Research On The Inversion Of Outcrop Of Coal Seam Fire Exploration Based On Infrared Imaging Technology

Coal spontaneous combustion is an outstanding natural disasters of coal mine, not onlyseriously affect the normal production of coal, results in substantial waste coal resources, butalso poses a threat to staffs’life and property. Therefore, it is important to find the fire sourcelocation at the beginning of coal burning.There are many external factors which cause the nature burning of coalfield, includingtopography, climate, geological structure, human mining activities and so on. At present,thereare five methods of detecting the location and extent of the coa l fire,including geophysical,geochemical, thermal exploration, drilling and remote sensing. It plagues the researchers tofind an effective detection methods based on spontaneous combustion character and formationmechanism.In this paper, a kind of GASA(Hybrid optimization inversion algorithm) is proposedwhich uses infrared thermal imaging method taking into account the infrared radiation basedon the physical characteristics of the outcrop of coal seam, heat transfer theory and thecharacteristics of spontaneous combustion and meanwhile, loose coal experiment is set up toverify the effectiveness of the algorithm. The algorithm combines the advantages of geneticalgorithm and simulated annealing algorithm to achieve the optimization of structure, behaviorand operation, and it also uses inversion method for the first time to inverse the temperatureand location of fire source. The GASA algorithm creates diffusion equation of point heatsource as the forward model to be the objective function, which combines the information ofsurface temperature field and inverses the location and intensity of fire source effectively.Point heat diffusion equation is given by dimensional analysis and heat transfer theory and it is a suitable model for arbitrary point heat source spreads to the surrounding by any medium.The GASA algorithm avoids the error of model order error caused by finite difference anddiscrete regularization, getting a certain degree of precision in time and inversion effectcompared to traditional method. In this paper, we conduct a detailed comparison and analysisof GA, SA, and GASA algorithm’s inversion results, providing a powerful source of practicalvalue and reference value for detecting fire source.