Open Access. Powered by Scholars. Published by Universities.®
Articles 1 - 3 of 3
Full-Text Articles in Engineering
Online Identification Of Turbogenerator's Dynamics Using A Neuro-Identifier, Wenxin Liu, Ganesh K. Venayagamoorthy, Donald C. Wunsch
Online Identification Of Turbogenerator's Dynamics Using A Neuro-Identifier, Wenxin Liu, Ganesh K. Venayagamoorthy, Donald C. Wunsch
Electrical and Computer Engineering Faculty Research & Creative Works
The increasing complexity of modern power systems highlights the need for effective system identification techniques for the successful control of power system. This paper proposes a robust continually online trained neuroidentifier to predict the outputs of turbogenerator - terminal voltage and speed deviation. The inputs to the neuro-identifier are the changes of the plant's outputs and plant's inputs. It overcomes the drawback of calculating deviation signals from reference signals for different operating points in previous work. Simulation results show that the neuro-identifier can provide accurate identification under different operating conditions. Furthermore, the neuro-identifier can learn the dynamics of the system …
Fdtd Data Extrapolation Using Multilayer Perceptron (Mlp), H. Goksu, David Pommerenke, Donald C. Wunsch
Fdtd Data Extrapolation Using Multilayer Perceptron (Mlp), H. Goksu, David Pommerenke, Donald C. Wunsch
Electrical and Computer Engineering Faculty Research & Creative Works
This work compares MLP with the matrix pencil method, a linear eigenanalysis-based extrapolator, in terms of their effectiveness in finite difference time domain (FDTD) data extrapolation. Matrix pencil method considers the signal as superposed complex exponentials while MLP considers each time step to be a nonlinear function of previous time steps.
Neural Networks Skin Tumor Diagnostic System, Zhao Zhang, William V. Stoecker, Randy Hays Moss
Neural Networks Skin Tumor Diagnostic System, Zhao Zhang, William V. Stoecker, Randy Hays Moss
Electrical and Computer Engineering Faculty Research & Creative Works
In this study, a malignant melanoma diagnostic system is designed using a straightforward neural network with the back-propagation learning algorithm. Eleven features are automatically extracted from skin tumor images. The correct diagnostic rate of this system is better than the average rate of 16 dermatologists who based their diagnosis with only the slide images.