How Deep Is Deep Enough for Deep Belief Network for Approximating Model Predictive Control Law

Wang, Gongming; Qiao, Junfei; Liu, Caixia; Shen, Zhaoxu

doi:10.1109/tase.2021.3074984

Cited by 7 publications

(4 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based upon the definition of input and output variables, the entire data-driven power flow calculation model is further constructed using Deep Belief Networks (DBN) (Zhang et al, 2018). DBN, as a form of deep learning, consists of multiple layers of Restricted Boltzmann Machines (RBM) (Zhang et al, 2018;Tao et al, 2020;Wang et al, 2022). In this network architecture, there are connections between layers, but units within each layer are not interconnected.…”

Section: The Establishment Of the Overall Modelmentioning

confidence: 99%

See 1 more Smart Citation

Research on data-driven power flow calculation method based on undirected-graph delooping-backtracking

Zhu,

Hu,

Liu

et al. 2024

Front. Energy Res.

View full text Add to dashboard Cite

As the scale of the power grid expands and distributed energy sources are integrated, along with the emergence of random loads, topological control of distribution networks has become a novel means of control. Therefore, data-driven power flow calculations must be capable of rapidly and accurately computing power flow results even when there are changes in the network’s topology. In this paper, a data-driven power flow calculation method is proposed to take topological changes into account. Based on initial loop data, we employ an undirected-graph delooping-backtracking method to generate a set of feasible topological samples. Using the Monte Carlo method on this basis, we generate feasible samples for the network’s topology and power injection, thereby establishing a training dataset. By training a deep neural network on these samples and adjusting network parameters, we effectively address power flow calculations in the presence of topological changes. Case study results demonstrate that the data-driven power flow calculation method, considering topological changes, can rapidly and accurately compute power flow results when topology alterations occur.

show abstract

Section: The Establishment Of the Overall Modelmentioning

confidence: 99%

“…In this network architecture, there are connections between layers, but units within each layer are not interconnected. After training the neural network parameters layer by layer, DBNs are effective in fitting a large number of data samples, enabling estimation and prediction tasks (Wang et al, 2022).…”

Section: The Establishment Of the Overall Modelmentioning

confidence: 99%

Research on data-driven power flow calculation method based on undirected-graph delooping-backtracking

Zhu,

Hu,

Liu

et al. 2024

Front. Energy Res.

View full text Add to dashboard Cite

show abstract

“…The architecture of a DNN depends significantly upon the complexity of the problem under consideration; it is related to how each layer is implemented, as well as the computational method used in each layer. The most widely used DL approaches are CNNs [40], recurrent neural networks (RNNs) [41], deep belief networks (DBNs) [42], and the recently introduced approach of generative adversarial networks (GANs) [43]. However, the most popular supervised DNNs for image processing, object recognition, image formation, and classification are based on CNN architectures [44].…”

Section: Deep Neural Networkmentioning

confidence: 99%

Model-Based Data Augmentation Applied to Deep Learning Networks for Classification of Micro-Doppler Signatures Using FMCW Radar

Rojhani

Passafiume

Sadeghibakhi

et al. 2023

IEEE Trans. Microwave Theory Techn.

View full text Add to dashboard Cite

Deep neural networks (DNNs) have become a relevant subject in the classification of radio frequency signals and remote sensing data. A primary challenge is a tradeoff between obtaining data that are suitable for DNN training and the effort that making experimental measurements requires. Hence, the quality and quantity of data used for the training and testing of models are crucial for effective classifier development. The training dataset should cover a wide range of cases that synthesize the actual scenarios being classified. This work proposes a novel data augmentation method based on a deterministic model to generate a simulated dataset of radar micro-Doppler signatures suitable for unmanned aerial vehicle (UAV) target classification, without requiring measurement data. It is shown that the DNN trained using the properly generated model-based data offers improved classification accuracy performance. Resultsare presented for a two-class classification of the number of UAV motors using a 77-GHz frequency-modulated continuouswave (FMCW) automotive radar system. The effectiveness of the proposed methodology is proven: a classification accuracy of 78.68% is achieved using a convolutional neural network (CNN) trained using the synthetic dataset, while an accuracy of 66.18% is achieved by using a typical signal processing data augmentation method on a limited measured dataset.

show abstract

“…In return, the DNN can provide a probability of each class discriminating normal and attack packets (Deng, Gao, Lu, and Gao, 2018), thus, the sensor can identify malicious attacks on the vehicle (Deng, Gao, Lu, & Gao, 2018;Eziama, Awin, Ahmed, Santos-Jaimes, Pelumi, & Corral-De-Witt, 2020). The research conducted an unsupervised pre-training method of deep belief networks (Wang, Qiao, Liu, and Shen, 2021), to include the stochastic gradient descent method (Sun, Qiao, Liao, and Li, 2020) can extract in-vehicular network packets (Wang, Qiao, Liu, and Shen, 2021;Sun, Quio, Liao, & Li, 2020). The research outlines related work for CAN protocol, which includes intrusion detection and machine learning, deep learning, a proposed technique for a proposed intrusion detection system, CAN packet feature, attack detection, data set, and performance evaluation.…”

Section: Safety Algorithm Contributorsmentioning

confidence: 99%

Exploring Vehicle Security Improvement Framework in the Safety Messaging System to Protect Against Hackers

2023

IJFMR

View full text Add to dashboard Cite

The transportation and automotive industry have seen the shifts and effects of innovation within the last decade, from the addition of critical messaging technologies, event response systems, to safety integration of each user initiated by safety messaging systems to almost all private utility vehicles to a public utility vehicle, all have seen either responsive or unresponsive layers of usability and functionality issues, together with the need for safety and protection of each users and riders. These assistive technologies are supposed to help drivers and operators in their daily driving routine, believed to be connected to a myriad of networked hardware and software, and arguably to be either be cost effective or safety limited. The study exploring vehicle security improvement framework in the safety messaging system to protect against hackers delivered responses from 6 subject matter experts within the public transit authority, cloud supply chain companies, and network providers that have invested greatly in vehicle ad hoc network technologies. Thematic analysis was chosen as a qualitative method design to extract themes from the critical answers through the semi-structured interview retrieved from general responses from the subject matter experts. The general findings were extracted from the major themes 1 through 5. The general findings are the following (1) Automotive Organizations and Transit Authority Leaders have to Manage Vehicle Cyber Risks Throughout the Vehicle Lifecycle, (2) Automotive Organization and Transit Authority Should Be Involved in Engineering a Secure Vehicle by Design, (3) Engineering Vendor Teams Should Detect and Respond to Major Incidents within the Vehicle Lifecycle.

show abstract

How Deep Is Deep Enough for Deep Belief Network for Approximating Model Predictive Control Law

Cited by 7 publications

References 46 publications

Research on data-driven power flow calculation method based on undirected-graph delooping-backtracking

Research on data-driven power flow calculation method based on undirected-graph delooping-backtracking

Model-Based Data Augmentation Applied to Deep Learning Networks for Classification of Micro-Doppler Signatures Using FMCW Radar

Exploring Vehicle Security Improvement Framework in the Safety Messaging System to Protect Against Hackers

Contact Info

Product

Resources

About