Wyniki wyszukiwania - BazTech

1

An effective estimate for selecting the regularization parameter in the 3D inversion of magnetotelluric data

Ghaedrahmati Reza, Moradzadeh Ali, Moradpouri Farzad

Acta Geophysica

|

2022

|

Vol. 70, no 2

609--621

EN

The magnetotelluric (MT) inverse problem is a nonlinear and strongly ill-posed problem. Therefore, to avoid the problem of non-uniqueness of response, this problem is mainly solved by Tikhonov regularization method. The purpose of this study is to present a suitable method for selecting the regularization parameters in the 3D MT inverse problem, with regard to the accuracy and speed of the inversion. In this research, the regularization parameter is simply estimated in each iteration of inversion as the ratio of the data misfit to sum of the data misfit and model norm in the pre-iteration. This scheme is applied in the well-known 3D inversion algorithm, WSInv3DMT, instead of the discrepancy principle method. The accuracy of this scheme is assessed by performing the inversion on synthetic models and real data. Results from the inversion for the synthetic and real data indicate that the data misfit and the model norm are reduced with an acceptable rate during the inversion operation. The inverse model has been smoothly converged to an appropriate model and that unrealistic structures have not been included in the model. The results also show that estimation of the regularization parameter by the discrepancy principle method and continuing the inversion to achieve the target data misfit may lead to the production of a model with non-realistic structures, while in the proposed scheme the inversion has not encountered this problem and it converges to an appropriate model after fewer iterations of inversion. In addition, the results show that the time consumed for the inversion of a set of real data with 41 stations and 16 measurement frequencies would decrease up to 27 percent compared to the time devoted for inverting the same set of data by the discrepancy principle method. Also the inversion does not deviate toward unrealistic models and it closely converges to the model of real geological structures.

2

3D crosswell electromagnetic inversion based on radial basis function neural network

Fang Sinan, Zhang Zhansong, Chen Wei, Pan Heping, Peng Jun

Acta Geophysica

|

2020

|

Vol. 68, no. 3

711--721

EN

Crosswell electromagnetic (EM) method has fundamentally improved the horizontal detection ability of well logging and will become an increasingly promising approach for the secondary exploration of hydrocarbon reservoir. We applied orthogonal least squares (OLS) radial basis function neural network (RBFNN) based on improved Gram–Schmidt (G–S) procedure to three-dimensional (3D) crosswell EM inversion problems. In the inversion process of the simplifed crosswell model with single-grid conductivity anomalies and normal oil reservoir, compared the inversion results of other fve neural networks, OLS-RBFNN was proved to have the best global optimization ability and the fastest sample learning speed and the average inversion error of low conductivity anomalies model (4%) and oil reservoir model (9%) can meet the inversion requirements of crosswell EM method. Only the OLS-RBFNN could achieve ideal inversion results in the most concerned central area of crosswell model, and the inversion accuracy of this algorithm will be more outstanding when the model becomes more complex. Merely using the three-component time-domain crosswell EM data of two wells, the inversion of 3D medium conductivity in the crosswell dominant exploration area can be efectively realized through the nonlinear approximation of the OLS-RBFNN.

3

The electrical resistivity signature of a fault controlling gold mineralization and the implications for Mesozoic mineralization: a case study from the Jiaojia Fault, eastern China

Zhang K., Lü Q., Yan J., Hu H, Fu G.

Acta Geophysica

|

2017

|

Vol. 65, no. 4

727--741

EN

We use 3D audio magnetotelluric method to the south segment of Jiaojia fault belt, and obtain the 3D electrical model of this area. Regional geophysical data were combined in an analysis of strata and major structural distribution in the study area, and included the southern segment of the Jiaojia fault zone transformed into two fault assemblages. Together with the previous studies of the ore-controlling action of the Jiaojia fault belt and deposit characteristics, the two faults are considered to be favorable metallogenic provinces, because some important features coupled with them, such as the subordinate fault intersection zone and several fault assemblages in one fault zone. It was also suggested the control action of later fault with reversed downthrows to the ore distribution. These studies have enabled us to predict the presence of two likely target regions of mineralization, and are prospecting breakthrough in the southern section of Jiaojia in the Shandong Peninsula, China.