Applied Acoustics最新文献

Evaluating metric and contrastive learning in pretrained models for environmental sound classification

IF 3.4 2区物理与天体物理

Applied Acoustics Pub Date : 2025-02-11 DOI: 10.1016/j.apacoust.2025.110593

Feilong Chen , Zhenjun Zhu , Chengli Sun , Linqing Xia

{"title":"Evaluating metric and contrastive learning in pretrained models for environmental sound classification","authors":"Feilong Chen , Zhenjun Zhu , Chengli Sun , Linqing Xia","doi":"10.1016/j.apacoust.2025.110593","DOIUrl":"10.1016/j.apacoust.2025.110593","url":null,"abstract":"<div><div>Environmental Sound Classification (ESC) has advanced significantly with the advent of deep learning techniques. This study conducts a comprehensive evaluation of contrastive and metric learning approaches in ESC, introducing the ESC51 dataset, an extension of the ESC50 benchmark that incorporates noise samples from quadrotor Unmanned Aerial Vehicles (UAVs). To enhance classification performance and the discriminative power of embedding spaces, we propose a novel metric learning-based approach, SoundMLR, which employs a hybrid loss function emphasizing metric learning principles. Experimental results demonstrate that SoundMLR consistently outperforms contrastive learning methods in terms of classification accuracy and inference latency, particularly when applied to the lightweight MobileNetV2 pretrained model across ESC50, ESC51, and UrbanSound8K (US8K) datasets. Analyses of confusion matrices and t-SNE visualizations further highlight SoundMLR’s ability to generate compact, distinct feature clusters, enabling more robust discrimination between sound classes. Additionally, we introduce two innovative modules, Spectral Pooling Attention (SPA) and the Feature Pooling Layer (FPL), designed to optimize the MobileNetV2 backbone. Notably, the MobileNetV2 + FPL model, equipped with SoundMLR, achieves an impressive 92.16 % classification accuracy on the ESC51 dataset while reducing computational complexity by 24.5 %. Similarly, the MobileNetV2 + SPA model achieves a peak accuracy of 91.75 % on the ESC50 dataset, showcasing the complementary strengths of these modules. These findings offer valuable insights for the future development of efficient, scalable, and robust ESC systems. The source code for this study is publicly available at <span><span>https://github.com/flchenwhu/ESC-SoundMLR</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"232 ","pages":"Article 110593"},"PeriodicalIF":3.4,"publicationDate":"2025-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143378692","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

The ecological validity of laboratory experiments in soundscape and landscape research: A systematic review and meta-analysis

IF 3.4 2区物理与天体物理

Applied Acoustics Pub Date : 2025-02-11 DOI: 10.1016/j.apacoust.2025.110582

Yichun Lu, Siu-Kit Lau

{"title":"The ecological validity of laboratory experiments in soundscape and landscape research: A systematic review and meta-analysis","authors":"Yichun Lu, Siu-Kit Lau","doi":"10.1016/j.apacoust.2025.110582","DOIUrl":"10.1016/j.apacoust.2025.110582","url":null,"abstract":"<div><div>Soundscape and landscape research considers people’s perceptions of the audio-visual environment. Laboratory experiments that enable manipulation and randomization are crucial for testing hypotheses and causal relationships in these fields. However, concerns have been raised regarding the generalizability of laboratory findings to real-world contexts. To address this issue, researchers are increasingly investigating the ecological validity of laboratory experiments, that is, the degree to which laboratory results reflect real-world ones. This systematic review examined 104 papers investigating the ecological validity of laboratory experiments in soundscape and landscape research. Descriptors for measuring ecological validity and potential experimental factors influencing it were identified. Meta-analysis was conducted on 45 eligible papers to examine the effects of these experiment factors on ecological validity, highlighting the importance of factors such as auralization, visualization, and human–computer interaction. Lastly, a framework for ecological validity research in soundscape and landscape studies is proposed, offering guidance for future research.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"232 ","pages":"Article 110582"},"PeriodicalIF":3.4,"publicationDate":"2025-02-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143378686","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Weak time-varying fault indicator detection of the complex gearbox via the meshing order modulation-aided scaling reassigning chirplet transform

IF 3.4 2区物理与天体物理

Applied Acoustics Pub Date : 2025-02-07 DOI: 10.1016/j.apacoust.2025.110580

Shaodan Zhi , Hengshan Wu , Haikuo Shen , Weidong Cheng , Yancai Xiao , Tianyang Wang

