Insulator equivalent salt density optical fiber test method based on particle swarm support vector machine

A technology of support vector machine and equivalent salt density, which is applied in the field of value salt density optical fiber detection and value salt density detection, can solve the problems of low efficiency, real-time online detection, etc., and avoid economic loss and the harsh environment of live measurement , reduce pollution flashover accidents, save manpower and material resources

Inactive Publication Date: 2010-02-24
XIAN UNIV OF TECH
0 Cites 20 Cited by

AI-Extracted Technical Summary

Problems solved by technology

[0017] The purpose of the present invention is to provide an insulator equivalent salt-dense optical fiber detection method based on particle swarm support vector machine, to solve the problems of low ef...
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

Method used

Utilize support vector machine (Support Vector Machine, SVM) to set up above-mentioned nonlinear regression model in solving the advantages such as small sample, nonlinearity, high dimension, local minimum point, and utilize part...
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

Abstract

The invention discloses a high voltage insulator equivalent salt density optical fiber test method based on a particle swarm support vector machine. In the method, firstly luminous flux attenuation and relative humidity are tested to be transmitted to a supervisory computer in a manner of wireless communication; then the luminous flux attenuation is subject to normalization pretreatment; referringto a standard relation curve graph of dust ratio with the luminous flux attenuation and the relative humidity, corresponding dust ratio is searched; and the particle swarm support vector machine is used for establishing a nonlinearity regression model among an insulator equivalent salt density value, the luminous flux attenuation, the relative humidity, and the dust ratio to obtain the insulatorequivalent salt density value. The high voltage insulator equivalent salt density optical fiber test method based on the particle swarm support vector machine in the invention realizes on-line test ofinsulator salt density with high precision, is suitable for various voltage levels, avoids the defects of the traditional salt density measurement such as time consuming, effort consuming and poor measurement effect, and improves reliability of power supply system.

Application Domain

Technology Topic

Nonlinear regressionComputational physics +10

Image

  • Insulator equivalent salt density optical fiber test method based on particle swarm support vector machine
  • Insulator equivalent salt density optical fiber test method based on particle swarm support vector machine
  • Insulator equivalent salt density optical fiber test method based on particle swarm support vector machine

Examples

  • Experimental program(1)

Example Embodiment

[0075] Example
[0076] Step 1: Install the inspection system that has been fully tested in the laboratory on a 110KV high-voltage transmission line. The insulator is a suspension glass insulator LXP-160. The on-site detection device detects the luminous flux attenuation at five different time points Φ 1 And environmental relative humidity RH, as shown in Table 1.
[0077] Table 1 The luminous flux attenuation at five different time points Φ 1 And relative humidity RH
[0078]
[0079] Attenuate the luminous flux Φ 1 , The relative humidity RH is transmitted to the management computer through wireless communication;
[0080] Step 2: Manage the attenuation of the received light flux of the computer Φ 1 And relative humidity RH data information, attenuation of luminous flux Φ 1 Perform normalization pretreatment according to the following formula to obtain the luminous flux attenuation Φ after normalization pretreatment 2 :
[0081] Φ 2 = β i - β o β i X 100 %
[0082] Where β i Is the incident light intensity, β o Is the intensity of the emitted light.
[0083] The processing results are shown in the following table:
[0084] Table 2 Luminous flux attenuation Φ after normalized pretreatment 2
[0085]
[0086] Step 3: Refer to dust ratio X and luminous flux attenuation Φ 2 , The standard curve diagram of relative humidity RH, such as figure 2 As shown, according to the detected luminous flux attenuation Φ 2 And relative humidity RH, find the corresponding dust ratio X, as shown in Table 3.
[0087] Table 3 Luminous flux attenuation Φ 2 , Relative humidity RH and corresponding dust ratio X
[0088]
[0089] Step 4: Use particle swarm support vector machine to establish the equivalent salt density value ESDD and luminous flux attenuation Φ of the insulator surface 2 The nonlinear regression model between RH, relative humidity RH and dust ratio X is implemented according to the following steps:
[0090] (1) Collect sample data in artificial pollution laboratories and power production sites, including light attenuation, relative humidity, equivalent salt density, and dust ratio, correct and preprocess the sample data, and divide the sample data into training samples And the test sample;
[0091] (2) Perform initial settings, including setting the group size, number of iterations, and randomly giving γ and σ as the initial positions of the particles;
[0092] (3) Using the γ and σ corresponding to the individual particles, establish the learning prediction model of LS-SVM, and use the prediction error rate of the model to the test sample as the fitness value y of the particle i;
[0093] (4) Adapt the particle to the value y i Optimal fitness value Compare, if y i Less than Then replace the fitness value of the previous round with the new fitness value, and replace the particles in the previous round with the new particles, namely y pbest i = y i , x best i = x i ;
[0094] (5) The optimal fitness value of each particle Respectively with the optimal fitness value of all particles in the previous round Compare if Less than The optimal fitness value of each particle is used to replace the fitness value of the original particle, and the current state of the particle is saved;
[0095] (6) Judge whether the fitness value or the number of iterations meets the requirements, if it does not meet the requirements, perform a new round of calculations, according to the formula v=w·v+c 1 r 1 (p best -p)+c 2 r 2 (g best -p) and p=p+β·v move the particles to generate new particles, return to step (3), if the fitness value meets the requirements, the calculation ends, the individual particle corresponds to the most suitable LS-SVM γ And σ to establish a nonlinear regression model based on LS-SVM. The input parameters of the model are luminous flux attenuation, relative humidity, and dust ratio, and the output parameters are equivalent salt density values;
[0096] Step 5: Attenuate the luminous flux obtained in step 2 Φ 2 , The relative humidity RH and the dust ratio X obtained in step 3 are substituted into the nonlinear regression model obtained in step 4 to obtain the equivalent salt density value ESDD on the surface of the insulator. The results obtained are shown in Table 4:
[0097] Table 4 Equivalent salt density value ESDD on the surface of insulator
[0098]
[0099] It can be seen from Table 4 that the relative error of the equivalent salt density value of the high-voltage insulator predicted by the method of the present invention and the equivalent salt density value measured regularly by the traditional method is within the range of 4.5%-6.9%, which meets the requirements of the system. The measurement error is less than 10% requirement.
[0100] The high-voltage insulator equivalent salt density fiber monitoring method based on the particle swarm support vector machine of the present invention realizes the online detection of the insulator salt density, has high accuracy, is suitable for various voltage levels, and avoids the time-consuming, laborious and measurement effect of the traditional salt density measurement method The bad shortcomings improve the reliability of the power supply system.
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

Description & Claims & Application Information

We can also present the details of the Description, Claims and Application information to help users get a comprehensive understanding of the technical details of the patent, such as background art, summary of invention, brief description of drawings, description of embodiments, and other original content. On the other hand, users can also determine the specific scope of protection of the technology through the list of claims; as well as understand the changes in the life cycle of the technology with the presentation of the patent timeline. Login to view more.
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

Similar technology patents

Powder metallurgy start chainwheel and gear

Owner:ZAOZHUANG RUNYUAN POWDER METALLURGY

Classification and recommendation of technical efficacy words

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