ES Estimation Based on RBF Neural Networks and ROLS Algorithm
The paper proposed an ES (Expert System) lifetime estimation method based on RBFNN and ROLS algorithm. It is considered that the determination of center parameters will directly influence the learning capability of RBFNN,so the ROLS algorithm and the FCM algorithm for training RBFNN are adopted. The algorithm improves the original OLS algorithm and it doesnt need to decide the number of RBF centers in advance and the effective centers of network can be obtained by adopting backward selection algorithm,which reduces the number of the centers and computational efforts and it can accelerate the training of the RBFNN substantially. The satisfactory results are achieved by using this algorithm in the expert system of concrete safety estimation. It illustrates that the ROLS algorithm is effective and adaptive. Finally,the estimation data,which are calculated by training RBF and BP respectively,are showed and it is proved that RBF and its learning algorithm are better than that of BP.
ES Estimation RBF Neural Network ROLs Algorithm
Chengqian Ma Jingling Yuan Luo Zhong Qiong Jiang
Department of Computer Science and Technology,Wuhan University of Technology,430070,China
国际会议
The First International Conference on Management Innovation(ICMI 2007)(管理创新会议)
上海
英文
444-446
2007-06-04（万方平台首次上网日期，不代表论文的发表时间）