|
|
|
| Optimization of bolting scheme based on combination of principal component analysis and BP neural network |
| WU Shu-liang, CHEN Jian-hong, YANG Shan |
| School of Resources and Safety Engineering, Central South University, Changsha 410083, China |
|
|
|
Abstract Roadway characteristics and bolting can be seen as a non-linear mapping, using the general mathematical methods are difficult to represent non-linear mapping between roadway bolting scheme and its influencing factors. Neural network has been widely used in optimization of bolting scheme and achieved good results. Based on the fact that there existing deficiencies when only using neural networks to predict bolting scheme, the optimization model of bolting scheme was set up combining principal component analysis with neural networks.Principal component analysis was used to do neural network input data pre-processing, which could reduce the input data and make input data unrelated, accelerate the convergence rate of the network,and prediction accuracy was over 90%.The results show that combining principal components analysis with BP neural networks to optimize bolting scheme has a high prediction accuracy. Compared with the single BP neural network,it can improve the precision accuracy．
|
|
Published: 15 April 2012
|
基于主成分分析与BP网络的锚杆支护方案优选
巷道特征与锚杆支护之间可以看作是一种非线性映射关系,用一般的数学方法难以表达巷道支护方案与其影响因素之间的非线性映射关系.神经网络已广泛应用于锚杆支护方案优选,并取得较好的效果.基于单一神经网络预测锚杆支护方案存在一些不足,构建了主成分分析与BP网络相结合的巷道锚杆支护方案优选模型.利用主成分分析对神经网络的输入数据进行预处理,使输入数据减少且不相关,加快网络的收敛速度,并且预测精度均在90%以上.研究结果表明:将主成分分析与BP神经网络结合优选巷道的锚杆支护方案,具有很高的预测精度;与单一BP神经网络相比,提高了预测精度．
关键词：
锚杆支护方案,
主成分分析,
BP神经网络,
综放回采巷道
|
Viewed |
|
|
|
Full text
|
|
|
|
|
Abstract
|
|
|
|
|
Cited |
|
|
|
|
| |
Shared |
|
|
|
|
| |
Discussed |
|
|
|
|
