Research On The Method Of Parameter Estimation And Model Updating Based On Probability Integral Method And Program Realization

Though the growth of coal production has been slowing down in the past two years,it is undeniable that coal takes decisive position in China.The surface movement after exploiting coal affects the safety of mine production,therefore it is necessary to predict surface movement and deformation,especially the surface of mined area.The accuracy of model parameters determines the precision of the results.Till now,the defect of fast convergent edge hasn't been thoroughly solved.This article selects the engineering examples in Wugou mine 1013 and nine other working face in Wanbei mine area.This paper introduces the four methods of least squares based on probability integral,linear least square method,modular vector method,genetic algorithm and comprehensive parameter negotiation.By using the nine working face in Wanbei mining area as an example,the empirical formula is finally derived.Results show that the expected accuracy of the legacy algorithm model is higher than that of the other three models by using the four methods listed above,while the common problem is that edge converging is too fast.In this paper,a new correction model is proposed.Based on the genetic algorithm,the correction model and the classical model are used for the 1013 working face of Wugou Mine,and the results are compared and analyzed.Using the MATLAB software development platform,this paper calculates the surface movement parameters separately for the four classical models and the correction model.Fast convergence edge problems have been basically solved in this way.And the revised genetic algorithm model has better applicability in calculating the parameters precisely.This is of guiding significance for other surface movement prediction work which has similar geological and mining conditions.