Least square support vector machine-based fluctuating wind velocity prediction method

A technology of support vector machine and pulsating wind speed, which can be used in forecasting, computer parts, instruments, etc., and can solve the problems of time-consuming and low simulation accuracy of support vector machine.

Inactive Publication Date: 2016-11-16
SHANGHAI UNIV
View PDF3 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] The purpose of the present invention is to provide a fluctuating wind speed prediction method based on the least squares support vector machine to solve the problems of low simulation accuracy and time-consuming of traditional support vector machines in wind speed prediction

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
  • Least square support vector machine-based fluctuating wind velocity prediction method
  • Least square support vector machine-based fluctuating wind velocity prediction method
  • Least square support vector machine-based fluctuating wind velocity prediction method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0094] see figure 1 , a fluctuating wind speed prediction method based on the least squares support vector machine, the specific steps are as follows:

[0095] 1) Select a super high-rise building and determine the parameters required for numerical simulation of fluctuating wind speed: the simulated building height and the heights of the simulated wind speed points, the average wind speed at a height of 10 meters, surface roughness coefficient, ground roughness index, simulation correlation function, a set number of fluctuating wind speed time histories uniformly distributed along the height generated by the numerical simulation of the ARMA method, as the limited original fluctuating wind speed sample data;

[0096] 2) Use the PSO algorithm to determine the parameters. Determine the population size and evolution times, set c 1 and c 2 、w max and w min The value of r is randomly generated 1 and r 2 ; According to the preliminary range of model parameters obtained by cros...

Embodiment 2

[0100] This is based on the fluctuating wind speed prediction method of the LSSVM model using the B-RBF combined kernel function. The specific steps are as follows:

[0101] The first step is to select a super high-rise building with a height of 150 meters in a city center, and take points every 10 meters along the height direction as the simulated wind speed points. Other relevant parameters are shown in Table 1:

[0102] Table 1 Related simulation parameters

[0103]

[0104] Indicates the average wind speed at a height of 10m.

[0105] In the second step, a certain number of fluctuating wind speed time histories uniformly distributed along the height generated by the ARMA numerical simulation are used as limited original fluctuating wind speed sample data.

[0106] In order to verify the validity of the prediction based on the machine learning method, a part of the sample data set needs to be used for machine learning, and another part of the sample data set is 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
Login to view more

PUM

No PUM Login to view more

Abstract

The invention proposes a least square support vector machine (LSSVM)-based fluctuating wind velocity prediction method. The method comprises the steps of firstly numerically simulating fluctuating wind velocities of 15 simulation points of a super high-rise building through an ARMA method to serve as sample data; secondly establishing a B-RBF combined kernel function by linearly combining a B-spline kernel function with a radial basis function (RBF) kernel function, further searching for optimal parameters of a B-RBF combined kernel function-based LSSVM model by adopting a particle swarm optimization (PSO) algorithm to minimize a prediction error of the model, and establishing the B-RBF kernel function-based LSSVM model by adopting the parameters after optimization; and finally predicting a fluctuating wind velocity of an intermediate layer through fluctuating wind velocity samples of upper and lower layers, adopting an average error, an average absolute error, a root-mean-square error and related coefficients as evaluation indexes, and performing comparison with prediction results of a single B-spline (including a first-order B-spline, a third-order B-spline and a fifth-order B-spline) kernel function-based LSSVM model and a single RBF kernel function-based LSSVM model.

Description

technical field [0001] The invention relates to a fluctuating wind speed prediction method based on a least squares support vector machine, based on a machine learning method of a least squares support vector machine (LSSVM) using a B-spline-radial basis (B-RBF) combined kernel function Time history method for predicting fluctuating wind speed. Background technique [0002] The standard support vector machine (SVM) method for function fitting is mainly to convert the input samples from the low-dimensional input space to a high-dimensional feature space through nonlinear mapping, and then minimize the loss function in this high-dimensional space to obtain Linear fit function. According to Mercer's theorem, for support vector machines, the function regression fitting problem can be described as solving a constrained quadratic programming problem, the number of constraints is equal to the capacity of the sample, although the related kernel function is used to avoid the explici...

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): G06Q10/04G06K9/62
CPCG06Q10/04G06F18/2411
Inventor 徐言沁李春祥
Owner SHANGHAI UNIV
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
Try Eureka
PatSnap group products