An Iterative Inversion Method For The Detection Of Irregular Defects In The Leakage Of Magnetic Flux Based On The Evolutionary Algorithm

In our national economy,the oil and natural gas occupies an important position in the national economy,and the transportation of the pipeline is the main mode of transportation.But the leakage of the pipeline can lead to the occurrence of major accidents such as explosion,combustion and casualties.Therefore,the pipeline defect detection is of great significance to the regular safety maintenance of the pipeline and the prediction of pipeline leakage.This thesis studies and analyzes the defect inversion method of the magnetic detection technology.The main work is as follows:Firstly,the background and significance of the thesis are introduced,and the two key factors of the defect inversion technique are the forward model of the magnetic field and the inverse algorithm,which will be introduced and analyzed in detail.The forward model of the leakage magnetic detection is constructed by the finite element analysis software ANSYS,the model size parameter is set,and the correction of the parameters is carried out according to the measured data of the pipeline.Finally the forward model is solved to be prepared for the following inversion problem.Then,the inverse algorithm of pipeline defect is studied,and a genetic algorithm is used as the inverse algorithm for defect inversion based on the forward finite element model.On the basis of the basic genetic algorithm,to solve the problem that the genetic algorithm is easy to fall into the local extreme value and the marginal degrees of freedom are more likely to lose the diversity,the three aspects of the fitness calculation method,the population crossing method and the population variation mode are respectively optimized,The experimental results show that the improved genetic algorithm has faster convergence and higher stability.In the end,we focus on the inverse performance of the particle swarm optimization as the inverse algorithm.On the basis of the basic particle swarm optimization,to solve the problem of precocious puberty,the particle swarm optimization is optimized from the two aspects of the inertia weight and the speed updating method,respectively.For the inertia weight of the particle swarm,the inertial weight adaptive to the current motion state of the particle is introduced.For the speed updating mode of particle swarm,by introducing other particles into the speed updating formula as the learning object of social experience,other high quality particles in the population also have the opportunity to be learned,thus enriching the evolution direction of particle swarm.Next,to solve the problem that the particle swarm optimization algorithm is not easy to jump out of the local extreme value under the requirement of higher convergence precision,an improved method based on the dynamic particle composition mode is proposed,and the idea of the genetic algorithm"inheritance" and the "variation" is used for reference.A particles with a good position is added to help jump out of the local minimum and to accelerate the optimization process.Finally,based on the above improvement,the degree of freedom of the inversion is expanded,and the performance of the improved algorithm after the degree of freedom expansion is compared.