scholarly journals Pipeline Leak Aperture Recognition Based on Wavelet Packet Analysis and a Deep Belief Network with ICR

2018 ◽  
Vol 2018 ◽  
pp. 1-8 ◽  
Author(s):  
Xianming Lang ◽  
Zhiyong Hu ◽  
Ping Li ◽  
Yan Li ◽  
Jiangtao Cao ◽  
...  
Keyword(s):  
Sound Velocity ◽  
Wavelet Packet ◽  
Ultrasonic Signal ◽  
Deep Belief Network ◽  
Independent Component ◽  
Fine Tuning ◽  
Local Optimum ◽  
Wavelet Packet Analysis ◽  
Belief Network ◽  
Oil Pipeline

The leakage aperture cannot be easily identified, when an oil pipeline has small leaks. To address this issue, a leak aperture recognition method based on wavelet packet analysis (WPA) and a deep belief network (DBN) with independent component regression (ICR) is proposed. WPA is used to remove the noise in the collected sound velocity of the ultrasonic signal. Next, the denoised sound velocity of the ultrasonic signal is input into the deep belief network with independent component regression (DBNICR) to recognize different leak apertures. Because the optimization of the weights of the DBN with the gradient leads to a local optimum and a slow learning rate, ICR is used to replace the gradient fine-tuning method in conventional DBN for improving the classification accuracy, and a Lyapunov function is constructed to prove the convergence of the DBNICR learning process. By analyzing the acquired ultrasonic sound velocity of different leak apertures, the results show that the proposed method can quickly and effectively identify different leakage apertures.

Coalbed methane content prediction using deep belief network

2020 ◽  
Vol 8 (2) ◽  
pp. T309-T321
Author(s):  
Fan Peng ◽  
Suping Peng ◽  
Wenfeng Du ◽  
Hongshuan Liu
Keyword(s):  
Machine Learning ◽  
Coal Seam ◽  
Coalbed Methane ◽  
Deep Belief Network ◽  
Training Data ◽  
Fine Tuning ◽  
Support Vector ◽  
Learning Models ◽  
Belief Network ◽  
Machine Learning Models

Accurate measurement of coalbed methane (CBM) content is the foundation for CBM resource exploration and development. Machine-learning techniques can help address CBM content prediction tasks. Due to the small amount of actual measurement data and the shallow model structure, however, the results from traditional machine-learning models have errors to some extent. We have developed a deep belief network (DBN)-based model with the input as continuous real values and the activation function as the rectified linear unit. We first calculated a variety of seismic attributes of the target coal seam to highlight the features of the coal seam, then we preprocessed the original attribute features, and finally developed the performance of the DBN model using the preprocessed features. We used 23,374 training data to train our model, 23,240 for pretraining, and 134 for fine-tuning. For the purpose of demonstrating the advantages of the DBN model, we compared it with two typical machine-learning models, including the multilayer perceptron model and the support vector regression model. These two models were trained based on the same labeled training data. The results, obtained from different models, indicated that the DBN model has the least error, which means that it is more accurate than the other two models when used to predict CBM content.

scholarly journals Self-Organizing Traffic Flow Prediction with an Optimized Deep Belief Network for Internet of Vehicles

Sensors ◽  
2018 ◽  
Vol 18 (10) ◽  
pp. 3459 ◽  
Author(s):  
Shidrokh Goudarzi ◽  
Mohd Kama ◽  
Mohammad Anisi ◽  
Seyed Soleymani ◽  
Faiyaz Doctor
Keyword(s):  
Traffic Flow ◽  
Traffic Congestion ◽  
Deep Belief Network ◽  
Fine Tuning ◽  
Superior Performance ◽  
Traffic Information ◽  
Series Data ◽  
Internet Of Vehicles ◽  
Belief Network ◽  
Self Organizing

To assist in the broadcasting of time-critical traffic information in an Internet of Vehicles (IoV) and vehicular sensor networks (VSN), fast network connectivity is needed. Accurate traffic information prediction can improve traffic congestion and operation efficiency, which helps to reduce commute times, noise and carbon emissions. In this study, we present a novel approach for predicting the traffic flow volume by using traffic data in self-organizing vehicular networks. The proposed method is based on using a probabilistic generative neural network techniques called deep belief network (DBN) that includes multiple layers of restricted Boltzmann machine (RBM) auto-encoders. Time series data generated from the roadside units (RSUs) for five highway links are used by a three layer DBN to extract and learn key input features for constructing a model to predict traffic flow. Back-propagation is utilized as a general learning algorithm for fine-tuning the weight parameters among the visible and hidden layers of RBMs. During the training process the firefly algorithm (FFA) is applied for optimizing the DBN topology and learning rate parameter. Monte Carlo simulations are used to assess the accuracy of the prediction model. The results show that the proposed model achieves superior performance accuracy for predicting traffic flow in comparison with other approaches applied in the literature. The proposed approach can help to solve the problem of traffic congestion, and provide guidance and advice for road users and traffic regulators.