{"title":"Weak time-varying fault indicator detection of the complex gearbox via the meshing order modulation-aided scaling reassigning chirplet transform","authors":"Shaodan Zhi , Hengshan Wu , Haikuo Shen , Weidong Cheng , Yancai Xiao , Tianyang Wang","doi":"10.1016/j.apacoust.2025.110580","DOIUrl":"10.1016/j.apacoust.2025.110580","url":null,"abstract":"<div><div>Faced with the weak fault detection of a complex planetary gearbox under time-varying operational conditions, a new time–frequency analysis method, namely the meshing order modulation-aided scaling reassigning chirplet transform (MOM-SRCT), is proposed. Specifically, a new meshing modulation analysis strategy suitable for time-varying speed and a novel scaling reassigning chirplet transform are separately designed and combined to realize time-varying weak fault indicator detection directly from the lower frequency area of the time–frequency representation with a considerably high resolution. The two main advantages of the MOM-SRCT are its ability of mesh modulation frequency band identification at time-varying speeds and accurate time-varying weak fault indicator capture. Two types of gear defects, the missing tooth fault and the gear tooth break fault, are employed on a planetary gearbox test rig to verify the effectiveness of the proposed method and display the superiority over conventional algorithms in identifying time-varying weak fault indicators.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"232 ","pages":"Article 110580"},"PeriodicalIF":3.4,"publicationDate":"2025-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143350383","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

High gain beamforming method in low-frequency band using virtual array migration

IF 3.4 2区物理与天体物理

Applied Acoustics Pub Date : 2025-02-06 DOI: 10.1016/j.apacoust.2025.110570

Yuan Hu , Jingwei Yin , Tianxing Li , Jianguo Cao , Liang Zhang

{"title":"High gain beamforming method in low-frequency band using virtual array migration","authors":"Yuan Hu , Jingwei Yin , Tianxing Li , Jianguo Cao , Liang Zhang","doi":"10.1016/j.apacoust.2025.110570","DOIUrl":"10.1016/j.apacoust.2025.110570","url":null,"abstract":"<div><div>When measuring the low-frequency radiated noise of underwater low-noise targets, a conflict arises between the large-scale array’s ability to achieve low-frequency gain and the reduction of propagation loss at a fixed measurement distance. This paper presents a method for virtual array element migration, based on the measured phase modulation of array elements. By increasing the phase differences between adjacent elements and further enhancing the phase differences between steering vectors in different directions, the proposed method improves the suppression of undesired directional interference signals, thereby enabling high-gain measurements under limited aperture conditions. To mitigate the reduction in robustness associated with virtual array construction, a virtual array factor is introduced for analysis. The results demonstrate that when the virtual array factor is set to 1, the proposed method performs comparably to conventional beamforming. When the virtual array factor exceeds 1, it can effectively enhance the suppression of interference signals in the low-frequency range. The optimal value of this factor should be selected based on the signal-to-noise ratio of the data to ensure the robustness of the algorithm. Both simulation and lake test data demonstrate that the proposed algorithm significantly outperforms conventional methods in terms of its ability to reject interference in low-frequency radiated noise measurements.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"232 ","pages":"Article 110570"},"PeriodicalIF":3.4,"publicationDate":"2025-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143261660","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A fault identification method of hydraulic pump fusing long short-term memory and synchronous compression wavelet transform

IF 3.4 2区物理与天体物理

Applied Acoustics Pub Date : 2025-02-05 DOI: 10.1016/j.apacoust.2025.110553

Shengnan Tang , Yixuan Jiang , Hong Su , Yong Zhu

{"title":"A fault identification method of hydraulic pump fusing long short-term memory and synchronous compression wavelet transform","authors":"Shengnan Tang , Yixuan Jiang , Hong Su , Yong Zhu","doi":"10.1016/j.apacoust.2025.110553","DOIUrl":"10.1016/j.apacoust.2025.110553","url":null,"abstract":"<div><div>The hydraulic pump (HP) is the “power heart” of the hydraulic transmission system. It is widely used in high-pressure, high flow, and flow regulation situations, such as hydraulic presses and engineering machinery. Due to the harsh working environment and complexity of the structure, faults in HPs are often not detected and treated in time, resulting in some economic losses and even serious accidents. This research introduces a long short-term memory (LSTM) and visual geometry group (VGG) fused fault diagnosis approach for HPs. First, the one-dimensional non-stationary acoustic signal is transformed into the high-frequency two-dimensional time–frequency image through synchronous compressed wavelet transform, which serves as the input for the convolutional neural network. Second, VGG is introduced for the fusion of extracted features to increase the attention of important features and thus enhance the representation ability of the network. Third, LSTM is incorporated for the mining of time series. Finally, the fault diagnosis results are verified through the Softmax layer. To verify the performance of the proposed model, five types of faults are identified through experiments. As compared with other traditional diagnostic methods, the proposed model has superior accuracy and stability.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"232 ","pages":"Article 110553"},"PeriodicalIF":3.4,"publicationDate":"2025-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143261659","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Adaptive active control of underwater low-frequency radiated noise based on acoustics energy flow method

