Evaluation of vehicle interior sound quality using a continuous restricted Boltzmann machine-based DBN

Huang, Hai; Li, Ren X.; Yang, Ming; Lim, Teik C.; Ding, Wei P.

doi:10.1016/j.ymssp.2016.07.014

Cited by 81 publications

(32 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The difference between the limited Boltzmann machine and the Boltzmann machine is that there is no connection between the same layer. The restricted Boltzmann machine is divided into visible layer and hidden layer [8][9] [10], as shown in Fig.3.…”

Section: Restricted Boltzmann Machinementioning

confidence: 99%

Intelligent Defect Detection Method of Photovoltaic Modules Based on Deep Learning

Ni¹,

Zou²,

Li³

et al. 2018

Proceedings of the 2018 International Conference on Transportation &Amp; Logistics, Information &Amp; Communication, Smart City

View full text Add to dashboard Cite

Currently, photovoltaic module manufacturers still rely on manual detection of EL images of photovoltaic modules to identify hidden defects. EL image detection is an important link in the quality control of photovoltaic modules production. Manual detection leads to slow detection speed, and the accuracy is affected by personal subjective judgment. In this paper, an intelligent defect detection method based on deep learning is proposed. The method first builds a network according to the sample characteristics. The initial network value is obtained through training. Then, the neural algorithm is used to adjust the network parameters to obtain the mapping relationship between training samples and defect-free templates. Finally, the comparison between reconstructed image and defect image is used to realize defect detection of test samples. Experiments show that the method based on deep learning can detect defects accurately and quickly.

show abstract

Section: Restricted Boltzmann Machinementioning

confidence: 99%

Intelligent Defect Detection Method of Photovoltaic Modules Based on Deep Learning

Ni¹,

Zou²,

Li³

et al. 2018

Proceedings of the 2018 International Conference on Transportation &Amp; Logistics, Information &Amp; Communication, Smart City

View full text Add to dashboard Cite

show abstract

“…Zwicker and Moore-Glasberg are two main loudness models. The total Zwicker loudness has been applied to not only the radiated noise of internal combustion engines [1][2][3], but also the interior noise of gasoline vehicles [4][5][6], electric vehicles, and high-speed trains [7,8].…”

Section: Introductionmentioning

confidence: 99%

Evaluation and Improvement of the Sound Quality of a Diesel Engine Based on Tests and Simulations

Yang

Feng

et al. 2020

Energies

View full text Add to dashboard Cite

Traditional acoustic evaluation of a diesel engine generally uses the A-weighted sound pressure level (AWSPL) and radiated sound power to assess the noise of an engine prototype present in an experiment. However, this cannot accurately and comprehensively reflect the auditory senses of human subjects during the simulation stage. To overcome such shortage, the Moore–Glasberg loudness and sharpness approach is applied to evaluate and improve the sound quality (SQ) of a 16 V-type marine diesel engine, and synthesizing noise audio files. Through finite element (FE) simulations, the modes of the engine’s block and the average vibrational velocity of the entire engine surface were calculated and compared with the test results. By further applying an automatically matched layer (AML) approach, the engine-radiated sound pressure level (SPL) and sound power contributions of all engine parts were obtained. By analyzing the Moore–Glasberg loudness and sharpness characteristics of three critical sound field points, an improvement strategy of the oil sump was then proposed. After improvement, both the loudness and sharpness decreased significantly. To verify the objective SQ evaluation results, ten noise audio clips of the diesel engine were then synthesized and tested. The subjective evaluation results were in accordance with the simulated analysis. Therefore, the proposed approach to analyze and improve the SQ of a diesel engine is reliable and effective.

show abstract

“…Many studies are currently using deep learning for time series forecasts [21][22][23][24]. Some results reveal a DBN model's superiority over a linear autoregressive and a conventional back-propagation neural network (i.e., ANN) model.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning Neural Networks Trained with MODIS Satellite-Derived Predictors for Long-Term Global Solar Radiation Prediction

et al. 2019

View full text Add to dashboard Cite

Solar energy predictive models designed to emulate the long-term (e.g., monthly) global solar radiation (GSR) trained with satellite-derived predictors can be employed as decision tenets in the exploration, installation and management of solar energy production systems in remote and inaccessible solar-powered sites. In spite of a plethora of models designed for GSR prediction, deep learning, representing a state-of-the-art intelligent tool, remains an attractive approach for renewable energy exploration, monitoring and forecasting. In this paper, algorithms based on deep belief networks and deep neural networks are designed to predict long-term GSR. Deep learning algorithms trained with publicly-accessible Moderate Resolution Imaging Spectroradiometer (MODIS) satellite data are tested in Australia’s solar cities to predict the monthly GSR: single hidden layer and ensemble models. The monthly-scale MODIS-derived predictors (2003–2018) are adopted, with 15 diverse feature selection approaches including a Gaussian Emulation Machine for sensitivity analysis used to select optimal MODIS-predictor variables to simulate GSR against ground-truth values. Several statistical score metrics are adopted to comprehensively verify surface GSR simulations to ascertain the practicality of deep belief and deep neural networks. In the testing phase, deep learning models generate significantly lower absolute percentage bias (≤3%) and high Kling–Gupta efficiency (≥97.5%) values compared to the single hidden layer and ensemble model. This study ascertains that the optimal MODIS input variables employed in GSR prediction for solar energy applications can be relatively different for diverse sites, advocating a need for feature selection prior to the modelling of GSR. The proposed deep learning approach can be adopted to identify solar energy potential proactively in locations where it is impossible to install an environmental monitoring data acquisition instrument. Hence, MODIS and other related satellite-derived predictors can be incorporated for solar energy prediction as a strategy for long-term renewable energy exploration.

show abstract

Evaluation of vehicle interior sound quality using a continuous restricted Boltzmann machine-based DBN

Cited by 81 publications

References 33 publications

Intelligent Defect Detection Method of Photovoltaic Modules Based on Deep Learning

Intelligent Defect Detection Method of Photovoltaic Modules Based on Deep Learning

Evaluation and Improvement of the Sound Quality of a Diesel Engine Based on Tests and Simulations

Deep Learning Neural Networks Trained with MODIS Satellite-Derived Predictors for Long-Term Global Solar Radiation Prediction

Contact Info

Product

Resources

About