The Intelligent Diagnosis of a Hydraulic Plunger Pump Based on the MIGLCC-DLSTM Method Using Sound Signals DOI Creative Commons
L. L. Ma,

Anqi Jiang,

Wanlu Jiang

и другие.

Machines, Год журнала: 2024, Номер 12(12), С. 869 - 869

Опубликована: Ноя. 29, 2024

To fully exploit the rich state and fault information embedded in acoustic signals of a hydraulic plunger pump, this paper proposes an intelligent diagnostic method based on sound signal analysis. First, were collected under normal various conditions. Then, four distinct features—Mel Frequency Cepstral Coefficients (MFCCs), Inverse Mel (IMFCCs), Gammatone (GFCCs), Linear Prediction (LPCCs)—were extracted integrated into novel hybrid cepstral feature called MIGLCCs. This fusion enhances model’s ability to distinguish both high- low-frequency characteristics, resist noise interference, capture resonance peaks, achieving complementary advantage. Finally, MIGLCC set was input double layer long short-term memory (DLSTM) network enable recognition pump’s operational states. The results indicate that MIGLCC-DLSTM achieved accuracy 99.41% test Validation CWRU bearing dataset data from high-pressure servo motor turbine system yielded overall accuracies 99.64% 98.07%, respectively, demonstrating robustness broad application potential method.

Язык: Английский

A Deep Learning Framework for Corrosion Assessment of Steel Structures Using Inception v3 Model DOI Creative Commons

Xinghong Huang,

Zhenhua Duan,

Shaojin Hao

и другие.

Buildings, Год журнала: 2025, Номер 15(4), С. 512 - 512

Опубликована: Фев. 7, 2025

Corrosion detection plays a crucial role in the effective lifecycle management of steel structures, significantly impacting maintenance strategies and operational performance. This study presents machine vision-based approach for classifying corrosion levels Q235 steel, providing valuable insights assessment decision-making. Accelerated salt spray tests were performed to simulate progression over multiple cycles, resulting comprehensive dataset comprising surface images corresponding eight loss measurements. A comparative evaluation with other architectures, namely, AlexNet, ResNet, VggNet, demonstrated that Inception v3 model achieved superior classification accuracy, exceeding 95%. method offers an precise solution evaluation, supporting proactive planning optimal resource allocation throughout structures. By leveraging advanced deep learning techniques, provides scalable efficient framework enhancing sustainability safety infrastructure.

Язык: Английский

Процитировано

2
Enhancing Respiratory Monitoring by CNN Using Mel Frequency Cepstral Coefficients DOI
Yajnaseni Dash,

Ajith Abraham,

Shivam Gupta

и другие.

Lecture notes in networks and systems, Год журнала: 2025, Номер unknown, С. 572 - 580

Опубликована: Янв. 1, 2025

Язык: Английский

Процитировано

0
A Real-Time Intelligent Acoustic IoT-Enabled Embedded Construction Site Monitoring and Alert System: Integrating Deep Learning–Based Machine-Listening Algorithms, Edge Computing, and Cloud Computing DOI
Oscar Poudel, Rayan H. Assaad

Journal of Construction Engineering and Management, Год журнала: 2025, Номер 151(7)

Опубликована: Апрель 24, 2025

Язык: Английский

Процитировано

0
Feature extraction for acoustic leakage detection in water pipelines DOI
Tao An,

Liang Ma,

Dazhi Li

и другие.

Automation in Construction, Год журнала: 2025, Номер 176, С. 106248 - 106248

Опубликована: Май 9, 2025

Язык: Английский

Процитировано

0
SpectroFusionNet a CNN approach utilizing spectrogram fusion for electric guitar play recognition DOI Creative Commons
Ganesh Kumar Chellamani,

N Aishwarya,

C Chandhana

и другие.

Scientific Reports, Год журнала: 2025, Номер 15(1)

Опубликована: Май 15, 2025

Music, a universal language and cultural cornerstone, continues to shape enhance human expression connection across diverse societies. This study introduces SpectroFusionNet, comprehensive deep learning framework for the automated recognition of electric guitar playing techniques. The proposed approach first extracts various spectrograms, including Mel-Frequency Cepstral Coefficients (MFCC), Continuous Wavelet Transform (CWT), Gammatone capture intricate audio features. These spectrograms are then individually processed using lightweight models (MobileNetV2, InceptionV3, ResNet50) extract discriminative features different sounds, with ResNet50 yielding better performance. To further classification performance nine distinct sound classes, two types fusion strategies adopted provide rich feature representation: One is early where combined before extraction other one late independent from concatenated via three approaches: weighted averaging, max-voting simple concatenation. Then, fused subsequently fed into machine classifiers, Support Vector Machine (SVM), Multilayer Perceptron (MLP), Logistic Regression, Random Forest etc., final classification. Experimental results demonstrate that MFCC-Gammatone provided best performance, achieving 99.12% accuracy, 100% precision, recall 9 classes. assess SpectroFusionNet's generalization ability, real-time dataset evaluated, demonstrating an accuracy 70.9%, indicating its applicability in real world scenarios.

