传统的基于机器学习的煤层突水预测方法需要大量的训练样本进行预测模型的训练。而在矿井突水问题中,大量训练样本的获得基本上是不可能的。主要研究在突水样本数据有限的情况下提高煤层突水预测结果的准确性。结合山西省某煤矿的实际情...传统的基于机器学习的煤层突水预测方法需要大量的训练样本进行预测模型的训练。而在矿井突水问题中,大量训练样本的获得基本上是不可能的。主要研究在突水样本数据有限的情况下提高煤层突水预测结果的准确性。结合山西省某煤矿的实际情况,提出了一种新颖的基于万有引力的煤层底板突水预测方法(Gravitational force based algorithm,GFA)。该算法采用半监督的学习方式,将万有引力公式引入到预测模型中,利用少量的突水训练样本作为引力的源点吸引测试样本进行突水安全状态的传递,进而实现突水测试样本安全性的预测。将提出的算法用于历史突水数据以及实际的煤层底板突水数据进行实验,实验结果表明,在突水训练数据有限的情况下,提出的基于万有引力的煤层底板突水预测算法可获得良好的预测效果。展开更多
We consider an Error-in-Variable partially linear model where the covariates of linear part are measured with error which follows a normal distribution with a known covariance matrix. We propose a corrected-loss estim...We consider an Error-in-Variable partially linear model where the covariates of linear part are measured with error which follows a normal distribution with a known covariance matrix. We propose a corrected-loss estimation of the covariate effect. The proposed estimator is asymptotically normal. Simulation studies are presented to show that the proposed method performs well with finite samples, and the proposed method is applied to a real data set.展开更多
文摘传统的基于机器学习的煤层突水预测方法需要大量的训练样本进行预测模型的训练。而在矿井突水问题中,大量训练样本的获得基本上是不可能的。主要研究在突水样本数据有限的情况下提高煤层突水预测结果的准确性。结合山西省某煤矿的实际情况,提出了一种新颖的基于万有引力的煤层底板突水预测方法(Gravitational force based algorithm,GFA)。该算法采用半监督的学习方式,将万有引力公式引入到预测模型中,利用少量的突水训练样本作为引力的源点吸引测试样本进行突水安全状态的传递,进而实现突水测试样本安全性的预测。将提出的算法用于历史突水数据以及实际的煤层底板突水数据进行实验,实验结果表明,在突水训练数据有限的情况下,提出的基于万有引力的煤层底板突水预测算法可获得良好的预测效果。
基金supported by National Natural Science Foundation of China(Grant Nos.10901020 and 11371062)the Fundamental Research Funds for the Central Universities,Beijing Center for Mathematics and Information Interdisciplinary Sciences,China Zhongdian Project(Grant No.11131002)
文摘We consider an Error-in-Variable partially linear model where the covariates of linear part are measured with error which follows a normal distribution with a known covariance matrix. We propose a corrected-loss estimation of the covariate effect. The proposed estimator is asymptotically normal. Simulation studies are presented to show that the proposed method performs well with finite samples, and the proposed method is applied to a real data set.