Title :
A Novel Algorithm on BP-Neural Network
Author_Institution :
Sch. of Electron. & Inf. Eng., Jinggangshan Univ., Jian, China
Abstract :
Because during the net work learning, BP fall into local minimum, this article proposed a prove algorithm that change the learning gradient, and adjust the weight and threshold, in order to speed up the network´s learning speed. At last, this method was used to, study sample at the fuzzy reasoning layer by BP calculation method, obtain professional experience and ways of thinking about bridage safety assessment via the network. Finally, compare the assessment results from the network with those from professionals. The comparison proves the artificial BP-neural network´s more feasibility and efficiency.
Keywords :
backpropagation; bridges (structures); fuzzy reasoning; neural nets; safety; structural engineering computing; BP calculation method; BP neural network; backpropagation; bridge safety assessment; fuzzy reasoning; learning gradient; network learning; Bridges; Communication cables; Educational institutions; Power cables; Safety; Structural beams; Vectors; BP Model; BP algorithm; BP neural network;
Conference_Titel :
Computer Science and Electronics Engineering (ICCSEE), 2012 International Conference on
Conference_Location :
Hangzhou
Print_ISBN :
978-1-4673-0689-8
DOI :
10.1109/ICCSEE.2012.117