Title :
An evolutionary RBF networks based on RPCL and its application in fault diagnosis
Author :
Zhao, Zeng-shun ; Hou, Zeng-Guang ; De Xu ; Tan, Min
Author_Institution :
Coll. of Inf. & Electr. Eng., Shandong Univ. of Sci. & Technol., Qingdao, China
Abstract :
The performance of a RBF neural network strongly depends on the network structure and parameters. Therefore, an algorithm that can automatically select the network configuration will be very beneficial. This paper presents a novel method where configuration of an RBF network can be learned by a hybrid training schema. Appropriate number of neural nets in the hidden layer and their coarse centers is obtained by a modified version of Rival Penalized Competitive Learning (RPCL). Then, Genetic Algorithm is applied to determine the optimal parameters including refined cluster centers, variance of Radial-Basis Function and weights. The application of proposed evolutionary RPCL-RBF networks is discussed and the result of simulation illustrate that proposed method is effective.
Keywords :
fault diagnosis; genetic algorithms; learning (artificial intelligence); radial basis function networks; RBF neural network; evolutionary RBF network; evolutionary RPCL-RBF network; fault diagnosis; genetic algorithm; hybrid training schema; network configuration; network parameters; network structure; optimal parameter; radial-basis function; rival penalized competitive learning; Cybernetics; Fault diagnosis; Genetic algorithms; Intelligent networks; Learning systems; Least squares methods; Machine learning; Neural networks; Radial basis function networks; Testing; Competitive Learning; Fault Diagnosis; Genetic Algorithm; RBF network;
Conference_Titel :
Machine Learning and Cybernetics, 2009 International Conference on
Conference_Location :
Baoding
Print_ISBN :
978-1-4244-3702-3
Electronic_ISBN :
978-1-4244-3703-0
DOI :
10.1109/ICMLC.2009.5212432
