摘要
以矿岩RQD、RMQ和RMR值为基础进行矿岩可崩性聚类综合分级,解决了同一地质测段分级结果的不一致性,明确了可崩性级别;利用样本中易测评判指标建立矿岩可崩性BP神经网络分析模型进行矿岩可崩性分级,解决了用钻孔资料进行矿岩可崩性分级精度不高的问题。程潮铁矿自然崩落法试验区段的矿岩可崩性分级表明矿岩可崩性属于极易一中等崩落,与生产实际相符,在生产中可应用该方法进行矿岩可崩性分级工作。
Based on ore rock RQD,RMD and RMR value to cluster grading of ore rock cavability, solve the same geology examine sections of hierarchical inconsistency of result ,confirm the rank of ore rock cavablity; Utilize sample apt to examine judging quota can set up the rank of ore rock cavablity by using BP neural network analysis ; solve the hierarchical precision problem of cavablity with bore informing. The grading results of ore rock cavability in Chengchao iron mine ore natural caving sysem test ore rock of sector indicate that ore rock cavablity belong to extremely apt-medium-sized caving , The result is in conformity with reality.
出处
《化工矿产地质》
CAS
2004年第2期112-116,共5页
Geology of Chemical Minerals
关键词
自然崩落采矿法
聚类分级
矿岩可崩性
BP神经网络
程潮铁矿
natural caving system, ore rock cavability, cluster grading, BP neural network, application, Chengchao iron mine
