PCA-BP模型在判别基于LIF技术煤矿突水水源的应用（英文）

发布时间：2018-04-13 09:50

本文选题：煤矿突水 + 水源判别　；参考：《光谱学与光谱分析》2017年03期

【摘要】：防治煤矿突水时需迅速精准地判别突水水源,激光诱导荧光(LIF)光谱技术具有灵敏度高、快速准确监测特点,为检测突水水源提供了一种新的方法。该研究引入该技术以获取突水荧光光谱数据。采用卷积(SG)平滑和多元散射校正(MSC)方法对光谱图进行预处理,以消除光谱采集过程中噪声干扰。采用主成分分析(PCA)方法提取特征信息,针对SG预处理后的数据,当主成分个数为3时,累积贡献率可达到99.76%,已基本保留原数据的全信息。选择3层结构BP神经网络建立分类判别模型,通过不同方式构造训练集和测试集,SG预处理数据构建的分类模型可以达到精准判别,而对于MSC预处理和原始数据出现很少的误判。实验结果表明SG预处理结果要优于MSC预处理。研究结果表明,将PCA和BP神经网络结合建立分类模型,能有效判别煤矿突水水源,且具有较强的自组织、自学习能力。
[Abstract]:It is necessary to distinguish water inrush source quickly and accurately when preventing and curing water inrush in coal mine. Laser induced fluorescence spectroscopy (LIF) technique has the characteristics of high sensitivity, fast and accurate monitoring, which provides a new method for detecting water inrush water source.In this study, the technique was introduced to obtain fluorescence spectrum data of water inrush.In order to eliminate the noise interference in spectral acquisition, convolutional SG-based smoothing and multivariate scattering correction (MSC) methods are used to preprocess the spectral images.The principal component analysis (PCA) method is used to extract the feature information. When the number of principal components is 3, the cumulative contribution rate can reach 99.76, and the whole information of the original data has been basically preserved.The three-layer BP neural network is selected to establish the classification and discrimination model. The classification model constructed by constructing training set and test set SG pretreatment data in different ways can achieve accurate discrimination, but there are few misjudgments for MSC preprocessing and raw data.The experimental results show that SG pretreatment is superior to MSC pretreatment.The results show that the classification model based on PCA and BP neural network can effectively distinguish the water inrush from coal mine, and it has strong self-organization and self-learning ability.
【作者单位】：安徽理工大学电气与信息工程学院;阜阳师范学院计算机与信息工程学院;
【基金】：Project supported by the"Twelfth Five-year"National Science and Technology Support Projects(2013BAK06B01) the National Science Foundation Project of China(51174258) the National Safe Production Critical Incident to Key Technologies Science and Technology Project(anhui-0001-2016AQ) the Provincial Natural Science Research Project(KJ2015A278,KJ2015A295) the College Young Talent Support Plan(gxyqzd2016194)
【分类号】：TD745

【相似文献】