Inversion Of Permeability Coefficient Of Goaf Based On Genetic Algorithm

The permeability coefficient of goaf is the key physical parameter affecting the air leakage and the air movement in the goaf.The air leakage in the goaf is the main cause of the spontaneous combustion disaster of the remaining coal in the goaf,and also affects the gas concentration distribution in the goaf..It can be seen that the determination of the permeability coefficient of the goaf is of great significance for the disaster prevention and control of the goaf.The traditional permeability coefficient measurement method often solves the permeability coefficient of the actual seepage field through field experiments combined with relevant analytical formulas.The analytical formula is more applicable under relatively simple boundary conditions,but due to the complicated boundary conditions of the goaf,this method is used to calculate The permeability coefficient is more difficult.Therefore,based on the numerical inversion idea,the measurable point pressure value inside the goaf is selected as the characteristic parameter to invert the permeability coefficient,and the error between the pressure inversion value and the pressure target value at the measuring point is the inversion objective function.The inversion model of the permeability coefficient of the goaf is established,and the inversion problem of the permeability coefficient is transformed into the minimum error optimization problem.The improved adaptive genetic algorithm is used to optimize the error minimum of the inversion target function of the goaf.Firstly,the genetic algorithm is used to encode the permeability coefficient in floating point number to generate the initial population of permeability coefficient.Then the individual individuals are brought into one by one.The finite volume method is used to calculate the pressure value,and the error between the set of corresponding measured point pressure inversion values and the set of pressure target values is calculated.According to the minimum value optimization principle,the adaptive genetic algorithm is used for global search to obtain the numerical model.The optimal numerical distribution of the permeability coefficient.This paper establishes a numerical model to verify the feasibility of the technical route and the accuracy of the inversion results in the inversion algorithm of the goaf in the goaf based on genetic algorithm,which lays a theoretical foundation for the future physical experiment and practical engineering application.In this paper,three different permeability coefficient distribution models are established for the inversion test.By analyzing the inversion results,we can see that the relative error between the pressure inversion results and the initial set values of the three numerical models is kept at 10% Below,and the corresponding relative error between permeability is also within the acceptable range;the comparison test of the influence degree of the number of measuring points on the inversion results shows that the number of measuring points plays a positive correlation on the accuracy of the inversion results of the permeability coefficient.That is,the larger the number of measuring points,the larger the coverage area,and the more accurate the inversion result of the obtained permeability coefficient is.The comparison experiments using the adaptive genetic algorithm and the simple genetic algorithm respectively show that the crossover and mutation operators are improved.The adaptive genetic algorithm is superior to the simple genetic algorithm in the accuracy of the inversion results and the inversion efficiency.In summary,the inversion method of the goaf in the goaf based on the adaptive genetic algorithm has a good inversion capability in the numerical model of the two-dimensional goaf.This method provides a numerical algorithm for the determination of the permeability coefficient of the goaf.The new ideas have guiding significance for the practical application of the project.The paper has 27 pictures,15 tables,and 55 references.