摘要
提出了巷道支护设计与决策的新方法──基于神经网络的自学习和模式自适应识别的方法。建立的人工神经元网络系统以实际支护方案为基础，从工程实例中学习支护决策的知识，在工程岩体地质特征与支护方案之间建立起智能推理网络，然后将其推广，类比出新开掘巷道的支护方案。推理结果表明，该网络在工程地质特征与支护方案间可以建立良好的推理关系，方法实用价值高，决策结果与实际相吻合。
A novel approach to select support system for underground openings is described. It is based on self-learning and adaptive recognition of neural network. Knowledge is first obtained from case study to recognize support systems for roadways. An intelligent reasoning network is established between geological conditions of rockmass of the projects and support systems. Then it is popularized. The results show that the neural network can establish a good reference relation between the geological characteristics of projects and support systems. This method is of great value. And the results of calculation coincide with the measured data.
出处
《煤炭学报》
EI
CAS
CSCD
北大核心
1995年第1期34-38,共5页
Journal of China Coal Society
基金
国家教委博士点基金
辽宁省博士启动资金
关键词
神经网络
巷道支护
设计
underground roadways,support system , neural network,inference, self-learning, adaptive recognition