IF 3.4 2区物理与天体物理

Applied Acoustics Pub Date : 2025-02-05 DOI: 10.1016/j.apacoust.2025.110569

Xuesong Zhang , Dejiang Shang

{"title":"Adaptive active control of underwater low-frequency radiated noise based on acoustics energy flow method","authors":"Xuesong Zhang , Dejiang Shang","doi":"10.1016/j.apacoust.2025.110569","DOIUrl":"10.1016/j.apacoust.2025.110569","url":null,"abstract":"<div><div>An adaptive control method is proposed to suppress far-field, low-frequency noise radiated from underwater target structures, with the normal sound energy flow in the structure's near field as the control target to reduce far-field radiated noise. An APSO-LM-BP neural network (APSO = adaptive particle swarm optimization; LM = Levenberg–Marquardt; BP = backpropagation) optimization algorithm is employed to train on sample data on the basis of placement of secondary sound sources and error sensors, establishing a mathematical model between the positions of the secondary sound sources, error sensors, and radiated sound power of the sound field after control. Numerical simulations and lake experiments are conducted on a double-layer, ribbed, cylindrical shell model. The results indicate that the neural network accurately predicts placement positions of secondary sound sources and error sensors and that the optimized placement strategy achieves effective global noise control in the low-frequency range. In addition, with increasing frequency, noise control can be achieved in specific regions, but total radiated sound power decreases with increase in number of secondary sources and error sensors.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"232 ","pages":"Article 110569"},"PeriodicalIF":3.4,"publicationDate":"2025-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143261697","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A zero-shot model for diagnosing unknown composite faults in train bearings based on label feature vector generated fault features

IF 3.4 2区物理与天体物理

Applied Acoustics Pub Date : 2025-02-03 DOI: 10.1016/j.apacoust.2025.110563

Deqiang He , Yuan Xu , Zhenzhen Jin , Qi Liu , Ming Zhao , Yanjun Chen

{"title":"A zero-shot model for diagnosing unknown composite faults in train bearings based on label feature vector generated fault features","authors":"Deqiang He , Yuan Xu , Zhenzhen Jin , Qi Liu , Ming Zhao , Yanjun Chen","doi":"10.1016/j.apacoust.2025.110563","DOIUrl":"10.1016/j.apacoust.2025.110563","url":null,"abstract":"<div><div>The composite faults in train traction motor bearings are diverse and often unknown. Current approaches heavily rely on extensive training datasets to guarantee the dependability of diagnostic outcomes. However, obtaining training samples of unknown composite faults in train bearings under real-world conditions is challenging. To tackle this problem, this study introduces a zero-shot diagnostic framework that utilizes acoustic signals captured by voiceprint sensors for diagnosing unknown composite faults. The approach builds upon a model that generates fault features from label feature vectors (LFV), enabling the diagnosis of unknown composite faults based on knowledge of single faults. First, a feature extraction approach using a spatially enhanced convolutional neural network is designed, introducing a spatial attention mechanism to strengthen the model’s attention to critical aspects of the features. Subsequently, a definition method for LFV is established to map the relationship between the extracted features and the LFV. A Wasserstein-generating adversarial network with a second-order dynamic gradient penalty is then proposed to generate virtual features based on the LFV. The designed second-order dynamic gradient penalty function helps the model explore the parameter space more efficiently and find the optimal solution, reducing the discrepancy between generated and real features. Finally, two independent acoustic datasets verified the model’s robustness. Without training on composite fault data, the model achieved a classification accuracy of 69.84% for four types of unknown composite faults in bearings, surpassing other methods for bearing composite fault diagnosis.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"232 ","pages":"Article 110563"},"PeriodicalIF":3.4,"publicationDate":"2025-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143261699","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Research on jet control and noise reduction mechanism of sound source for leading bogie region

