Wide-area electromagnetic method induced polarization information nonlinear extraction method based on learning
An extraction method and electromagnetic method technology, which is applied in the direction based on specific mathematical models, scientific instruments, biological models, etc., and can solve the problems of limiting the application of nonlinear inversion algorithms, large amount of calculation, and high computational cost.
Active Publication Date: 2021-04-23
湖南工商大学
View PDF6 Cites 1 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
However, the nonlinear inversion algorithm evaluates the optimal solution in the global solution space by repeatedly calling the forward algorithm. For the complex forward simulation in the EM inverse problem, there are problems of huge amount of calculation and high calculation cost, which greatly limits the Application of Nonlinear Inversion Algorithm in Extraction of IP Information by Wide-area Electromagnetic Method
Method used
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0105] A wide-area electromagnetic method IP information extraction method based on nonlinear inversion algorithm, such as figure 2 As shown, the algorithm process of this method is:
[0106] (1) Initialize the algorithm parameters and the population of the gray wolf optimization algorithm, each individual in the population represents a set of resistivity parameters and polarizability parameters of the wide-area electromagnetic method;
[0107] (2) Evaluate the individual using the real model (Algorithm 1);
[0108] In step (2), there is no training data of the proxy model at the beginning of the algorithm, so an initial iteration cycle parameter iter is set when the algorithm starts to execute b , in the initial iterative cycle, all individuals use the real model to evaluate the individual in order to obtain enough training data sets; after the initial iterative cycle, only when the elite solutions α, β, δ wolf are updated, the real model is used to evaluate the individual ...
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More PUM
Login to View More Abstract
The invention discloses a wide-area electromagnetic method induced polarization information nonlinear extraction method based on learning. The method comprises the steps that a grey wolf optimization algorithm is adopted as a main program of nonlinear inversion; a Bayesian neural network is used as an agent model with low calculation cost; the inversion process is optimized by learning historical data in the inversion process, an electromagnetic method inversion problem is solved by using a grey wolf optimization algorithm, and induced polarization information of a wide-area electromagnetic method is extracted. According to the method, a nonlinear inversion algorithm based on a learning mechanism is designed, on one hand, a self-adaptive grey wolf optimization algorithm is designed by utilizing historical effective parameters in an algorithm process; and on the other hand, establishing an agent model based on the Bayesian neural network in combination with forward evaluation data in the inversion process. The method provided by the invention has good applicability, can efficiently and quickly complete the inversion task for the complete nonlinear inversion method, saves the inversion time, and provides referential experience for realizing large-scale and high-precision geological data interpretation.
Description
technical field [0001] The invention belongs to the technical field of geophysics, and relates to a learning-based method for nonlinearly extracting IP information using a wide-area electromagnetic method. Background technique [0002] The Electromagnetics Method (EM for short) is a method based on the differences in electrical conductivity, magnetic permeability and dielectric properties of rocks, minerals (stones) in the earth's crust and surrounding media, and uses the principle of electromagnetic induction to observe and study artificial or natural formations. The distribution law of the electromagnetic field and then the geophysical method to solve various geological problems. Electromagnetic (EM) inversion is one of the most difficult problems in geophysics, because on the one hand, the EM inverse problem has nonlinear and ill-posed characteristics; on the other hand, the forward modeling of the electromagnetic method needs to solve Numerical problems are very computa...
Claims
the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More Application Information
Patent Timeline
Login to View More IPC IPC(8): G06F30/27G06F30/25G06N3/00G06N7/00G01V3/00G06F111/06G06F111/08G06F119/02
CPCY02A90/30
Inventor 董莉江沸菠王敏捷周志彬
Owner 湖南工商大学