Inversion Of Self-potential Data From Contaminant Monitoring Based On PSO-PF Algorithm

Self-potential(SP)method has the advantage of sensitivity and economic efficiency in the application of locating and tracking underground organic contaminants.However,the composition and mechanism of SP of organic contaminants are complex,and the electric field signal is easily disturbed,which increases the difficulty of inversion and interpretation of observed data.Aiming at this problem,a SP data inversion algorithm based on improved particle filter is proposed based on improved particle swarm optimization technology.Firstly,on the basis of analyzing the characteristics of diffusion and migration of organic contaminants in underground media and the macroscopic response of REDOX reaction,the complex SP source is equivalent to a simple geometrically polarized body to carry out forward modeling.Then,the iterative optimization of particle swarm optimization is used to replace the importance sampling process of particle filtering,and the problems such as insufficient coincidence between likelihood function and prior distribution and particle weight degradation in the late iteration are improved.A global optimization inversion algorithm for SP data is designed based on improved particle filter.The simulated data test with random noise shows that the inversion algorithm has good robustness,fast convergence speed and good tolerance to noise.At the same time,the particle statistics in the inversion process can also provide the posterior probability density distribution of the inversion parameters,which can be further used for reliability analysis of the inversion results.In order to further verify the performance of the inversion algorithm,the measured data from laboratory physical simulation are tested.The results show that the algorithm is suitable for the inversion and interpretation of complex SP data such as organic pollution with microbial participation,and can provide effective algorithm support for data processing and interpretation of actual contaminated sites.