Improvement of On-line Recognition Systems Using a RBF-Neural Network Based Writer Adaptation Module

2011 ◽  
Author(s):  
Lobna Haddad ◽  
Tarek M. Hamdani ◽  
Monji Kherallah ◽  
Adel M. Alimi
Keyword(s):  
Neural Network ◽  
Rbf Neural Network ◽  
On Line ◽  
Recognition Systems

On-Line Thickness Measurement of Thin Film Based on Neural Network

2014 ◽  
Vol 484-485 ◽  
pp. 307-310
Author(s):  
Li Cai ◽  
Yue Gang Tan ◽  
Qin Wei
Keyword(s):  
Neural Network ◽  
Thin Film ◽  
Film Thickness ◽  
Production Cost ◽  
Rbf Neural Network ◽  
Thickness Measurement ◽  
Monitoring Method ◽  
Line Thickness ◽  
On Line ◽  
Thickness Deviation

This paper proposes a on-line thickness measurement scheme of thin film based on the capacitance thickness sensor and introduces the composition and principle of thickness measurement system. Then it further states the principle and simulation of the RBF neural network, which can effectively predict the thickness deviation of thin film by setting the appropriate parameters. The monitoring method based on the RBF neural network will reduce production cost and make the film thickness uniformity better, combining the traditional film production line with an new idea of controlling the opening degree of wind ring.

Temperature Control of a SOFC and MGT Hybrid System

2011 ◽  
Vol 8 (5) ◽  
Author(s):  
Xiao-Juan Wu ◽  
Qi Huang ◽  
Xin-Jian Zhu ◽  
Chang-Hua Zhang
Keyword(s):  
Neural Network ◽  
Hybrid System ◽  
Temperature Control ◽  
Rbf Neural Network ◽  
Learning Ability ◽  
Inlet Temperature ◽  
Micro Gas Turbine ◽  
Proportional Integral ◽  
Internal Reforming ◽  
On Line

Transients in a load have a significant impact on the performance and durability of a solid oxide fuel cell (SOFC) integrated into a micro gas turbine (MGT) hybrid power system. One of the main reasons is that the SOFC operating temperature and turbine inlet temperature change drastically due to the load change. Therefore, in order to guarantee the temperature to operate within a specified range, an adaptive proportional-integral-derivative (PID) decoupling control strategy based on a dynamic radial basis function (RBF) neural network is presented to control the temperature of a natural gas fueled, tubular SOFC/MGT hybrid with internal reforming in this paper. Using the self-learning ability of the dynamic RBF neural network, the proportional, integral, and differential factor of the PID controller are tuned on-line. The simulation results show that it is feasible to build the adaptive PID decoupling controller for temperature control of the SOFC/MGT hybrid system.

Prediction of Flow Stress of Stainless Steel 0Cr13Mn by RBF Neural Network

2010 ◽  
Vol 139-141 ◽  
pp. 264-268
Author(s):  
Bai Lin Fan ◽  
Ling Qi Meng ◽  
Zhong Fu Li
Keyword(s):  
Neural Network ◽  
Stainless Steel ◽  
Flow Stress ◽  
Strain Rate ◽  
Numerical Models ◽  
Rbf Neural Network ◽  
Rbf Network ◽  
On Line ◽  
Temperature Strain ◽  
Developing Method

The flow stress of hot deformation about 0Cr13Mn stainless steel was experimentally studied by A Gleeble1500 thermo-mechanical simulator. The effects of deformed temperature, strain rate and strain to flow stress were analyzed. The prediction of RBF neural network with correlation between the flow stress and the chemical composition, deformed temperature, strain rate and strain, etc was established. Simulation data of the flow stress by RBF network with relationship between input and output for 0Cr13Mn stainless steel were stable. Accuracy of the prediction by RBF Neural network was higher than the regression precision by the multiple non-linear regression numerical models. Through a combination of the prediction by RBF neural network and numerical regression model of flow stress, the developing method of BPF neural network on-line calculation based on measured data of flow stress will be feasible under the condition of ensuring the accuracy of the premise.

New Method of Diagnosis and New Knowledge Acquirement for Fault Diagnosis Expert System of High Voltage Circuit Breaker

2012 ◽  
Vol 466-467 ◽  
pp. 1413-1417
Author(s):  
Yong Mei Yang ◽  
Nai Quan Sun ◽  
Hong Ke Xu
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Expert System ◽  
Rbf Neural Network ◽  
Circuit Breaker ◽  
Forward Reasoning ◽  
The Neural Network ◽  
New Knowledge ◽  
On Line ◽  
Improved Algorithm

In this paper, an improved algorithm of general radial basis (RBF) function neural network is introduced, based on improved algorithm, the neural network realized quickly fault diagnosis and self-update of neural network structure, and the neural network is applied to the on-line fault diagnosis expert system. The expert system deals with the fault data that send from on-line monitoring equipment by using neural network, and it can discover the fault type and give reasonable solution by forward reasoning. Meanwhile, the expert system has the ability of achieving new knowledge based on the application of self-update ability of RBF neural network.

Sign in / Sign up

Export Citation Format

Share Document