Estimating the relationship between isoseismal area and earthquake magnitude by a hybrid fuzzy-neural-network method

1999 ◽  
Vol 107 (2) ◽  
pp. 131-146 ◽  
Author(s):  
Chongfu Huang ◽  
Yee Leung
Keyword(s):  
Neural Network ◽  
Fuzzy Neural Network ◽  
Earthquake Magnitude ◽  
Neural Network Method ◽  
Network Method ◽  
Fuzzy Neural ◽  
The Relationship

Performance Prediction of Pre-Corroded Aluminum Alloy Using Genetic Algorithm-Neural Network and Fuzzy Neural Network

2008 ◽  
Vol 33-37 ◽  
pp. 1283-1288 ◽  
Author(s):  
Chao Hua Fan ◽  
Yu Ting He ◽  
Hong Peng Li ◽  
Feng Li
Keyword(s):  
Neural Network ◽  
Genetic Algorithm ◽  
Aluminum Alloy ◽  
Aluminum Alloys ◽  
Performance Prediction ◽  
Bp Neural Network ◽  
Fuzzy Neural Network ◽  
Neural Network Method ◽  
Network Method ◽  
Fuzzy Neural

Genetic algorithm is introduced in the study of network authority values of BP neural network, and a GA-NN algorithm is established. Based on this genetic algorithm-neural network method, a predictive model for fatigue performances of the pre-corroded aluminum alloys under a varied corrosion environmental spectrum was developed by means of training from the testing dada. At the same time, a fuzzy-neural network method is established for the same purpose. The results indicate that genetic algorithm-neural network and fuzzy-neural network can both be employed to predict the underlying fatigue performances of the pre-corroded aluminum alloy precisely.

Transient probabilistic design of flexible multibody system using a dynamic fuzzy neural network method with distributed collaborative strategy

2018 ◽  
Vol 233 (11) ◽  
pp. 4077-4090 ◽  
Author(s):  
Lu-Kai Song ◽  
Guang-Chen Bai ◽  
Cheng-Wei Fei ◽  
Rhea P Liem
Keyword(s):  
Neural Network ◽  
Probabilistic Analysis ◽  
Multibody Systems ◽  
Fuzzy Neural Network ◽  
Robot Manipulator ◽  
Neural Network Method ◽  
Surface Method ◽  
Network Method ◽  
Flexible Multibody ◽  
Fuzzy Neural

To improve the efficiency and accuracy of transient probabilistic analysis of flexible multibody systems, a dynamic fuzzy neural network method-based distributed collaborative strategy is proposed by integrating extremum response surface method and fuzzy neural network. Distributed collaborative dynamic fuzzy neural network method is mathematically modeled and derived by considering the high nonlinearity, strong coupling, and multicomponent characteristics of a flexible multibody system. The proposed method is demonstrated to perform the transient probabilistic analysis of a two-link flexible robot manipulator. We obtain the distributional characteristics, reliability degree, and sensitivity degree of robot manipulator, which are useful for the effective design of robot manipulator. By comparing the full-scale method, extremum response surface method, dynamic fuzzy neural network method, and distributed collaborative dynamic fuzzy neural network method, we find that the distributed collaborative dynamic fuzzy neural network method can be used to perform the transient probabilistic analysis of the robot manipulator and improve the computational efficiency while maintaining a good accuracy. Moreover, this study offers a useful insight for the reliability-based design optimization of flexible multibody systems, and enriches the field of mechanical reliability theory as well.

Relationship Between Distance of Photonic Crystal Rods Array and Wavelength of Light Propagating in Right-Angular, Cross, and Y-Shapes Waveguide

2008 ◽  
Author(s):  
Meng-Ju Lin
Keyword(s):  
Neural Network ◽  
Photonic Crystal ◽  
Optical Communication ◽  
Complex Geometry ◽  
Numerical Data ◽  
Neural Network Method ◽  
Bifurcation Phenomena ◽  
Network Method ◽  
Physical Phenomena ◽  
The Relationship

Photonic crystal is widely used in optical communication. Waveguide is one of important components used in optical communication. The performance of right-angular, cross, and Y shapes photonic crystal waveguide used in optical communication is investigated. Material used for photonic crystal is GaAs. Shape of photonic crystal rod is cylindrical rod. These rods are arranged to become matrix with constant distance. In the photonic crystal matrix, some rods are removed to form waveguide channel. Three shapes of waveguides are discussed. They are right-angular, cross, and Y shapes. The relationship between distances of circular photonic crystals and wavelength in propagation is discussed. Due to coupling of multi physical phenomena and complex geometry, a finite element method software called COMOSL is used to analyze performance of waveguide under different wavelength of light. It is found under different distance of photonic crystal rods array and wavelength, light could propagate along waveguide successfully or lost energy and decay. From results analyzed by COSMOL software, numerical data are obtained to classify light being able to pass waveguide successfully or not. By neural network method, equations between crystal rods array distance and wavelength to decide whether light can propagate through waveguide successfully or not are derived. They can be used to determine performance of light propagating in waveguide. For Y shape waveguide, bifurcation phenomena of light propagate in different channel of Y shape branch is found. It is found under different crystal pillar distance the light of different wavelength would propagate in one of the branch of the Y shape channel. The longer wavelength lights pass the upper channel and the shorter pass the lower. The results show that the bifurcation happens at the tendency if pillar distance increases the bifurcation wavelength also increases. From neural network method, equations to decide light propagate to upper or lower branch of Y shape waveguides are also obtained.

The Product Types Choosing Based on Product Service Characteristics Table

2013 ◽  
Vol 721 ◽  
pp. 631-635
Author(s):  
Hu Hao ◽  
Yu Jun Lu ◽  
Ju Yin Qin ◽  
Ling Lin Liang ◽  
Lu Lin
Keyword(s):  
Neural Network ◽  
Working Conditions ◽  
Fuzzy Neural Network ◽  
Research Approach ◽  
Design Requirement ◽  
Service Characteristics ◽  
Product Service ◽  
Fuzzy Neural ◽  
The Relationship

In order to define the relationship between various product working conditions and product types choosing design, product types choosing method according to product service characteristics as design requirement is put forward. First product service characteristics table was defined，secondly product service characteristics vector as fuzzy neural network's input, product service records were employed to train fuzzy neural network, then the result of product types choosing was gained through synthesizing output of fuzzy neural network based on the power of each product service characteristics vector, finally the example of railway carriage bearing was adopt to explain above research approach.

Sign in / Sign up

Export Citation Format

Share Document