您的位置:首页  > 论文页面

层次分析法和神经网络模型在矿井突水中的应用

发表时间:2009-02-28  浏览量:1755  下载量:852
全部作者: 宋晓猛,孔凡哲
作者单位: 中国矿业大学资源与地球科学学院
摘 要: 综合应用层次分析法和神经网络预测模型进行矿井突水预测及相关分析,运用相关性分析确定各影响因素之间及其与矿井突水状态的相关程度,结果表明:岩溶、水压和含水层厚度与突水状态成正相关,而隔水层厚度与突水状态成负相关。通过多目标的层次分析决策模型确定了各因素对突水量影响程度的排列顺序为水压、含水层厚度、岩溶、断裂构造、隔水层厚度。最后利用基于误差反向传播的神经网络预测3个样本的突水状态为:样本1不突水,样本2和样本3突水,与实际情况相吻合,可知运用神经网络进行矿井突水预测是可靠的。
关 键 词: 水文学及水资源;矿井突水;层次分析法;人工神经网络;相关性分析
Title: Application of analytical hierarchy process and neural network model in coal mining water inrush
Author: SONG Xiaomeng, KONG Fanzhe
Organization: School of Resource and Earth Science, China University of Mining and Technology
Abstract: The paper integrates correlation analysis, analytical hierarchy process (AHP) and artificial neural network (ANN) to do some analysis on forecast for coal mining water inrush. First, the paper determines the relativities between the impact factors and the state of water inrush by correlation analysis, and the results show that the water inrush is positive correlated with the karst, water pressure and the thickness of aquifer respectively, and negative correlated with the thickness of aquifuge. Then it determines the order of various factors’ impact degree by AHP, which is water pressure, the thickness of aquifer, karst, fault structure and the thickness of aquifuge. Finally, the paper does the forecast for the state of water inrush of three samples, and the result shows that the water inrush may not arise in sample 1, but may arise in sample 2 and 3, which tallies with the actual situation considerably.
Key words: hydrology and water resources; coal mining water inrush; analytical hierarchy process; artificial neural network; correlation analysis
发表期数: 2009年2月第4期
引用格式: 宋晓猛,孔凡哲. 层次分析法和神经网络模型在矿井突水中的应用[J]. 中国科技论文在线精品论文,2009,2(4):345-351.
 
0 评论数 0
暂无评论
友情链接