Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Method for combining RBF (Radial basis function) surrogate models by optimal point-by-point weighting

An approximate model and weight technology, which is applied in special data processing applications, instruments, electrical and digital data processing, etc., can solve the problems of inappropriate weight coefficients, poor effects, and inability to make full use of them, so as to reduce computational costs, improve prediction accuracy, The effect of improving optimization efficiency

Inactive Publication Date: 2015-08-19
DALIAN UNIV OF TECH
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The existing point-wise weight combination methods, the weight coefficients determined by them are either inappropriate or cannot fully utilize the advantages of different single RBF approximation models
Therefore, the performance is worse than the average weight combination method in some numerical tests

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
  • Method for combining RBF (Radial basis function) surrogate models by optimal point-by-point weighting
  • Method for combining RBF (Radial basis function) surrogate models by optimal point-by-point weighting
  • Method for combining RBF (Radial basis function) surrogate models by optimal point-by-point weighting

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] The whole algorithm has three key points: 1) Determine the weight of known sample points; 2) Construct the optimal weight function for each single RBF approximate model; 3) Construct the combined RBF approximate model.

[0025] In order to make the purpose, technical solution and advantages of the present invention clearer, the present invention will be further described below in conjunction with the accompanying drawings and specific examples.

[0026] Consider a one-dimensional function and a set of sample points X={x 1 ,x 2 ,...,x m}, choose MQ basis function, TPS basis function and GA basis function to construct three different single RBF approximate models, see figure 2 . The specific implementation process of the combined RBF approximate model is as follows:

[0027] Step 1. Determine the weight of known sample points.

[0028] The prediction error of a single RBF approximation model at each known sample point is calculated by cross-validation method. At th...

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 present invention, belonging to the field of engineering design and optimization, relates to a method for combining surrogate models by optimal point-by-point weighting, which improves the prediction accuracy of RBF surrogate models. The method is characterized by comprising the following steps: 1) confirming a weighting coefficient of each single RBF surrogate model at a known sample point by a cross validation method; 2) establishing an optimal weighting function by using an inverse-distance model and optimal attenuation coefficients, according to the weighting coefficients of the known sample points; and 3) establishing the combined RBF surrogate model by using the optimal weighting function and single RBF surrogate models based on different primary functions. The method for combining the RBF surrogate models according to the present invention, is higher and more stable in prediction accuracy, and has a higher practical value on optimized project designs based on numerical simulation analysis of computers.

Description

technical field [0001] The present invention is in the field of engineering design and optimization. Specifically, it relates to a combined method for improving the prediction accuracy of an RBF approximate model. Background technique [0002] In recent years, high-precision computer numerical simulation models (Simulation models) have been widely used in engineering structure analysis and optimization design problems. The computer numerical simulation model can effectively simulate the physical characteristics such as the stress state of the actual engineering structure, and provide design basis for designers, so it can significantly improve product design efficiency and save design cost. However, the process of high-precision computer numerical simulation is usually time-consuming. For the analysis of complex engineering structures, it usually takes hours, days or even longer. In order to save computing resources, approximate model (Surrogate models) technology is widely...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06F17/50
Inventor 徐胜利刘海涛王晓放
Owner DALIAN UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
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