The Study Of Prediction Method Of The Mining Subsidence Based On Genetic Algorithm

The subsidence caused by underground coal mining do increasingly harm to economic development and ecological environment. Therefore, the analysis and prediction of ground movement and deformation around the mining areas is significant. The probability integration method based on the theory of stochastic mechanics is presently widely used in mining subsidence prediction. The parameter error caused by inaccurate expected parameter calculation is one of the main expected errors. The author tries to use genetic algorithm to calculate the expected parameters of the probability integration method and use the data of any observation line in the observation materials of surface movement to obtain the expected parameters and analyze the accuracy and reliability.Considering the conditions and the exploitation of mining geological conditions, the author of this paper takes the Yun Jialing Mine in Wu'an, Hebei province as an example, firstly a group of expected sinkage parameters is assumed, which is used to expect values of surface subsidence. The predicted data is used as inverse initial data. After that the genetic algorithm should be used to calculate the corresponding expected sinkage parameters, which will be compared with the pre-set parameters. Then man can determine the accuracy and reliability of the inverse probability integration of the genetic algorithm. Based on the genetic algorithm the author using MATLAB programming based on the corresponding surface movement and deformation prediction equations, obtain the expected parameters for the main dip-slip section of surface movement basins during semi-infinite mining, the expected parameters for the main dip-slip section of surface movement basins during limited mining and the expected parameters for the main strike-slip section of surface movement basins during limited mining, and also draws many kinds of changing contours of the coalface, and thus proves that, the use of genetic algorithm can accurately calculate the expected parameter values inversely.