Vehicle structural road noise prediction based on an improved Long Short-Term Memory method

by Xiongying Yu, Ruxue Dai, Jian Zhang, Yingqi Yin, Sha Li, Peisong Dai, Haibo Huang

Sound & Vibration, Vol.59, No.1, 2025;

The control of vehicle interior noise has become a critical metric for assessing noise, vibration, and harshness (NVH) in vehicles. During the initial phases of vehicle development, accurately predicting the impact of road noise on interior noise is essential for reducing noise levels and expediting the product development cycle. In recent years, data-driven methods based on machine learning have gained significant attention due to their robust capability in navigating complex data mapping relationships. Notably, surrogate models have demonstrated exceptional performance in this domain. Numerous researchers have integrated diverse intelligent algorithms into the study of vehicle noise, leveraging advantages such as the elimination of precise modeling requirements, extensive solution space exploration, continuous learning from data, and robust algorithmic versatility. However, in NVH engineering applications, data-driven models face inherent limitations, particularly in interpretability and stability. To address these issues, this paper introduces an improved Long Short-Term Memory (LSTM) network that combines knowledge and data. Inspired by the physical information neural network concept, this approach incorporates values calculated through empirical formulas into the neural network as constraints. Comparative assessments with traditional LSTM networks highlight the advantages of this deep learning model. By integrating empirical formulas constraints, the model not only enhances interpretability but also achieves robust generalization with fewer data samples. The proposed method is validated on a specific vehicle model, showing significant improvements in prediction accuracy and efficiency.

Open Access

Articles

Article ID: 2242

Engineering vibration recognition using CWT-ResNet

by Wei Huang, Jian Xu

Sound & Vibration, Vol.59, No.1, 2025;

Multi-source signal recognition is a common problem in engineering vibration control. Given that traditional methods often primarily rely on prior knowledge and expertise, which can limit efficiency and accuracy, this study proposed a vibration recognition model based on ResNet, utilizing continuous wavelet transform to combine signal processing with deep learning techniques. The continuous wavelet transform converts the original one-dimensional vibration signals into two-dimensional time-frequency representations with richer feature information, which are then input into the convolutional layers for automatic feature extraction, culminating in vibration recognition through the Softmax layer. To evaluate the model’s performance, 20 sets of measured vibration data were tested. The results show that the proposed model achieves a recognition accuracy of 99%, excelling in both component recognition and the separation of vibration signals. Therefore, this study is of great significance for engineering vibration diagnosis, the front-end design of vibration control, and the analysis and optimization of control effectiveness.

Open Access

Articles

Article ID: 1944

Design and research of numerical control simulation platform in discrete manufacturing disturbed by vibration

by Ganlong Wang, Yue Wang, Yanxia Wu, Guoyin Zhang, Jianxun Chen

Sound & Vibration, Vol.59, No.1, 2025;

It is of great significance to study the deep integration of manufacturing technology and new-generation information communication technology under vibration interference of machine tools to improve the intelligence level of CNC machine tools. In this paper, a numerical control manufacturing workshop affected by vibration in discrete manufacturing is taken as the research background, and a solution for a digital workshop operation simulation platform based on the industrial internet is proposed. By constructing the simulation environment of the operation process of the digital factory, the generation and transmission of manufacturing information in the digital factory are simulated. The application architecture of the machining workshop based on a numerical control simulation platform is proposed, and the business process of the numerical control machining workshop is analyzed. Then, the key technologies of NC machine tool modeling, synchronous mapping of data and model, data integration, and fusion are studied. Through the integration and implementation of the NC machine tool simulation platform in the machining workshop, the top-down data instructions can be issued accurately, and the bottom-up feedback information can be confirmed in time. Finally, the system is applied to the electronic information and ship machining workshop to verify the effectiveness of the system framework and method proposed in this paper.

Open Access

Articles

Article ID: 2073

Sound insulation prediction and optimization of wooden support structure for high-speed train floor based on machine learning

by Haiyang Ding, Ruiqian Wang, Xuefei Zhang, Ziyan Xu, Ancong Zhang, Lei Xu

Sound & Vibration, Vol.59, No.1, 2025;

In order to improve the sound insulation performance of high-speed train floors, this study first obtained the necessary data for model training based on the reverberation test method, and then conducted data sorting and feature selection. Next, the maximum mutual information minimum redundancy (mRMR) feature selection algorithm was used to calculate the selected features and screen out a subset of significant features. Subsequently, the decision tree, BP neural network, and support vector machine regression (SVR) methods were applied in sequence, and the standardized feature data were used for the high-speed train floor under the same evaluation criteria of the mean square error (MSE) and coefficient of determination (R²). We conducted training and validation of the sound insulation prediction models for timber-framed support structures. The prediction accuracy of the trained model was compared and evaluated with the finite element statistical energy analysis (FE-SEA) prediction model. Finally, the SVR model was used to optimize the design under constraint conditions. The research results show that based on the research object, sample library, and model training in this article, compared with the FE-SEA model, the prediction error of the SVR model is only 0.3 dB, showing better performance. In engineering practice, the SVR model can effectively optimize the wooden support structure in the floor under certain constraints, and it predicts that the weighted sound insulation of the entire floor is 50.45 dB, which has important engineering application value.

Open Access

Articles

Article ID: 2048

Design optimization of vibration amplitude reduction based on virtual prototype and machine learning

by Hong Bao, Jinxuan Tao, Jing Yang, Bin Cao, Liuxian Zhao

Sound & Vibration, Vol.59, No.1, 2025;

The traditional design optimization of vibration amplitude reduction mainly has the disadvantages of low modeling and prediction accuracy as well as low optimization efficiency. Therefore, this paper presents a design optimization method for vibration amplitude reduction based on virtual prototyping and machine learning, which combines the high accuracy of numerical calculations with the efficiency of machine learning, overcoming the shortcomings of traditional methods. Firstly, sample points are collected through the design of experiments and virtual prototype simulation. Then, based on the sampled data, a prediction model for the relationship between the design parameters and the amplitude of the product is established using Genetic Algorithm-Support Vector Regression (GA-SVR). On the basis of the GA-SVR prediction model, a multi-objective optimization model of product is established, and Multiple Objectives Particle Swarm Optimization -entropy weight- Technique for Order Preference by Similarity to Ideal Solution (MOPSO-entropy weight-TOPSIS) is used to solve for the optimal design parameters. Finally, the washing machine suspension system is used as an example to verify the effectiveness of the model. The results show that, compared with the original design scheme, the design scheme obtained by the model can reduce the amplitude of the washing machine suspension system by 12.68%, and reduce the total weight of the counterweight by 7.35%. This method is conducive to the intelligent and efficient design optimization of vibration amplitude reduction, and is of great significance to product life cycle design.