Magnetic anomaly inversion using magnetic dipole reconstruction based on the pipeline section segmentation method

Abstract

In the model of a horizontal straight pipeline of finite length, the segmentation of the pipeline elements is a significant factor in the accuracy and rapidity of the forward modeling and inversion processes, but the existing pipeline segmentation method is very time-consuming. This paper proposes a section segmentation method to study the characteristics of pipeline magnetic anomalies—and the effect of model parameters on these magnetic anomalies—as a way to enhance computational performance and accelerate the convergence process of the inversion. Forward models using the piece segmentation method and section segmentation method based on magnetic dipole reconstruction (MDR) are established for comparison. The results show that the magnetic anomalies calculated by these two segmentation methods are almost the same regardless of different measuring heights and variations of the inclination and declination of the pipeline. In the optimized inversion procedure the results of the simulation data calculated by these two methods agree with the synthetic data from the original model, and the inversion accuracies of the burial depths of the two methods are approximately equal. The proposed method is more computationally efficient than the piece segmentation method—in other words, the section segmentation method can meet the requirements for precision in the detection of pipelines by magnetic anomalies and reduce the computation time of the whole process.