scholarly journals Design Flow of Accelerating Hybrid Extremely Low Bit-Width Neural Network in Embedded FPGA

2018 ◽  
Author(s):  
Junsong Wang ◽  
Qiuwen Lou ◽  
Xiaofan Zhang ◽  
Chao Zhu ◽  
Yonghua Lin ◽  
...  
Keyword(s):  
Neural Network ◽  
Design Flow

scholarly journals Assessment of a Neural-Network-Based Optimization Tool: A Low Specific-Speed Impeller Application

2011 ◽  
Vol 2011 ◽  
pp. 1-11 ◽  
Author(s):  
Matteo Checcucci ◽  
Federica Sazzini ◽  
Michele Marconcini ◽  
Andrea Arnone ◽  
Mario Coneri ◽  
...  
Keyword(s):  
Neural Network ◽  
Fluid Dynamic ◽  
Optimization Technique ◽  
Three Dimensional ◽  
Design Procedure ◽  
Design Flow ◽  
Centrifugal Impeller ◽  
Geometrical Constraints ◽  
Low Specific Speed ◽  
Specific Speed

This work provides a detailed description of the fluid dynamic design of a low specific-speed industrial pump centrifugal impeller. The main goal is to guarantee a certain value of the specific-speed number at the design flow rate, while satisfying geometrical constraints and industrial feasibility. The design procedure relies on a modern optimization technique such as an Artificial-Neural-Network-based approach (ANN). The impeller geometry is parameterized in order to allow geometrical variations over a large design space. The computational framework suitable for pump optimization is based on a fully viscous three-dimensional numerical solver, used for the impeller analysis. The performance prediction of the pump has been obtained by coupling the CFD analysis with a 1D correlation tool, which accounts for the losses due to the other components not included in the CFD domain. Due to both manufacturing and geometrical constraints, two different optimized impellers with 3 and 5 blades have been developed, with the performance required in terms of efficiency and suction capability. The predicted performance of both configurations were compared with the measured head and efficiency characteristics.

Design Optimization of a Centrifugal Compressor Impeller Using Radial Basis Neural Network Method

2009 ◽  
Author(s):  
Jin-Hyuk Kim ◽  
Jae-Ho Choi ◽  
Kwang-Yong Kim
Keyword(s):  
Neural Network ◽  
Design Optimization ◽  
Centrifugal Compressor ◽  
Optimization Techniques ◽  
Design Flow ◽  
Radial Basis Neural Network ◽  
Neural Network Method ◽  
Network Method ◽  
Radial Basis ◽  
Isentropic Efficiency

This paper presents a procedure for design optimization of a centrifugal compressor. The centrifugal compressor consists of a centrifugal impeller, vaneless diffuser and volute. And, optimization techniques based on radial basis neural network method are used to optimize the impeller of the centrifugal compressor. Latin hypercube sampling of design of experiments is used to generate thirty design points within design spaces. Three-dimensional Reynolds-averaged Navier-Stokes equations with shear stress transport turbulence model are discretized by using finite volume approximations and solved on hexahedral grids to evaluate the objective function of an isentropic efficiency. Four variables defining impeller hub and shroud contours are selected as design variables in this optimization. The results of optimization show that the isentropic efficiency of the optimized shape at the design flow coefficient is enhanced by 1.0% and the efficiencies at the off-design points are also improved significantly by the design optimization.

scholarly journals Design of artificial neural network based power split controller for hybrid electric vehicle

2018 ◽  
Vol 1 (1) ◽  
pp. 72-80
Author(s):  
Aqsa Kk
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Dynamic Programming ◽  
Electric Vehicle ◽  
Hybrid Electric Vehicle ◽  
Design Flow ◽  
Parallel Hybrid Electric Vehicle ◽  
Parallel Hybrid ◽  
Artificial Neural ◽  
Hybrid Electric

