Research On Transparency System Of Working Face Based On Multi-Source Data Collaboration And Intelligent Algorithms

As the basis for the realization of intelligent and precise mining in coal mines,transparency of the working face has become the future development trend of the industry.In view of the uncertainty of internal geological information in the area to be mined and the asymmetry of multi-source data generated by people,equipment and environment in the process of coal face transparency construction at present,the means of field investigation,theoretical analysis,case test,numerical simulation and engineering practice were employed to studying the reconstruction model of geological anomaly in coal face and cooperative fusion mechanism of multi-source data around the working face systematically.Firstly,based on the widely used technology of electromagnetic wave penetration,,the matrix solution problem in the traditional inversion calculation process was converted into a functional extremum solution problem in this paper aiming at the shortcomings of traditional inversion algorithms.The objective function model of geological anomalies tomography inversion for electromagnetic wave based on intelligent algorithm was established,which providing a new idea for the inversion of geological anomalies in working face.Secondly,the evolutionary mechanism of the single genetic algorithm(SGA)with strong global search ability in the process of solving the objective function was studied,and the crossover and mutation operators are revealed to be the key factors affecting the computational performance of SGA.A multi-group co-evolution operator and genetic parameters adaptive adjustment operator were designed to optimize SGA,and a geological anomaly reconstruction model based on adaptive multiple population genetic algorithm(AMPGA)was established.Through case testing and numerical simulation,it was found that the intelligent algorithm of AMPGA effectively enhances the global and local search capabilities of the SGA,ensures the diversity of the population,and improves the accuracy,stability,and noise resistance of the inversion results.Thirdly,three kinds of constraint models based on prior data in electromagnetic wave tomography inversion were proposed,including range constraint,proximity constraint and mean constraint.The influence of multi-source prior data constraint on the inversion result of AMPGA was analyzed.Through numerical simulation,it was found that the accuracy of the inversion results corresponding to the three constraint models gradually increases with the increase of the priori constraint data exposed in the roadway,drilling and mining.Among them,the average constraint model can better use the constraints data to effectively adjust the evolution direction of the AMPGA in the inversion calculation process,showing higher accuracy and stability.Then,based on a large number of on-site investigations,the characteristics,classification,and transparency of multi-source data generated by people,equipment,and environment related to the production of the working face were analyzed.This paper presents a collaborative sharing mechanism and platform architecture of coal mine multi-source data,which uses graphical engineering data as the spatial guidance and integrates multi-disciplinary information.It not only provides numerical support for the inversion and prediction of geological anomalies in the working face,but also effectively improves the collaborative management and transparency of multi-source data in coal mines,and provides conditions for the full utilization of multi-source big data in coal mines.Finally,based on the multi-source data collaboration platform of coal mine,the intelligent algorithm of AMPGA based on the average value constraint realizes efficient and accurate prediction of internal geological anomalies in the 8208 working face of the Datong mining area under the constraints of the prior data exposed by the roadway,drilling and mining.The mining design scheme was also discussed according to the inversion results.In addition,based on the feedback of on-site application effects of multiple coal mining enterprises,it is revealed that the multi-source data collaboration platform helps coal mining enterprises achieve greater scope and higher dimensions of transparency,and provides data support for intelligent and precise mining.There are 101 figures,28 tables,and 267 references in this dissertation.