Unlock instant, AI-driven research and patent intelligence for your innovation.

Self-adaptive parameter optimization method of SVM approximation model

A technology of approximate model and optimization method, which is applied in the field of approximate model, can solve the problem of complex and time-consuming selection of SVM approximate model parameters, and achieve the effect of improving accuracy

Inactive Publication Date: 2018-06-01
NANJING UNIV OF SCI & TECH
View PDF2 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide an adaptive SVM approximate model parameter optimization method, effectively solve the problem of complex and time-consuming selection of SVM approximate model parameters, and improve model accuracy and generalization ability

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
  • Self-adaptive parameter optimization method of SVM approximation model
  • Self-adaptive parameter optimization method of SVM approximation model
  • Self-adaptive parameter optimization method of SVM approximation model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0025] The present invention will be described in further detail below in conjunction with the accompanying drawings.

[0026] combine Figure 1 to Figure 4 , an adaptive SVM approximation model parameter optimization method, the steps are as follows:

[0027] Step 1. Use the optimal Latin hypercube design method combined with the analysis of the physical model f to obtain the training sample set S={(x(i),f x (i)), i=1,2,...,n s};details as follows:

[0028] Step 1-1. Divide each dimension variable of the training sample into n within its value range s intervals, a sampling point is randomly generated in each interval, and randomly combined to form an n s ×m training sample matrix X, which is the training sample points, turn to step 1-2.

[0029] Step 1-2, set OUT=1, IN=1, optimal training sample matrix X best =X; go to step 1-3.

[0030] Step 1-3. Exchange any two elements in the IN column of the training sample matrix X for A times to construct a batch of new training...

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 discloses a self-adaptive parameter optimization method of an SVM approximation model which is used for increasing precision of the SVM approximation model. The method comprises the following steps: obtaining a training sample set and a test sample set by combining an optimal Latin Hypercube experiment design method with analyses of a physical model; taking relative errors of the test sample set as fitness values and optimizing parameters of the SVM approximation model with a genetic algorithm in order to obtain an optimal parameter SVM approximation model; taking relative errorsof the test sample set as a criterion and updating training and test sets with a greedy algorithm; and building a novel SVM approximation model on the basis of a novel sample and iterating the modeltill the SVM approximation model meets the requirement for precision. Parameters of the SVM approximation model are easy to select and precision is easily enhanced. Analysis efficiency of a complex physical model in engineering is greatly increased. The self-adaptive parameter optimization method plays a big role in engineering.

Description

technical field [0001] The invention relates to the technical field of approximate models of complex physical models, in particular to an adaptive SVM approximate model parameter optimization method. Background technique [0002] Approximate model is a technology for optimization. The optimization method based on approximate model is one of the best hope methods for solving large nonlinear problems. Due to its high efficiency and time saving, it is widely used in the field of engineering optimization. Approximate model technology is mainly composed of two parts: experimental design method and approximate model construction, both of which directly affect the prediction accuracy and construction efficiency of the approximate model. [0003] In fact, the selection of the most representative combination of reasonable test sample points in the design space directly determines the number and spatial distribution of sample points required to construct the approximate model, and aff...

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): G06F19/00
CPCG16Z99/00
Inventor 赵天枝葛建立曹杰王雪嫣孙全兆杨国来王浩
Owner NANJING UNIV OF SCI & TECH