The Fault Diagnosis for Electro-Hydraulic Servo Valve Based on the Improved Genetic Neural Network Algorithm

2006 ◽  
Author(s):  
Lian-dong Fu ◽  
Kui-sheng Chen ◽  
Jun-sheng Yu ◽  
Liang-cai Zeng
Keyword(s):  
Neural Network ◽  
Fault Diagnosis ◽  
Servo Valve ◽  
Network Algorithm ◽  
Hydraulic Servo ◽  
Neural Network Algorithm ◽  
Genetic Neural Network

Nuclear Magnetic Image Segmentation of Rectal Cancer Based on Improved Genetic Neural Network

2020 ◽  
Vol 10 (7) ◽  
pp. 1644-1653
Author(s):  
Danyang Li ◽  
Yumei Sun ◽  
Wanqing Liu ◽  
Bing Hu ◽  
Jianlin Wu ◽  
...  
Keyword(s):  
Neural Network ◽  
Rectal Cancer ◽  
Image Segmentation ◽  
Fitness Function ◽  
Segmentation Method ◽  
Magnetic Image ◽  
Network Algorithm ◽  
Neural Network Algorithm ◽  
Genetic Neural Network ◽  
Nuclear Magnetic

Image segmentation is the basis of image analysis and understanding, and has an unshakable position in the field of computer vision. In order to improve the accuracy of nuclear magnetic image segmentation of rectal cancer, this paper proposes an improved genetic neural network algorithm for the problems of traditional BP neural network algorithm. In order to enhance the network performance, this paper improves the genetic neural network from the two aspects of fitness function and genetic operator, which makes the training speed and convergence precision greatly improved. Target samples were analyzed by image histogram analysis, and the improved genetic neural network was used to learn the samples to obtain the training network. Taking the pixel matrix of the image as the input vector, it is put into the trained network for classification, and finally the segmentation is realized. The simulation experiment proves that compared with the classical image segmentation method, the improved genetic neural network image segmentation method has a good segmentation effect and is a feasible image segmentation method.

Application of Materials Design Based on Genetic Neural Network

2007 ◽  
Vol 280-283 ◽  
pp. 1837-1840
Author(s):  
Yilai Zhang ◽  
Xue Jian Li ◽  
Ling Ke Zeng ◽  
Cheng Kang Chang
Keyword(s):  
Neural Network ◽  
Materials Design ◽  
Aluminum Titanate ◽  
Network Algorithm ◽  
Neural Network Algorithm ◽  
Genetic Neural Network ◽  
Design Materials

Neural network (NN) is an effective method in the filed of materials design, but the convergent speed is decided by initial weights. This paper proposes genetic neural network algorithm (GNNA) to design materials. Aluminum titanate modification is studied by the method of GNNA. The results indicate the algorithm works well.

Sign in / Sign up

Export Citation Format

Share Document