An accelerated particle swarm optimization training algorithm for RBF neural network in river stage forecasting in typical Karst area, South China

2013 ◽  
pp. 351-356
Keyword(s):  
Neural Network ◽  
Particle Swarm Optimization ◽  
South China ◽  
Rbf Neural Network ◽  
Particle Swarm ◽  
Karst Area ◽  
Training Algorithm ◽  
Swarm Optimization ◽  
River Stage ◽  
Accelerated Particle Swarm Optimization

New Neural Network Inverse Control of Two-Motor Drive System

2013 ◽  
Vol 416-417 ◽  
pp. 447-453
Author(s):  
Mei Kang ◽  
Wen Xiang Zhao ◽  
Jing Hua Ji ◽  
Guo Hai Liu
Keyword(s):  
Neural Network ◽  
Particle Swarm Optimization ◽  
Motor System ◽  
Rbf Neural Network ◽  
Particle Swarm ◽  
Drive System ◽  
Motor Drive ◽  
Swarm Optimization ◽  
Synchronous System ◽  
Motor Drive System

Two-motor drive system is a multi-variable, nonlinear and strongly coupled system. A new synchronous control strategy for two-motor system is proposed based on radial basis function (RBF) neural network inverse with particle swarm optimization. To enhance the system performance, the particle swarm optimization is adopted to optimize the RBF nerve center, an optimized RBF neural network inverse and a two-motor system is connected in series to form composite pseudo-linear system. This two-motor synchronous system can be decoupled into two independent linear subsystems for speed and tension. Then, the decoupled control is implemented by designing a linear closed-loop adjustor. The experimental results verify that the two-motor synchronous system can be decoupled well for speed and tension based on the proposed neural network inverse system. Also, the proposed system can deal with external disturbances with strong robustness.

Algorithm Researching of RBF Neural Network Based on Improved PSO

2011 ◽  
Vol 179-180 ◽  
pp. 233-238 ◽  
Author(s):  
Hua Chen ◽  
Yi Ren Fan ◽  
Shao Gui Deng
Keyword(s):  
Neural Network ◽  
Particle Swarm Optimization ◽  
Rbf Neural Network ◽  
Particle Swarm ◽  
Parameter Vector ◽  
Swarm Optimization ◽  
Improved Particle Swarm Optimization ◽  
Whole Space ◽  
Best Parameter ◽  
Improved Pso

In view of the defect of particle swarm optimization which easily gets into partial extremum, the paper put out an improved particle swarm optimization, and applies the algorithm to the selecting of parameter of RBF neural network basal function. It searches the best parameter vector in the whole space, according to coding means, iterative formula, adapted function which the paper puts forwards. The experiment proves that RBF neural network based on improved PSO has faster convergent speed, and higher error precision.

Sign in / Sign up

Export Citation Format

Share Document