IF 3.4 2区物理与天体物理

Applied Acoustics Pub Date : 2025-02-03 DOI: 10.1016/j.apacoust.2025.110567

Li Zhuo-ming , Li Qi-liang , Yang Zhi-gang

{"title":"Research on jet control and noise reduction mechanism of sound source for leading bogie region","authors":"Li Zhuo-ming , Li Qi-liang , Yang Zhi-gang","doi":"10.1016/j.apacoust.2025.110567","DOIUrl":"10.1016/j.apacoust.2025.110567","url":null,"abstract":"<div><div>In order to achieve sound source reduction in the leading bogie region, turbulent flow and near-filed acoustic pressure were solved based on large eddy simulation and acoustic perturbation equations. The noise control effectiveness of the jet control was confirmed through 1:5 scale high-speed trains model wind tunnel test and the numerical simulation method was validated. The effect of jet control at different positions, velocities and directions were evaluated. Notably, the Bogie-Jet-0.35 V scheme, which involves applying a jet flow at 0.35 times the incoming flow velocity at the leading edge of the bogie, exhibited superior control performance. Compared to the base condition, the optimal scheme achieved optimized control of acoustic and turbulent pressures, resulting in reductions of 2.0 dB and 2.7 dB, respectively. An improved reduced order method based on sample entropy was developed to explore the jet control mechanism. It was found that in the snapshots with high acoustic energy, the acoustic energy was primarily concentrated at the front and rear edges of the bogie. Jet flow reduced the turbulent pressure fluctuation energy near the front wheelset of the bogie. Additionally, jet flow reduced the probability of unsteady flow state transitioning to clusters with high acoustic pressure.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"232 ","pages":"Article 110567"},"PeriodicalIF":3.4,"publicationDate":"2025-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143261698","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Broaden noise reduction range in low frequency by a HR + MPP structure based on impedance matching method

IF 3.4 2区物理与天体物理

Applied Acoustics Pub Date : 2025-02-01 DOI: 10.1016/j.apacoust.2025.110572

Hanbo Shao, Weiyu Chen, Dong Jiang

{"title":"Broaden noise reduction range in low frequency by a HR + MPP structure based on impedance matching method","authors":"Hanbo Shao, Weiyu Chen, Dong Jiang","doi":"10.1016/j.apacoust.2025.110572","DOIUrl":"10.1016/j.apacoust.2025.110572","url":null,"abstract":"<div><div>Broadening the noise reduction range in low frequency has always been an important problem in the research of acoustic metamaterials. Traditional muffler like Helmholtz resonance cavities (HR) only have high absorption coefficient in its resonant frequency. By designing multi-layer or gradient air cavities, the range of frequency broadened is still limited. As is well known, microperforated plate (MPP) has a wide sound absorption frequency in the middle and low frequency. Taking advantage of the physical property of HR and MPP structure, we designed a composite structure by the method of surface impedance matching. Based on the noise reduction coefficient higher than 0.9, this structure realized a 967 Hz width noise reduction in the frequency below 1000 Hz. More than twice as the multi cavity HR structure we researched before. It is demonstrated that, by the impedance matching method, there is almost no reflection in the contact surface between HR and MPP. Therefore, the low resonant frequency and relatively medium frequency can be well absorbed in HR and MPP structures respectively.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"232 ","pages":"Article 110572"},"PeriodicalIF":3.4,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143261704","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Assessing the effect of construction noise frequency on mental workload of construction workers with varying task difficulty using EEG data

IF 3.4 2区物理与天体物理

Applied Acoustics Pub Date : 2025-02-01 DOI: 10.1016/j.apacoust.2025.110571

Samuel Oluwadamilare Olatunbosun , Francis Xavier Duorinaah , Chan-Hoon Haan , Chang-Yong Yi , Min-Koo Kim

{"title":"Assessing the effect of construction noise frequency on mental workload of construction workers with varying task difficulty using EEG data","authors":"Samuel Oluwadamilare Olatunbosun , Francis Xavier Duorinaah , Chan-Hoon Haan , Chang-Yong Yi , Min-Koo Kim","doi":"10.1016/j.apacoust.2025.110571","DOIUrl":"10.1016/j.apacoust.2025.110571","url":null,"abstract":"<div><div>Noise pollution is a critical concern on construction sites because it adversely affects the cognitive performance of workers. While prior research has explored the effect of construction noise on workers’ cognitive performance, gaps remain in understanding how these effects change under varying task difficulty and noise frequency. To address this limitation, this study investigates the combined effects of construction noise frequency and task difficulty on mental workload using EEG. To this end, a 3-level N-back memory task was conducted under three noise conditions while EEG data of participants was recorded simultaneously. Furthermore, behavioral and subjective data were collected. This study found that low-frequency noise has a more negative impact on mental workload. The frontal and left occipital-parietal brain regions show more responses to changes in mental workload. This study is expected to help in the development of mental workload assessment models and construction noise regulations for safety interventions.</div></div>","PeriodicalId":55506,"journal":{"name":"Applied Acoustics","volume":"232 ","pages":"Article 110571"},"PeriodicalIF":3.4,"publicationDate":"2025-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143261700","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"物理与天体物理","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0