Research On Quality Control Of Control Survey In Modern Large-scale Mine

With the continuous improvement of China’s economic and scientific strength,the safety production of mines has been paid more and more attention by the state,and mine control measurement as the basis of mine safety has gradually started to receive the attention of safety supervisors.However,the continuous development of science and technology and measuring instruments makes the traditional mine control measurement quality control technology and certain accuracy indicators on the "Coal Mine Measurement Regulations"(hereinafter referred to as "Coal Regulations")promulgated in 1989 in China have serious lagging problems,far from meeting the development needs of today’s mines.Therefore,how to improve the quality control of mine control measurement is an urgent problem to be solved.To solve this problem,this paper firstly discusses the importance of mine safety and the status of mine control measurement in mine construction.Then from three aspects of the mine control measurement work(correction of the limit difference of the index of the Coal Regulation,determination of the position of the gyro-directed edge in the error prediction and the processing of the control measurement data),the paper presents its views on the quality control of the mine control measurement.The work and conclusions of the paper are as follows.(1)In terms of the correction of the triangular elevation limit difference of underground coal regulations,the thesis firstly deduces the source of the original triangular elevation limit difference of underground coal mines,and then analyzes the error propagation theorem and the actual measurement experience to obtain that the original triangular elevation limit difference of underground coal mines is on the large side,which is no longer applicable to the present-day coal mine measurement requirements.Next,a new limit difference is proposed to meet the different inclination environments of the coal mine by classifying and discussing the specific measured data of a coal mine,and the results show that the new limits can meet the measurement requirements under different roadway inclinations.(2)For the determination of gyro-directed edge position,the U-Net neural network in deep learning is used to predict the gyro-directed edge position.This paper first introduces the U-Net neural network model,then describes the environment configuration for model training,and further gives the formula for the evaluation index,and finally analyzes and evaluates the results of the whole experiment.The qualitative(comparison between the original data image,the predicted position image and the effect image produced by superimposing the predicted position image onto the original image)and quantitative(evaluation index Accuracy value of 0.9678,MIOU value of 0.9567,MPA value of 0.9795 and fwIoU value of 0.9998)analysis of the experimental results in the paper shows that the U-Net model achieves better results in determining the gyro-directed edge positions.And it is concluded by a specific example that the gyro-directed edge predicted by U-Net neural network can meet the through error requirement when applied to the through error prediction.That is,deep learning is feasible in predicting the gyro-directed edge position,which also provides a reference value for the research of gyro-directed edge position determination method.(3)In the data processing of mine control survey,the paper uses Python language to compile the data processing involved in mine control survey,including the data processing of mine control survey underground conductor and elevation measurement,the calculation of single point and DXF graphics file conversion in different coordinate systems,the calculation of meridian convergence angle and gyro orientation edge,the calculation of wellhead center coordinates,the calculation of Gaussian projection The practical procedures such as the conversion calculation of coordinates.Then the measurement data processing procedures compiled in the paper are verified by specific measured data.The compiled data processing software is consistent with the calculation results of traditional Excel tables,and the pages are easy to operate,which is of practical value for mine control measurement data processing.