Magnetotelluric inversion method based on full convolutional neural network
A convolutional neural network and convolutional neural technology, applied in the field of magnetotelluric inversion based on full convolutional neural network, can solve the problems of loss of position information, overfitting, slow network convergence, etc., and achieve high fitting accuracy , fast convergence speed, and the effect of reducing the loss of computing memory and time
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Example Embodiment
[0076] Example 1: Two-dimensional abtenometry
[0077] Establish high, low-resistance constant model, and use a rectangular mesh to pass the cross-line interpolation finite unit method to calculate the input data of the simulation test model.
[0078] image 3 It is a test model sample and a grid score for the embodiment of the present invention. image 3 Distance image 3 (A) represents low-resistance abnormalities, (b) represents high resistance; image 3 Medium abandoned area ( image 3 The peripheral area of (a), image 3 The middle rectangular area of (b)) represents a high resistance, and the resistivity range is between 1000-1500 Ω · m, the light region ( image 3 The middle rectangular area of (a), image 3 The peripheral area of (b) represents a low-resistance, and the resistivity range is from 100 to 600 Ω · m. image 3 The middle rectangular area of (a) and (b) indicates an abnormally, and other areas are uniform surrounding rocks. The lateral and longitudinal dimension...
Example Embodiment
[0086] Example 2: Measured data inversion
[0087] This example selects a line in the electromagnetic field detection application test for inversion testing, which is 12km long, measurement point distance of 500m, and removes partial interference measures, a total of 20 test points. The MT measurement in the work area uses the self-developing equipment IEM-I electromagnetic method, the receiver is a DRU-1C type, and the magnetic sensor is an IMC-03 type. The system operating frequency range is from 0.0001 to 10 kHz, and the acquisition time is designed according to the minimum frequency of the work area, which is more than 8 hours. Intercepting the observation data of the 320-0.088Hz band, the frequency is distributed, with a total of 48 frequency points.
[0088] Figure 7 It is a schematic view of the measured inversion model of the second embodiment of the present invention. Figure 7 As shown, according to known geological materials, Figure 7 The sample set shown, using a recta...
PUM
Property | Measurement | Unit |
---|---|---|
Resistivity value | aaaaa | aaaaa |
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap