摘要
应用人工神经网络系统理论，采用机器学习的方法，建立了岩石的可爆性指数与岩体的爆炸漏斗体积Ｖ、大块率Ｋ＿１、平均合格率Ｋ＿２、小块率Ｋ＿３和波阻抗Ｚ之间的非线性映射关系，并将其用神经网络、网络连接权值矩阵和节点阈值向量分布式表达出来。对于新的岩石，网络采用并行推理的方法预报出其可爆性。实践表明，神经网络方法科学、具有较强的非线性动态处理的能力。
With the application of artificial neural network theory and machine learn-ing method,this paper establishes a nonlinear mapping between the blastability of rockand its affected factors such as volume of explosion crater,mass ratio of big rockblocks,ratio of small rock blocks,qualified ratio of blasting and wave impedence andrepresents them distributedly on neural network, connection weights and threshold ofnodes.The blastability of new type of rock mass is predicted by means of the method ofparallel inference.The results show that the proposed method has some more importantad vantag es than traditional ones and it has strong ability for nonlinear dynam ic processing.
出处
《爆炸与冲击》
EI
CAS
CSCD
北大核心
1994年第4期298-306,共9页
Explosion and Shock Waves
基金
国家八.五重点攻关项目
关键词
可爆性
神经网络
非线性映射
爆破
岩石力学
rock mass, blastability,neural netwotk,nonlinear mapping,self-learn-ing,parallel inference.
