Optimization Of Hydrogeological Parameters Based On Intelligent Algorithm

In the unsteady flow test,the curve fitting and the linear graphic method for solving hydrogeological parameters are two widely used methods in engineering practice with trict mathematical derivation and proof process,and are fast and convenient.However,both rely too much on the experience of the curve fitting staff.In the curve fitting,a slight change in position will result in a change in the result,not accurate enough.Aiming at this problem,this paper proposes to use the intelligent algorithm global optimization curve fitting and MATLAB fitting function to fit the observation points,thus achieving accurate parameter solving.Because the calculation of different kinds of well functions is involved in the curve fitting calculation,the MATLAB software used in this paper as the operating environment,its powerful numerical calculation and symbolic computing function solves the problem of numerical solution of well function.On the other hand,based on the basic principle of curve fitting,namely the least squares fitting theory,under the groundwater unsteady flow,Newman model and diffusion coefficient theory,the optimal equation for curve fitting is constructed.Using intelligent algorithms,find the best position and standard curve that fits the original data within the entire standard curve or standard curve family,and then accurately solve the hydrogeological parameters.Through comparative analysis:(1)The basic particle swarm optimization algorithm can accurately and quickly fit the curve fitting calculation of the original data in a single standard curve,but for high-dimensional optimization problems,such as the curve fitting of the curve family under the Newman model,its global optimization ability Not strong,it is easy to cause the algorithm to fall into local optimum,the original data is not fitted to the optimal standard curve,the movement is not optimal,and the final solution accuracy is affected.In view of this,for the intelligent optimization of the standard curve family,it is recommended to adopt an improved particle swarm optimization algorithm,such as a random weight particle swarm,whose powerful global optimization ability ensures the accuracy of the fitting curve and the accuracy of the final calculation result.(2)In the intelligent optimization curve fitting of diffusion coefficient,the optimization performance of the standard particle swarm optimization algorithm is sufficient to ensure the accuracy and stability of the results compared with the ant colony algorithm and the differential evolution algorithm.(3)In the intelligent algorithm used in this paper,the biggest influence on the optimization performance of the algorithm is the group size,the maximum number of iterations and the inertia weight.Therefore,in order to ensure that the algorithm can be optimized to the global best,it needs to be set according to the complexity of the specific equation.(4)Using the fitting function in MATLAB to artificially fit the allowable value of the linear slope change,the process of maximizing the retention of the original data can be achieved.