Petroleum Pipe Leakage Detection and Location Embeded in SCADA

2004 ◽  
Author(s):  
Likun Wang ◽  
Jian Li ◽  
Ke Peng ◽  
Shijiu Jin ◽  
Zhuang Li
Keyword(s):  
Wavelet Analysis ◽  
Data Exchange ◽  
Wavelet Packet ◽  
Economic Loss ◽  
Frequency Component ◽  
False Alarms ◽  
Leakage Detection ◽  
Wavelet Packet Analysis ◽  
Oil Pipeline ◽  
Scada System

With the increase of the age of the transport oil pipeline and the man-made destruction to pipeline, leaks are often found. The system for pipeline leakage detection and location must be established to find leakage and locate the leak positions to reduce serious environmental pollution and economic loss caused by leakage. The negative pressure wave method is an effective way to locate the leak position, because over 98 percent pipe leakage in China is paroxysmal. There is a SCADA (supervisory control and data acquisition) system to monitor operation for long transport petroleum pipe, but the function of leakage detection and location is not included in existing SCADA system in China. This paper used Dynamic Data Exchange (DDE) method to obtain pipe operation parameters such as pressure, flow rate, temperature, bump current, valve position and so on from the SCADA system. That takes full advantage of the abundant data collection function of the SCADA system to provide data for leakage detection and location. The wavelet packet analysis-based fault diagnosis method can directly use the change of parameters such as energy of frequency component to detect faults without system model. In the paper, a wavelet packet analysis-based characteristic extraction method is used to extract the characteristic information of leak pressure signals. The eigenvector indexes along with the parameters obtained from the SCADA system can be used to avoid false alarms. Wavelet analysis was used to locate leak positions accurately in this paper. Such a wavelet analysis-based leakage detection and location scheme embedded in the SCADA system has been successfully applied to a pipeline in PetroChina. Practical run demonstrated its well effect.

Fault diagnosis of rolling bearing based on multimodal data fusion and deep belief network

2021 ◽  
pp. 095440622110084
Author(s):  
Defeng Lv ◽  
Huawei Wang ◽  
Changchang Che
Keyword(s):  
Fault Diagnosis ◽  
Data Fusion ◽  
Rolling Bearing ◽  
Typical Feature ◽  
Deep Belief Network ◽  
Fine Tuning ◽  
Multimodal Data ◽  
Belief Network ◽  
Time Frequency ◽  
Multimodal Data Fusion

Aiming at raw vibration signal of rolling bearing with long time series, a fault diagnosis model based on multimodal data fusion and deep belief network is proposed in this paper. First, multimodal data composed of artificial features and model features can be obtained by time-frequency domain analysis and unsupervised learning based on restricted Boltzmann machine (RBM). Second, canonical correlation analysis method is used to extract the typical feature pairs from the multimodal data to realize the feature-level multimodal data fusion. Third, deep belief network is applied to extract deep feature mapping between typical feature pairs and fault types. After greedy layer-wise pre-training and fine-tuning, it is available to achieve the trained model for fault diagnosis of rolling bearing. Typical rolling bearing datasets are used to testify the effectiveness of the proposed method. It is verified that the robustness and accuracy of the proposed method are superior to common methods.

scholarly journals Research on DBN-based Evaluation of Distribution Network Reliability

2021 ◽  
Vol 242 ◽  
pp. 03004
Author(s):  
Kaiyu Zhang ◽  
Shanshan Shi ◽  
Shu Liu ◽  
Junjie Wan ◽  
Lijia Ren
Keyword(s):  
Reliability Analysis ◽  
Network Reliability ◽  
Distribution Network ◽  
Simulation Method ◽  
Deep Belief Network ◽  
Fine Tuning ◽  
Layer By Layer ◽  
Boltzmann Machine ◽  
Training Process ◽  
Belief Network

In order to accurately and efficiently analyze the reliability of distribution network, this paper proposes a method of analyzing the reliability of distribution network based on a deep belief network. The Deep Belief Network (DBN) is composed of limiting Boltzmann machine layer-by-layer stacking. It has a strong advantage of automatic feature extraction, which overcomes the shortcomings of traditional neural networks in extracting data features. The entire training process of DBN can be roughly divided into two stages: pre-training and fine-tuning.First of all, the pre-training of the DBN model is realized by training the Restricted Boltzmann Machine (RBM) layer by layer, then the BP algorithm is used for reverse fine-tuning to complete the training process of the entire network. finally, the reliability analysis of distribution network is performed by the trained DBN. Compared with the BP neural network method and the traditional Monte Carlo simulation method, it is verified that the proposed model of distribution network reliability analysis has high accuracy.

scholarly journals Deep Belief Network for Feature Extraction of Urban Artificial Targets

2020 ◽  
Vol 2020 ◽  
pp. 1-13
Author(s):  
Xiaoai Dai ◽  
Junying Cheng ◽  
Yu Gao ◽  
Shouheng Guo ◽  
Xingping Yang ◽  
...  
Keyword(s):  
Hyperspectral Image ◽  
Image Data ◽  
Original Data ◽  
Deep Belief Network ◽  
Fine Tuning ◽  
Model Parameters ◽  
Layer By Layer ◽  
Restricted Boltzmann Machines ◽  
Belief Network ◽  
Network Algorithm

Reducing the dimension of the hyperspectral image data can directly reduce the redundancy of the data, thus improving the accuracy of hyperspectral image classification. In this paper, the deep belief network algorithm in the theory of deep learning is introduced to extract the in-depth features of the imaging spectral image data. Firstly, the original data is mapped to feature space by unsupervised learning methods through the Restricted Boltzmann Machine (RBM). Then, a deep belief network will be formed by superimposed multiple Restricted Boltzmann Machines and training the model parameters by using the greedy algorithm layer by layer. At the same time, as the objective of data dimensionality reduction is achieved, the underground feature construction of the original data will be formed. The final step is to connect the depth features of the output to the Softmax regression classifier to complete the fine-tuning (FT) of the model and the final classification. Experiments using imaging spectral data showing the in-depth features extracted by the profound belief network algorithm have better robustness and separability. It can significantly improve the classification accuracy and has a good application prospect in hyperspectral image information extraction.

Sign in / Sign up

Export Citation Format

Share Document