Transfer learning-based thermal error prediction and control with deep residual LSTM network

2021 ◽  
pp. 107704
Author(s):  
Jialan Liu ◽  
Chi Ma ◽  
Hongquan Gui ◽  
Shilong Wang
Keyword(s):  
Transfer Learning ◽  
Thermal Error ◽  
Error Prediction ◽  
Prediction And Control ◽  
Lstm Network ◽  
And Control

Prediction and Control for Profile Angle Error of Slotting Cutter of Herringbone Gear in Turbine Redactor

2010 ◽  
Vol 139-141 ◽  
pp. 1864-1868 ◽  
Author(s):  
Bi Wen Li ◽  
Jin Feng Xiao ◽  
Chun Liang Zhang ◽  
Liang Bin Hu
Keyword(s):  
Nuclear Power ◽  
Error Control ◽  
Economic Benefits ◽  
Error Prediction ◽  
Frequency Noise ◽  
Angle Error ◽  
Profile Angle ◽  
Prediction And Control ◽  
Herringbone Gear ◽  
And Control

Relative to gear shaping and gear hob, using gear slotting produces herringbone gear which is used in nuclear power turbine speed redactor has more obvious technical and economic benefits, but profile angle error of slotting cutter causes profile error and base pitch deviation of herringbone gear. It will result in high-frequency noise which damages the tactical and technical performance of warship. By analyzed the wire cutting system of rack cutter used in MAAG type gear machining for herringbone gear, the mathematical model of error prediction for rack cutter profile angle was founded. On the base of model of error prediction, the principle of error control and the method of revision control are presented. Finally, the example of prediction and control is provided.

An insight into Case Management and Challenges in Veteran Suicide Prevention

2020 ◽  
pp. 14-25
Author(s):  
Tamara Green
Keyword(s):  
Case Management ◽  
Suicide Prevention ◽  
Suicide Risk ◽  
Digital Transformation ◽  
Prevention Science ◽  
Historical Records ◽  
Military Members ◽  
Prediction And Control ◽  
And Control ◽  
Insight Into

Much of the literature, policies, programs, and investment has been made on mental health, case management, and suicide prevention of veterans. The Australian “veteran community is facing a suicide epidemic for the reasons that are extremely complex and beyond the scope of those currently dealing with them.” (Menz, D: 2019). Only limited work has considered the digital transformation of loosely and manual-based historical records and no enablement of Artificial Intelligence (A.I) and machine learning to suicide risk prediction and control for serving military members and veterans to date. This paper presents issues and challenges in suicide prevention and management of veterans, from the standing of policymakers to stakeholders, campaigners of veteran suicide prevention, science and big data, and an opportunity for the digital transformation of case management.

A case study of wear-type rail corrugation prediction and control using speed variation

2009 ◽  
Vol 325 (1-2) ◽  
pp. 85-105 ◽  
Author(s):  
P.A. Meehan ◽  
P.A. Bellette ◽  
R.D. Batten ◽  
W.J.T. Daniel ◽  
R.J. Horwood
Keyword(s):  
Rail Corrugation ◽  
Speed Variation ◽  
Prediction And Control ◽  
And Control

Phenomenological Foundations for the Study of Suicide

1973 ◽  
Vol 4 (3) ◽  
pp. 195-208
Author(s):  
Keith Hoeller
Keyword(s):  
Great Emphasis ◽  
Prediction And Control ◽  
And Control

Is death the “enemy” to be avoided at all costs or is it to be faced, engendering liberation and rebirth? Contemporary suicidology concerns itself with the “causes” of suicide, placing great emphasis on prediction and control However, when the “meaning” of suicide is studied, understanding it as a human phenomenon becomes of major concern. Part of this understanding requires one to view “dread” as implying the possibility of making one's existence one's own, rather than something that must be prevented. In the study of suicide, revolutionary insights can emerge if less emphasis is placed on death as the “enemy” and more attention is placed on “dread” as a potential liberator.

scholarly journals Real-Time Detection of Dictionary DGA Network Traffic Using Deep Learning

2021 ◽  
Vol 2 (2) ◽  
Author(s):  
Kate Highnam ◽  
Domenic Puzio ◽  
Song Luo ◽  
Nicholas R. Jennings
Keyword(s):  
Neural Network ◽  
Deep Learning ◽  
Real Time ◽  
Network Traffic ◽  
Short Term Memory ◽  
Domain Names ◽  
Control Networks ◽  
Detection Techniques ◽  
Lstm Network ◽  
And Control

AbstractBotnets and malware continue to avoid detection by static rule engines when using domain generation algorithms (DGAs) for callouts to unique, dynamically generated web addresses. Common DGA detection techniques fail to reliably detect DGA variants that combine random dictionary words to create domain names that closely mirror legitimate domains. To combat this, we created a novel hybrid neural network, Bilbo the “bagging” model, that analyses domains and scores the likelihood they are generated by such algorithms and therefore are potentially malicious. Bilbo is the first parallel usage of a convolutional neural network (CNN) and a long short-term memory (LSTM) network for DGA detection. Our unique architecture is found to be the most consistent in performance in terms of AUC, $$F_1$$ F 1 score, and accuracy when generalising across different dictionary DGA classification tasks compared to current state-of-the-art deep learning architectures. We validate using reverse-engineered dictionary DGA domains and detail our real-time implementation strategy for scoring real-world network logs within a large enterprise. In 4 h of actual network traffic, the model discovered at least five potential command-and-control networks that commercial vendor tools did not flag.

scholarly journals Subgraphs of Interest Social Networks for Diffusion Dynamics Prediction

Entropy ◽  
2021 ◽  
Vol 23 (4) ◽  
pp. 492
Author(s):  
Valentina Y. Guleva ◽  
Polina O. Andreeva ◽  
Danila A. Vaganov
Keyword(s):  
Social Networks ◽  
Sampling Method ◽  
Building Blocks ◽  
Global Dynamics ◽  
Diffusion Dynamics ◽  
Different Types ◽  
Prediction And Control ◽  
Path Lengths ◽  
And Control ◽  
Diffusion Prediction

Finding the building blocks of real-world networks contributes to the understanding of their formation process and related dynamical processes, which is related to prediction and control tasks. We explore different types of social networks, demonstrating high structural variability, and aim to extract and see their minimal building blocks, which are able to reproduce supergraph structural and dynamical properties, so as to be appropriate for diffusion prediction for the whole graph on the base of its small subgraph. For this purpose, we determine topological and functional formal criteria and explore sampling techniques. Using the method that provides the best correspondence to both criteria, we explore the building blocks of interest networks. The best sampling method allows one to extract subgraphs of optimal 30 nodes, which reproduce path lengths, clustering, and degree particularities of an initial graph. The extracted subgraphs are different for the considered interest networks, and provide interesting material for the global dynamics exploration on the mesoscale base.

Sign in / Sign up

Export Citation Format

Share Document