 In this paper the work represent the design flow of artificial neural network (ANN) for the parallel hybrid electric vehicle using the dynamic programming strategy, for the better fuel economy and power for the real time driving condition. In this paper the artificial neural network for the parallel hybrid electric vehicle is first trained from the input/output data generated by a dynamic programming. The power spilt between electric motor (EM) and  internal combustion engine (ICE) an is prescribe by using this artificial neural network controller. One input layer is used and one output layer is used with 2 hidden layers. For the training of the data the numpy-library is used and matlab-simulink is used for the implementation. The trained data is used. The data is tasted on three driving cycle named NEDC, US06 and FTP-75 for both the thermal and hybrid vehicles.

scholarly journals AHEAD: Automatic Holistic Energy-Aware Design Methodology for MLP Neural Network Hardware Generation in Proactive BMI Edge Devices

Energies ◽  
2020 ◽  
Vol 13 (9) ◽  
pp. 2180
Author(s):  
Nan-Sheng Huang ◽  
Yi-Chung Chen ◽  
Jørgen Christian Larsen ◽  
Poramate Manoonpong
Keyword(s):  
Neural Network ◽  
Low Power ◽  
Design Methodology ◽  
Hardware Design ◽  
Design Flow ◽  
Floating Point ◽  
Energy Aware ◽  
Mlp Neural Network ◽  
Neural Recordings ◽  
Neural Network Hardware

The prediction of a high-level cognitive function based on a proactive brain–machine interface (BMI) control edge device is an emerging technology for improving the quality of life for disabled people. However, maintaining the stability of multiunit neural recordings is made difficult by the nonstationary nature of neurons and can affect the overall performance of proactive BMI control. Thus, it requires regular recalibration to retrain a neural network decoder for proactive control. However, retraining may lead to changes in the network parameters, such as the network topology. In terms of the hardware implementation of the neural decoder for real-time and low-power processing, it takes time to modify or redesign the hardware accelerator. Consequently, handling the engineering change of the low-power hardware design requires substantial human resources and time. To address this design challenge, this work proposes AHEAD: an automatic holistic energy-aware design methodology for multilayer perceptron (MLP) neural network hardware generation in proactive BMI edge devices. By taking a holistic analysis of the proactive BMI design flow, the approach makes judicious use of the intelligent bit-width identification (BWID) and configurable hardware generation, which autonomously integrate to generate the low-power hardware decoder. The proposed AHEAD methodology begins with the trained MLP parameters and golden datasets and produces an efficient hardware design in terms of performance, power, and area (PPA) with the least loss of accuracy. The results show that the proposed methodology is up to a 4X faster in performance, 3X lower in terms of power consumption, and achieves a 5X reduction in area resources, with exact accuracy, compared to floating-point and half-floating-point design on a field-programmable gate array (FPGA), which makes it a promising design methodology for proactive BMI edge devices.

Neural Network Aided GPS/INS Integration for UAV Navigation

2013 ◽  
Vol 397-400 ◽  
pp. 1598-1601
Author(s):  
Qing Ji Meng
Keyword(s):  
Neural Network ◽  
Kalman Filter ◽  
Integrated System ◽  
Design Flow ◽  
New Method ◽  
Test Results ◽  
Navigation Solution ◽  
Primary Test ◽  
Uav Navigation

Traditional INS/GPS integrated system would degrade sharply when GPS outages, a new method is proposed in this paper, which is a NN(Neural Network) aided KF (Kalman Filter) integration method.The structure of the method was shown summarily, the NN and KF design flow was discussed briefly, test experimentations have been done to prove the availability of the proposed method. Primary test results have shown that the proposed method can reduce the navigation solution during GPS outages, by about 70%, in comparison with INS stand alone results in GPS outage of 60 seconds.

Autofluorescence spectroscopy of oral leukoplakia using an artificial neural network

2000 ◽  
Vol 25 (4) ◽  
pp. 325-325
Author(s):  
J.L.N. Roodenburg ◽  
H.J. Van Staveren ◽  
N.L.P. Van Veen ◽  
O.C. Speelman ◽  
J.M. Nauta ◽  
...  
Keyword(s):  
Neural Network ◽  
Artificial Neural Network ◽  
Oral Leukoplakia ◽  
Autofluorescence Spectroscopy ◽  
Artificial Neural
Sign in / Sign up

Export Citation Format

Share Document