摘要
为了帮助煤矿寻找井田内开采地质条件较好的区块,将研究区划分为若干个单元,分别统计各单元断裂信息维及其相关因素数据.通过对已揭露区统计数据进行灰色关联分析和逐步回归分析,发现断裂信息维能综合反映断层条数、断层密度、断层长度、断层强度等特征,且与断裂构造的相对复杂程度呈正相关关系.将已揭露区各单元断裂信息维及其影响因素(不含断层类指标)统计数据作为样本进行人工神经网络训练,达到精度要求后,用于预测井田不同单元的断裂信息维.对比东庞井田已揭露区各单元断裂信息维的统计值与预测值,准确率达90%左右.图2,表1,参9.
In order to help coalmine search better regions of mining geological conditions in the mining field,an area studied was divided into cells,and the data,including fault information dimensions and its affection factors,of each cell in worked sections were separately counted.The results of gray relationship analysis and progressive regression analysis of the data showed that the fault information dimension was a synthetical reflection of fault characteristics such as number,density,length,intensity,and there was a positive correlation between fault information dimension and relative complexity degree of mine structures.An artificial nerve network was trained with the pattern of fault information dimensions and its affection factors except fault indexes from every cell in worked sections,and the network meet precision expectations was used to forecast fault information dimension of different cells in mining field. It is validated that the veracity is about 90% by the contrast between statistics and forecast data of the fault information dimension from every cell in worked sections of Dongpang Coal Mine.2figs.,1tab.,9refs.
出处
《湖南科技大学学报(自然科学版)》
CAS
北大核心
2005年第2期1-4,共4页
Journal of Hunan University of Science And Technology:Natural Science Edition
关键词
断裂信息雏
矿井构造
相对复杂程度
预测
fault information dimension
mine structures
relative complexity degree
forecast