65 / 2023-03-31 14:57:05

Research on Prediction Parameter Estimation of Probability Integral Method Based on Unconventional Observation Station Data

probability integration method; prediction parameters; unconventional observation stations; mining subsidence; parameter inversion

Predicting the surface movement caused by underground mining is of great significance to the theoretical research and production practice of mining subsidence. At present, the most reliable method for obtaining prediction parameters is based on field measured data, and selecting an appropriate prediction model to fit the measured data to obtain prediction parameters. The conventional method for obtaining prediction parameters about the probability integration method has strict requirements for the layout of surface movement observation stations in mining areas. Only when the surface movement observation stations are arranged on the major section, the results of prediction parameters are reliable. However, due to the influence of surface topography and landform, the layout of surface movement observation stations in mining areas often cannot be arranged on the major section, and needs to be arranged in an unconventional way. For unconventional stations, it is difficult to estimate the prediction parameters about the probability integration method, and the accuracy of the estimated parameters is low. For this reason, on the basis of considering the influence of the inflection point offset on the coordinates calculated by the probability integration method, the inversion model of the three-dimensional predicted parameters of the probability integration method is constructed, and when the parameters are obtained, the traditional method to obtain the parameters is used, and adopted a cross-iteration method based on an improved genetic algorithm, and considered the correlation between the predicted parameters of the probability integral method, combined with the measured data to add constraints to the genetic algorithm, making the inversion model more reasonable, Finally, the full parameter inversion is carried out for the predicted parameters of the probability integration method. In order to verify the effectiveness of the method, this study carried out the full parameter inversion of the probability integration method based on the measured data (subsidence and horizontal movement) of an unconventional surface movement observation station in a mining area, combined with the method proposed in this paper. Through comparison, it can be obtained that the mean square error of predicting subsidence is 112mm, and the relative error of predicting subsidence is 5.64%; the mean square error of predicting horizontal movement is 80mm, and the relative error of predicting horizontal movement is 12%; the relative error of predicting subsidence is less than 10% , the relative error of predicting horizontal movement is less than 20% , both can meet the project's estimated accuracy requirements.