Abstract: Billions of tonnes of coal are compressed under oil and gas pipelines in the mining areas of Shaanxi and Mongolia,
China,which seriously restricts the high-intensity mining and sustainable development of regional coal mines. The key to
solve this problem is to accurately predict the surface subsidence caused by underground mining,so as to provide an important
basis for the formulation of scientific mining schemes. However,due to the widespread distribution of thick bedded argillaceous
weakly cemented sandstone with different particle sizes in the regional overburden,the surface subsidence law is quite different
from the conventional understanding,which makes the prediction results of the probability integral method established by the
discontinuous medium model significantly different,especially in the stage of insufficient mining,the surface subsidence is
small. Based on this,this paper proposes a prediction idea of surface subsidence under insufficient mining considering overburden
structure and rock properties. Firstly,according to the fault exploration results of Jining No. 3 Mine and Shilawusu Mine,
the distribution differences of faults in the two mining areas in the east and west are compared and analyzed,and the reasons for
the small subsidence of deep mining in the extremely thick and weakly cemented mining area are discussed. It is found that the
faults in the eastern coal mines are dense,and the faults with a drop value of more than 20 m are common,while the faults in
the western coal mines are basically undeveloped,and the integrity of the extremely thick and weakly cemented overburden is better. Based on this and combined with the field measured data,the Vlasov thick plate theory and the equivalent subsidence
space principle are used to establish a surface subsidence prediction model for the non-full mining of extremely thick and weakly
cemented overburden,and the Shilawusu Mine is taken as an example for application. The results show that the root mean
square error of the traditional probability integral method is 1 125. 3 mm,and it is still 83. 84 mm after correcting the parameters.
The prediction error of the model is reduced to 8. 73 mm,and the prediction error is reduced by 89. 59%,which verifies
the reliability and accuracy of the model. The study results can provide technical support for the coordinated protection of coal
mining and oil and gas pipelines in Shaanxi-Inner Mongolia mining area.