Title :
An Original RBF Network Based on Attribute Similarity
Author :
Changbiao, Li ; Jianping, Song ; Kewen, Xia ; Lei, Wang
Author_Institution :
Sch. of Electron. & Inf. Eng., Xi´´an Jiaotong Univ.
Abstract :
According to the thought of biological receptive field, a model of original radial basis function neural network with attribute similarity (ASRBF) is presented, it not only has evident physical and biologic meanings, but also can simplify topology structure, decrease operation and save cost. Eventually, forecast oil reservoir production with the trained network, the simulation shows that, the application effect of this improved RBF network algorithm is very good; and the algorithm is not only superior to that of tradition RBF network, but also with high fitting precision and quick rate of convergence
Keywords :
learning (artificial intelligence); radial basis function networks; rough set theory; attribute similarity; biological receptive field; forecast oil reservoir production; original radial basis function neural network; Biological system modeling; Cost function; Discrete event simulation; Fitting; Hydrocarbon reservoirs; Network topology; Petroleum; Predictive models; Production; Radial basis function networks; RBF network; attribute reduction; attribute significance; attribute similarity; reservoir production;
Conference_Titel :
Mechatronics and Automation, Proceedings of the 2006 IEEE International Conference on
Conference_Location :
Luoyang, Henan
Print_ISBN :
1-4244-0465-7
Electronic_ISBN :
1-4244-0466-5
DOI :
10.1109/ICMA.2006.257412
