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

Multifrequency ultrasonic testing regression prediction method for acid value of transformer oil

A regression prediction and transformer oil technology, applied in the direction of using sound wave/ultrasonic wave/infrasonic wave to analyze fluids, instruments, complex mathematical operations, etc. It can solve problems such as poor fitting degree and large data error

Inactive Publication Date: 2018-02-02
ELECTRIC POWER SCI RES INST OF GUIZHOU POWER GRID CO LTD
View PDF4 Cites 18 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The technical problem solved by the present invention is to provide a multi-frequency ultrasonic test regression prediction method for the acid value of transformer oil, which uses the genetic algorithm to optimize the acid value regression model of transformer oil built by the support vector machine to predict the acid value in transformer oil, so as to solve the problem of traditional The prediction results of the standard support vector machine model and the actual transformer oil measurement results are poorly fitted and the data errors are large

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
  • Multifrequency ultrasonic testing regression prediction method for acid value of transformer oil
  • Multifrequency ultrasonic testing regression prediction method for acid value of transformer oil
  • Multifrequency ultrasonic testing regression prediction method for acid value of transformer oil

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0042] A transformer oil acid value multi-frequency ultrasonic test regression prediction method, comprising the following steps:

[0043] Step 1: Perform multi-frequency ultrasonic characteristic test and acid value measurement on transformer oil samples, and use support vector regression machine SVR (Support Vector Regression) to establish a regression prediction model for transformer oil acid value: conduct multi-frequency ultrasonic waves on transformer oil samples with different operating years Characteristic testing, and at the same time measure the acid value in transformer oil samples based on the indicator method, with the multi-frequency ultrasonic test data as input and the acid value test results in oil as output, use the support vector regression machine SVR to establish a regression prediction model for the acid value of transformer oil;

[0044] Step 2: Use the genetic algorithm GA (genetic algorithm) to optimize the combination of the penalty function penalty fact...

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 multifrequency ultrasonic testing regression prediction method acid value of transformer oil. The method comprises the following steps of 1), performing multifrequency ultrasonic characteristic testing and acid value measuring on a transformer oil sample, establishing a transformer oil acid value regression prediction model by means of a support vector regression machine;2), performing optimized combination of a punishment function punishment factor C, a kernel function parameter sigma and an insensitive loss function lost factor epsilon of the transformer coil acidvalue regression prediction model by means of a genetic algorithm; 3), establishing a transformer oil acid value multifrequency ultrasonic wave testing regression prediction model of the support vector regression machine based on the genetic algorithm; and 4), acquiring a multifrequency ultrasonic wave acoustic frequency spectrum of the transformer oil sample, and predicating the acid value in thetransformer oil. The multifrequency ultrasonic testing regression prediction method realizes acquisition of a predicated value of the acid value in the transformer oil through prediction and calculation, high fitting degree between the predicated value of the transformer oil acid value regression model which is established by the support vector machine through genetic algorithm optimization, andlow data error.

Description

technical field [0001] The invention relates to the technical field of transformer oil aging degree detection, in particular to a regression prediction method for transformer oil acid value multi-frequency ultrasonic testing. Background technique [0002] As the hub of the power system, large oil-immersed power transformers are the most expensive and important equipment in the power system. They play a pivotal role in the process of power transmission and transformation. Their safe and reliable operation is of decisive significance to the entire power system. Once the power transformer fails, it will bring huge losses to the national economy. The transformer oil in operation will gradually age under the influence of electricity, heat, force, water and oxygen, and produce a series of oxides that dissolve in the transformer oil, among which the most harmful ones are acidic substances such as formic acid, acetic acid and levulinic acid. The appearance and increase of acidic su...

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): G06K9/62G06F17/18G01N29/02
CPCG06F17/18G01N29/02G06F18/2111G06F18/2411
Inventor 余鹏程张英刘喆吴国卿蒋震牧灏李军卫
Owner ELECTRIC POWER SCI RES INST OF GUIZHOU POWER GRID CO LTD
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