Язык: Английский

Процитировано

0
Weighted adaptive active transfer learning for imbalanced multi-object classification in construction site imagery DOI Creative Commons
Karunakar Reddy Mannem, Samuel A. Prieto, Borja García de Soto

и другие.

Automation in Construction, Год журнала: 2025, Номер 176, С. 106297 - 106297

Опубликована: Май 26, 2025

Язык: Английский

Процитировано

0
Action Reliability Assessment Framework for Automated Construction Labor Measurements: Case Study on Plastering Operations DOI

Panming He,

Liping Qin, Yuan Yuan

и другие.

Journal of Construction Engineering and Management, Год журнала: 2025, Номер 151(8)

Опубликована: Май 29, 2025

Язык: Английский

Процитировано

0
A multi modal fusion coal gangue recognition method based on IBWO-CNN-LSTM DOI Creative Commons

Wenchao Hao,

Haiyan Jiang, Qinghui Song

и другие.

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Дек. 5, 2024

Accurate identification of coal and gangue is a crucial guarantee for efficient safe mining top caving face. This article proposes coal-gangue recognition method based on an improved beluga whale optimization algorithm (IBWO), convolutional neural network, long short-term memory network (CNN-LSTM) multi-modal fusion model. First, the mutation library mechanisms are introduced into to explore solution space fully, prevent falling local optimum, accelerate convergence process. Subsequently, image mapping audio signal vibration performed extract Mel-Frequency Cepstral Coefficients (MFCC) features, generating rich sample data CNN-LSTM. Then multi-head attention mechanism CNN-LSTM speed up training improve classification accuracy. Finally, IBWO-CNN-LSTM model constructed by optimal hyperparameter combination obtained IBWO realize automatic coal-gangue. The benchmark function proves that superior other algorithms. By building experimental platform impact tail beam hydraulic support, multiple collection carried out. results show proposed has better performance than models, accuracy rate reaches 95.238%. strategy helps robustness recognition.

Язык: Английский

Процитировано

2
Research on Intelligent Diagnosis of Corrosion in the Operation and Maintenance Stage of Steel Structure Engineering Based on U-Net Attention DOI Creative Commons
Zhenhua Duan,

Xinghong Huang,

Jia Hou

и другие.

Buildings, Год журнала: 2024, Номер 14(12), С. 3972 - 3972

Опубликована: Дек. 14, 2024

Intelligent corrosion diagnosis plays a crucial role in enhancing the efficiency of operation and maintenance for steel structures. Presently, detection primarily depends on manual visual inspections non-destructive testing methods, which are inefficient, costly, subject to human bias. While machine vision has demonstrated significant potential controlled laboratory settings, most studies have focused environments with limited background interference, restricting their practical applicability. To tackle challenges posed by complex backgrounds multiple interference factors field-collected images components, this study introduces an intelligent grading method designed specifically containing elements. By integrating attention mechanism into traditional U-Net network, we achieve precise segmentation component pixels from engineering images, attaining accuracy up 94.1%. The proposed framework is validated using collected actual sites. A sliding window sampling technique divides on-site several rectangular windows, filtered based Attention results. Leveraging dataset plate known grades, train Inception v3 classification model. Transfer learning techniques then applied determine grade each window, culminating weighted average estimate overall target component. This provides quantitative index assessing large-scale structure corrosion, significantly impacting improvement construction quality while laying solid foundation further research development related fields.

Язык: Английский

Процитировано

0
The Intelligent Diagnosis of a Hydraulic Plunger Pump Based on the MIGLCC-DLSTM Method Using Sound Signals DOI Creative Commons
L. L. Ma,

Anqi Jiang,

Wanlu Jiang

и другие.

Machines, Год журнала: 2024, Номер 12(12), С. 869 - 869

Опубликована: Ноя. 29, 2024

To fully exploit the rich state and fault information embedded in acoustic signals of a hydraulic plunger pump, this paper proposes an intelligent diagnostic method based on sound signal analysis. First, were collected under normal various conditions. Then, four distinct features—Mel Frequency Cepstral Coefficients (MFCCs), Inverse Mel (IMFCCs), Gammatone (GFCCs), Linear Prediction (LPCCs)—were extracted integrated into novel hybrid cepstral feature called MIGLCCs. This fusion enhances model’s ability to distinguish both high- low-frequency characteristics, resist noise interference, capture resonance peaks, achieving complementary advantage. Finally, MIGLCC set was input double layer long short-term memory (DLSTM) network enable recognition pump’s operational states. The results indicate that MIGLCC-DLSTM achieved accuracy 99.41% test Validation CWRU bearing dataset data from high-pressure servo motor turbine system yielded overall accuracies 99.64% 98.07%, respectively, demonstrating robustness broad application potential method.

Язык: Английский

Процитировано

0