Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Deep neural network-based multi-part reactive solute transport parameter inversion method

A deep neural network and multi-component response technology, applied in neural learning methods, biological neural network models, neural architectures, etc., can solve problems such as parameter identification

Active Publication Date: 2021-04-30
JILIN UNIV
View PDF6 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The technical problem to be solved in the present invention is to provide a multi-component reactive solute transport parameter inversion method based on a deep neural network, which can be used to estimate and set key parameters in the multi-component reactive solute transport model, thereby Solving Parameter Identification Problems During Multicomponent Reactive Solute Simulations

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 more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Deep neural network-based multi-part reactive solute transport parameter inversion method
  • Deep neural network-based multi-part reactive solute transport parameter inversion method
  • Deep neural network-based multi-part reactive solute transport parameter inversion method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0063] This example is a hypothetical experimental model of a reactive solute migration simulation column in porous media, with a length of 0.08 m. The solute in the initial state solution is mainly 1mmol / L NaNO 3 and 0.2mmol / LKNO 3 . One end of the column is injected with 0.6mmol / L CaCl at a flow rate of 0.1m / h 2 solution. The porosity of the porous medium is 0.3, and the permeability is 7.0×10-12m 2 , the density is 2650kg / m 3 . The main chemical reaction in this system is the cation exchange process, and its expressions are mainly as follows:

[0064] Na + +K-X=K + +Na-X;

[0065] Na + +0.5Ca-X 2 =0.5Ca 2+ +Na-X;

[0066] K + and Ca 2+ Take Na + The cation exchange coefficients for reference are: K Na / K = 0.1995 and K Na / Ca =0.3981, the cation exchange capacity is 0.01779meq / 100g.

[0067] The initial concentration and boundary concentration of each ion are shown in Table 1:

[0068] Table 1 The initial concentration and boundary concentration of various...

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

No PUM Login to View More

Abstract

The invention relates to a deep neural network-based multi-part reactive solute transport parameter inversion method, which comprises the following steps of: establishing a multi-part reactive solute transport model according to observation data, and determining a parameter to be inverted and a parameter value range; preparing a training sample set; carrying out data normalization training a forward neural network; training an inversion neural network; estimating model parameters by using an inversion neural network; calculating a forward neural network error at the parameter estimation point; by adaptively updating the forward neural network, the local precision of the forward neural network is improved, and the error of the inversion result is gradually reduced. The multi-part reactive solute transport parameter inversion method can solve the problem of multi-part reactive solute transport parameter inversion, and provides technical support for application of a multi-part reactive solute transport simulation technology.

Description

technical field [0001] The invention relates to a multi-component reactive solute migration parameter inversion method based on a deep neural network, which belongs to the research on the inverse problem of groundwater numerical simulation. Background technique [0002] Multi-component reactive solute migration simulation is an important tool for analyzing porous media flow systems and chemical component migration laws, and has a wide range of research applications in groundwater pollution, carbon dioxide geological storage, and nuclear waste burial safety assessment. [0003] In the research process of using numerical models to solve practical problems, accurate setting of model parameters is the prerequisite for reliable simulation results of numerical models. In the simulation of multicomponent reactive solute transport, many parameters are difficult to obtain directly by current measurement methods. To solve this problem, using actual observation data and inversion calc...

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
no application Login to View More
IPC IPC(8): G06F30/20G06N3/08G06N3/04
CPCG06F30/20G06N3/08G06N3/045
Inventor 戴振学陈骏骏杨志杰
Owner JILIN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com