人工神经网络估算超细玻璃纤维毡声传输损失的适用性

Q1 Arts and Humanities

Acta Acustica united with Acustica Pub Date : 2019-07-01 DOI:10.3813/AAA.919345

Fei Wang, Zhaofeng Chen, Cao Wu

{"title":"人工神经网络估算超细玻璃纤维毡声传输损失的适用性","authors":"Fei Wang, Zhaofeng Chen, Cao Wu","doi":"10.3813/AAA.919345","DOIUrl":null,"url":null,"abstract":"In the present study, the sound transmission loss (STL) of ultrafine glass fiber felts in terms of areal density and sound frequency has been modeled by artificial neural network (ANN), the Law of Theoretic Mass and fitting polynomial, respectively. The STL of ultrafine glass fiber\n felts with the areal density ranging from 0 to 300 g/m2 and at the sound frequency ranging from 500 to 6300 Hz was employed as training data for ANN. By the optimization of ANN structure, the number of neurons in the two hidden layers was determined to 8 and 4 respectively. The\n mean squared error of the ANN model was only 0.191 and the correlation coefficient was 0.9989, which showed high accuracy for estimating the STL of the felts. Compared with other two models, the ANN model showed excellent agreement with the measured results and it's very appropriate for the\n estimation of acoustic properties of ultrafine glass fiber felts.","PeriodicalId":35085,"journal":{"name":"Acta Acustica united with Acustica","volume":"54 1","pages":""},"PeriodicalIF":0.0000,"publicationDate":"2019-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"2","resultStr":"{\"title\":\"Applicability of Artificial Neural Network to Estimate Sound Transmission Loss of Ultrafine Glass Fiber Felts\",\"authors\":\"Fei Wang, Zhaofeng Chen, Cao Wu\",\"doi\":\"10.3813/AAA.919345\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"In the present study, the sound transmission loss (STL) of ultrafine glass fiber felts in terms of areal density and sound frequency has been modeled by artificial neural network (ANN), the Law of Theoretic Mass and fitting polynomial, respectively. The STL of ultrafine glass fiber\\n felts with the areal density ranging from 0 to 300 g/m2 and at the sound frequency ranging from 500 to 6300 Hz was employed as training data for ANN. By the optimization of ANN structure, the number of neurons in the two hidden layers was determined to 8 and 4 respectively. The\\n mean squared error of the ANN model was only 0.191 and the correlation coefficient was 0.9989, which showed high accuracy for estimating the STL of the felts. Compared with other two models, the ANN model showed excellent agreement with the measured results and it's very appropriate for the\\n estimation of acoustic properties of ultrafine glass fiber felts.\",\"PeriodicalId\":35085,\"journal\":{\"name\":\"Acta Acustica united with Acustica\",\"volume\":\"54 1\",\"pages\":\"\"},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2019-07-01\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"2\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Acta Acustica united with Acustica\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/10.3813/AAA.919345\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"Arts and Humanities\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Acta Acustica united with Acustica","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.3813/AAA.919345","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"Arts and Humanities","Score":null,"Total":0}

引用次数: 2

摘要

本文采用人工神经网络、理论质量定律和拟合多项式分别对超细玻璃纤维毡面密度和声频的传声损失进行了建模。采用面密度为0 ~ 300 g/m2，声频为500 ~ 6300 Hz的超细玻璃纤维毡的STL作为人工神经网络的训练数据。通过对ANN结构的优化，确定两个隐藏层的神经元数量分别为8个和4个。人工神经网络模型的均方误差仅为0.191，相关系数为0.9989，对毛毡STL的估计精度较高。与其他两种模型相比，人工神经网络模型与实测结果吻合良好，非常适合超细玻璃纤维毡声学性能的估计。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

查看原文本刊更多论文

Applicability of Artificial Neural Network to Estimate Sound Transmission Loss of Ultrafine Glass Fiber Felts

In the present study, the sound transmission loss (STL) of ultrafine glass fiber felts in terms of areal density and sound frequency has been modeled by artificial neural network (ANN), the Law of Theoretic Mass and fitting polynomial, respectively. The STL of ultrafine glass fiber felts with the areal density ranging from 0 to 300 g/m2 and at the sound frequency ranging from 500 to 6300 Hz was employed as training data for ANN. By the optimization of ANN structure, the number of neurons in the two hidden layers was determined to 8 and 4 respectively. The mean squared error of the ANN model was only 0.191 and the correlation coefficient was 0.9989, which showed high accuracy for estimating the STL of the felts. Compared with other two models, the ANN model showed excellent agreement with the measured results and it's very appropriate for the estimation of acoustic properties of ultrafine glass fiber felts.

求助全文

通过发布文献求助，成功后即可免费获取论文全文。去求助

来源期刊

Acta Acustica united with Acustica 物理-声学

CiteScore

2.60

自引率

0.00%

发文量

审稿时长

6.8 months

期刊介绍： Cessation. Acta Acustica united with Acustica (Acta Acust united Ac), was published together with the European Acoustics Association (EAA). It was an international, peer-reviewed journal on acoustics. It published original articles on all subjects in the field of acoustics, such as • General Linear Acoustics, • Nonlinear Acoustics, Macrosonics, • Aeroacoustics, • Atmospheric Sound, • Underwater Sound, • Ultrasonics, • Physical Acoustics, • Structural Acoustics, • Noise Control, • Active Control, • Environmental Noise, • Building Acoustics, • Room Acoustics, • Acoustic Materials and Metamaterials, • Audio Signal Processing and Transducers, • Computational and Numerical Acoustics, • Hearing, Audiology and Psychoacoustics, • Speech, • Musical Acoustics, • Virtual Acoustics, • Auditory Quality of Systems, • Animal Bioacoustics, • History of Acoustics.