基于混沌优化-神经网络的单桩竖向极限承载力预测.pdfVIP

- * 刘勇健 张伯友 ( 510640) : BP , , - , , , , : BP PREDICTION METHOD OF VERTICAL ULTIMATE BEARING CAPACITY OF PILE BASED ON CHAOS OPTIMAL METHOD AND NEURAL NETWORKS Liu Yongjian Zhang Boyou ( Guangzhou Institute of Geo hemistry , Chinese A ademy of S ien es Guangzhou 510640) Abstract : The hara teristi s of BP neural network and haos optimal method are analyzed. By integrating haos optimal method with gradient-de line method, an optimal method of ombination sear h is reated. Combining a large amount of pilesc testing data of the Pearl River Delta, the predi tion model of verti al ultimate bearing apa ity of single pile is established, based on haos optimal method and BP algorithm. Real example resear h shows the predi ted data agree well with the measured ones. It shows that the method is feasible, effe tive and satisfied with engineering demand, whi h an be used widely in proje ts. Ke words : single pile verti al ultimate bearing apa ity haos optimal method BP neural network predi tion 0 , BP , , , , , BP ( ) ( Chao Optimize, CO) , , ( Artifi ial Neural Network, ANN) , , COBP , [ 1] , 1 BP ,