Check patentability & draft patents in minutes with Patsnap Eureka AI!

Deep learning-based slope permanent displacement prediction model training method

A technology of deep learning and training methods, applied in neural learning methods, biological neural network models, instruments, etc., can solve problems such as poor prediction accuracy of permanent displacement prediction models, and achieve accurate, high-precision, and predictive effects of earthquake landslide risk assessment Good results

Active Publication Date: 2021-10-12
SOUTHWEST JIAOTONG UNIV
View PDF3 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The present invention intends to provide a training method for slope permanent displacement prediction model based on deep learning to solve the problem of poor prediction accuracy of the permanent displacement prediction model in the prior art

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 learning-based slope permanent displacement prediction model training method
  • Deep learning-based slope permanent displacement prediction model training method
  • Deep learning-based slope permanent displacement prediction model training method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0033] The present invention will be described in further detail in conjunction with the accompanying drawings and embodiments:

[0034] like figure 1 Shown a permanent displacement of slope prediction model training method based on the depth of learning, comprising:

[0035] Step 1: obtaining seismic data, vibration data base to build the database;

[0036] Specifically, access to the world's existing seismic database, including the NGA-west2 database, Kik-net database, collecting seismic data, which excludes low quality, unreliable, incomplete records, aftershocks and non-free-field ground motion records, follows standard shallow crust selected seismic record: moment magnitude M w ≧ 4; faults from R rup ≤300km; peak ground acceleration PGA≥0.001g. Using Matlab program, using all of the selected seismic records were compiled unity, and build the database.

[0037] Step 2: base motion parameters of seismic data, the vibration intensity is calculated parameters, and stored into the...

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 belongs to the technical field of slope displacement prediction, and discloses a deep learning-based slope permanent displacement prediction model training method, and the method comprises the steps of obtaining seismic oscillation data, and constructing a database according to the seismic oscillation data; according to seismic characteristic parameters in the seismic data, obtaining seismic oscillation intensity parameters through calculation, and storing the seismic oscillation intensity parameters in corresponding positions in a database; according to the soil parameters, the side slope geometric parameters and the environmental parameters of the site corresponding to the ground vibration data, calculating to obtain the side slope critical acceleration, and storing the side slope critical acceleration to a corresponding position in a database; according to the acceleration time history of the earthquake data, obtaining the permanent displacement of the side slope under the earthquake action of the Newmark method through calculation, and storing the permanent displacement of the side slope in a corresponding position in a database; and constructing a deep learning model, training the deep learning model through the database to obtain a parameter-adjusted deep learning model, and taking the parameter-adjusted deep learning model as a slope permanent displacement prediction model. The invention is high in prediction accuracy of the permanent displacement of the side slope, and is beneficial to judging the potential risk of landslide of the side slope caused by an earthquake.

Description

Technical field [0001] The present invention relates to a technical field slope displacement prediction, and particularly to a permanent displacement of slope depth prediction model training method based on learning. Background technique [0002] Seismic landslide hazard analysis region refers to a potential seismic case, the analysis of its spatial probability distribution of seismic induced landslides, describe elements landslide danger provide a specific location, volume and the like. The main steps of the current regional seismic landslide hazard analysis is as follows: Slope parameter information acquisition, Newmark permanent slope displacement prediction model and establish regional seismic hazard assessment of landslide. Newmark displacement prediction model in which the permanent establishment, the expert is currently used mainly simple regression methods were established, taking samples of ground motion there is insufficient adequacy, and effectiveness should be improve...

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): G06F30/27G06N3/08G06F16/2458G06Q50/26
CPCG06F30/27G06N3/08G06F16/2465G06Q50/26
Inventor 程印王建锋刘同同张迎宾何毅余海洪
Owner SOUTHWEST JIAOTONG UNIV
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com