Deep learning for smart manufacturing: Methods and applications

Wang, Jinjiang; Ma, Yukun; Zhang, Laibin; Gao, Robert X.; Wu, Dazhong

doi:10.1016/j.jmsy.2018.01.003

Cited by 1,245 publications

(533 citation statements)

References 71 publications

Supporting

Mentioning

522

Contrasting

Unclassified

Order By: Relevance

“…The green dots denote the input power while the blue dots represent the power output. It can be observed that both the power input and output tend to maintain a positive correlation with the wave height when it is higher than 0.2 m. The positive correlation diverges when the wave height is higher than 0.6 m. In general, these trends coincide with calculations using the wave energy [30] that varies with the square of wave height. It can also be seen that the device remains inactive when the wave height is below approximately 0.25 m, indicating the start wave height of this device is 0.25 m. When comparing these two power curves, it is found that the efficiency from wave energy to hydraulic power output shows little difference between 0.2 m and 0.6 m. Nevertheless, it increases smoothly when the wave height is higher than 0.6 m; this could reveal the mechanism of input and output power efficiency of this particular device.…”

Section: Power Curvessupporting

confidence: 66%

Prediction of Wave Power Generation Using a Convolutional Neural Network with Multiple Inputs

2018

Energies

View full text Add to dashboard Cite

Successful development of a marine wave energy converter (WEC) relies strongly on the development of the power generation device, which needs to be efficient and cost-effective. An innovative multi-input approach based on the Convolutional Neural Network (CNN) is investigated to predict the power generation of a WEC system using a double-buoy oscillating body device (OBD). The results from the experimental data show that the proposed multi-input CNN performs much better at predicting results compared with the conventional artificial network and regression models. Through the power generation analysis of this double-buoy OBD, it shows that the power output has a positive correlation with the wave height when it is higher than 0.2 m, which becomes even stronger if the wave height is higher than 0.6 m. Furthermore, the proposed approach associated with the CNN algorithm in this study can potentially detect the changes that could be due to presence of anomalies and therefore be used for condition monitoring and fault diagnosis of marine energy converters. The results are also able to facilitate controlling of the electricity balance among energy conversion, wave power produced and storage.

show abstract

Section: Power Curvessupporting

confidence: 66%

Prediction of Wave Power Generation Using a Convolutional Neural Network with Multiple Inputs

2018

Energies

View full text Add to dashboard Cite

show abstract

“…Its deep architecture, with many hidden layers, is a multi-level non-linear operation, transferring each layer's features from original input into more abstracted features in the higher layers to find complicated inherent structures. Unlike DNNs, shallow NNs approaches the extraction of the features and the construction of the model are done separately, where each module is formed by step-bystep model training [Wan18].…”

Section: Deep Learning For Map Digitizationmentioning

confidence: 99%

“…A CNN is a multi-layer feed-forward artificial deep neural network that has fewer connections and parameters than other NN architectures originally intended for two-dimensional image processing [LeC98], [Wan18], [Kri12]. The structure of CNN is illustrated in Figure 3.…”

Section: Map Digitization: Cnn Approachmentioning

confidence: 99%

“…The convolutional layers convolve multiple local kernel filters with input data in order to generate invariant local features [Wan18]. The pooling layers extract the most significant features over sliding windows, which have a fixedlength using different pooling operations.…”

Section: Map Digitization: Cnn Approachmentioning

confidence: 99%

“…Typical pooling operations are average pooling and max pooling. Average pooling calculates the mean value of the neurons in the region (sliding window) and takes it as the output value for that region [Wan18]. Not surprisingly, max pooling selects the maximum value in the region, so the pooling output is the maximum activation over rectangular regions that do not overlap [Cir12].…”

Section: Map Digitization: Cnn Approachmentioning

confidence: 99%

See 2 more Smart Citations

Deep Learning for Historical Cadastral Maps Digitization: Overview, Challenges and Potential

Ignjatić¹,

Nikolić²,

Rikalović³

2018

CSRN

View full text Add to dashboard Cite

Cartographic heritage of historical cadastral maps represent remarkable geospatial data. Historical cadastral maps are generally regarded as an essential part of the land management infrastructure (buildings, streets, canals, bridges, etc.). Today these cadastral maps are still in use in a digital raster form (scanned maps). Digitization of cadastral maps is time consuming and it is a challenge for scientists and engineers to find ways to automatically convert raster into vector maps. The process of map digitization typically involves several stages: preprocessing, visual object detection and classification, vector representation postprocessing and extracting information from text. Although neural networks have had a long history of use in the domain, their applications remain limited to extracting the information from text. Recent convergence of advancements in the domains of training deep neural networks (DNN) and GPU hardware allowed DNNs to achieve state-of-the-art results in computer vision applications, beyond hand-written text recognition. This paper provides an overview of different approaches to historical cadastral maps digitization, focusing of the challenges and the potential of using deep neural networks in map digitization.

show abstract

Computing and Service Architecture for Intelligent IoT

Yang

Tan

et al. 2021

Intelligent IoT for the Digital World

View full text Add to dashboard Cite

Deep learning for smart manufacturing: Methods and applications

Cited by 1,245 publications

References 71 publications

Prediction of Wave Power Generation Using a Convolutional Neural Network with Multiple Inputs

Prediction of Wave Power Generation Using a Convolutional Neural Network with Multiple Inputs

Deep Learning for Historical Cadastral Maps Digitization: Overview, Challenges and Potential

Computing and Service Architecture for Intelligent IoT

Contact Info

Product

Resources

About