采空区危险性评价方法优化

来源期刊:中南大学学报(自然科学版)2013年第7期

论文作者:冯岩 王新民 程爱宝 张钦礼 赵建文

文章页码:2881 - 2888

关键词:危险性评价;采空区;主成分分析法;BP神经网络;方法优化

Key words:risk evaluation; underground goaf; principal component analysis; BP-neural networks; method optimization

摘    要:为了更合理精确地评价采空区危险性,对利用神经网络评价的方法进行优化,建立主成分分析法与神经网络结合的采空区危险性评价模型。从地质和工程条件出发,综合考虑影响采空区稳定性的13项主要因素,统计样本数据。运用主成分分析法对影响采空区稳定性的样本进行预处理,将分析结果作为神经网络的输入数据,减少输入变量,消除变量之间的相关性,从而加快数据处理速度,提高预测精度。将该方法应用于锡矿山的采空区危险性评价。结果表明,预测误差在8%以内,较未经主成分分析的神经网络预测精度有了很大提高。利用主成分分析法和神经网络结合建立的采空区危险性评价模型具有分析速度快、预测精度高等优点。该方法科学合理,为采空区危险性评价提供了一种更完善的评价体系。

Abstract: The method of neural networks was optimized, and the principal component analysis and neural networks were used to construct a model about underground goaf risk evaluation to evaluate the underground goaf risk more accurately and reasonably. On the basis of geological and engineering conditions, the 13 main factors that affecting the stability of goaf were considered to count the sample data. The principal component analysis was conducted to the samples, taking the results as the input data for neural network. By reducing the number of input variables, and eliminating the correlation between variables, the data processing was sped up, and the prediction accuracy was improved. The method was applied to evaluate the risk underground goafs in Xi mine. The results show that the prediction error is controlled within 8%, and the accuracy is improved greatly compared with that predicted by neural network without principal component analysis. Assessing the risk of goaf with principal component analysis and neural network has the advantage of rapid analysis and prediction accuracy. The method is scientific and reasonable, and a better evaluation system for the underground risk evaluation is provided.

有色金属在线官网  |   会议  |   在线投稿  |   购买纸书  |   科技图书馆

中南大学出版社 技术支持 版权声明   电话:0731-88830515 88830516   传真:0731-88710482   Email:administrator@cnnmol.com

互联网出版许可证:(署)网出证(京)字第342号   京ICP备17050991号-6      京公网安备11010802042557号