Title :
Flaw Identification Based on Layered Multi-subnet Neural Networks
Author :
Jianye, Liu ; Yongchun, Liang ; Jianpeng, Bian ; Xiaoyun, Sun ; Aihua, Li
Author_Institution :
Sch. of Electr. Technol. & Inf. Sci., Hebei Univ. of Sci. & Technol., Shijiazhuang, China
Abstract :
Pointed to the disadvantages such as low recognizing precision, long training time and limited recognizing range of single neural network in eddy current testing, layered multi-subnet neural network is presented. It is composed by a sumnet and several layered subnets, and can divide a complex task into a series of subtasks, so it could quickly identify whether the defect is existed, and also the defect location and dimension. Because of Fisher ratio method used to select the RBF centers, the network structure is simplified much. The result shows that layered multi-subnet neural network is suitable to online eddy current testing.
Keywords :
eddy current testing; neural nets; Fisher ratio method; eddy current testing; flaw identification; layered multisubnet neural network; sumnet; Eddy current testing; Educational institutions; Electronic mail; Information science; Intelligent networks; Intelligent systems; Mechanical engineering; Neural networks; Pollution; Sun; Fisher Ratio method; RBF; layered multi-subnet neural network;
Conference_Titel :
Intelligent Networks and Intelligent Systems, 2009. ICINIS '09. Second International Conference on
Conference_Location :
Tianjin
Print_ISBN :
978-1-4244-5557-7
Electronic_ISBN :
978-0-7695-3852-5
DOI :
10.1109/ICINIS.2009.39