Research on Prediction Model of Gas Emission
Based on Lasso Penalty Regression Algorithm
Qian Chen
1(&) and Lianbing Huang
2
1 China Coal Research Institute, Beijing 100013, China
365039211@qq.com
2 Institute of Manned Space System Engineering, Beijing 100094, China
Abstract. Researches show that the amount of mine gas emission is influenced
by many factors, including the buried depth of coal seams, coal thickness, gas
content, CH 4 concentration, daily output, coal seam distance, permeability,
volatile yield, air volume, etc. Its high-dimensional characteristics could easily
lead to dimension disaster. In order to eliminate the collinearity of attributes and
avoid the over-fitting of functions, Lasso algorithm is used to reduce the
dimension of variables. After low-redundancy feature subset is obtained, the
best performance model is selected by 10-fold cross-validation method. Finally,
the gas emission is predicted and analyzed based on public data from coal mine.
The results show that the prediction model based on Lasso has higher accuracy
and better generalization performance than principal component analysis prediction model,and the accurate prediction of gas emission can be realized more
effectively.
Keywords: Gas Á The amount of emission Á Feature selection Á Penalty
regression Á Prediction model
1 Introduction
In recent years, with the number of safety accidents and death toll have decreased year
by year, China’s coal mine safety situation has improved, but the security situation is
still grim, since coal mine accidents occur frequently and major accidents happen from
time to time. According to incomplete statistics, there are 1945 coal mine accidents
from 2013 to 2017, and the death toll was 3771. Among them, the number of gas
accidents and deaths accounted for 11.21% and 30.17%, respectively, becoming the
second largest safety accident after the roof disaster accident. Therefore, gas is still one
of the main factors causing coal mine safety accidents.
Many scholars have done a large number of researches on the prediction model of
gas emission for coal mine. Qi and Xia [1] used the different-source prediction method
to predict the mine gas emission. Li [2] used the grey system theory to study the mine
gas emission. Using the l 1 regularized outlier isolation and regression method (LOIRE)
with the TLBO optimization algorithm, Hu et al. [3] established the TLBO-LOIRE
optimization prediction model to calculate and analyze the relevant influencing factors
and predicted the gas emission from the coal mining face.
© Springer Nature Singapore Pte Ltd. 2020
Q. Liang et al. (Eds.): Artificial Intelligence in China, LNEE 572, pp. 165–172, 2020.
https://doi.org/10.1007/978-981-15-0187-6_19
Based on Lasso Penalty Regression Algorithm
Qian Chen
1(&) and Lianbing Huang
2
1 China Coal Research Institute, Beijing 100013, China
365039211@qq.com
2 Institute of Manned Space System Engineering, Beijing 100094, China
Abstract. Researches show that the amount of mine gas emission is influenced
by many factors, including the buried depth of coal seams, coal thickness, gas
content, CH 4 concentration, daily output, coal seam distance, permeability,
volatile yield, air volume, etc. Its high-dimensional characteristics could easily
lead to dimension disaster. In order to eliminate the collinearity of attributes and
avoid the over-fitting of functions, Lasso algorithm is used to reduce the
dimension of variables. After low-redundancy feature subset is obtained, the
best performance model is selected by 10-fold cross-validation method. Finally,
the gas emission is predicted and analyzed based on public data from coal mine.
The results show that the prediction model based on Lasso has higher accuracy
and better generalization performance than principal component analysis prediction model,and the accurate prediction of gas emission can be realized more
effectively.
Keywords: Gas Á The amount of emission Á Feature selection Á Penalty
regression Á Prediction model
1 Introduction
In recent years, with the number of safety accidents and death toll have decreased year
by year, China’s coal mine safety situation has improved, but the security situation is
still grim, since coal mine accidents occur frequently and major accidents happen from
time to time. According to incomplete statistics, there are 1945 coal mine accidents
from 2013 to 2017, and the death toll was 3771. Among them, the number of gas
accidents and deaths accounted for 11.21% and 30.17%, respectively, becoming the
second largest safety accident after the roof disaster accident. Therefore, gas is still one
of the main factors causing coal mine safety accidents.
Many scholars have done a large number of researches on the prediction model of
gas emission for coal mine. Qi and Xia [1] used the different-source prediction method
to predict the mine gas emission. Li [2] used the grey system theory to study the mine
gas emission. Using the l 1 regularized outlier isolation and regression method (LOIRE)
with the TLBO optimization algorithm, Hu et al. [3] established the TLBO-LOIRE
optimization prediction model to calculate and analyze the relevant influencing factors
and predicted the gas emission from the coal mining face.
© Springer Nature Singapore Pte Ltd. 2020
Q. Liang et al. (Eds.): Artificial Intelligence in China, LNEE 572, pp. 165–172, 2020.
https://doi.org/10.1007/978-981-15-0187-6_19