Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Power transformer potential fault diagnosis method

A technology for power transformers and fault diagnosis, which is applied in the measurement of electrical variables, instruments, and electrical measurement, and can solve problems such as heavy subjectivity and lack of objectivity.

Inactive Publication Date: 2018-05-11
朱震
View PDF3 Cites 4 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, if you directly assign BPA, the subjectivity is too heavy; if you assign the processed neural network output to BPA, the reliability coefficient is still an empirical value
The methods mentioned above all have certain limitations, and the method of determining BPA through expert experience lacks objectivity

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
  • Power transformer potential fault diagnosis method
  • Power transformer potential fault diagnosis method
  • Power transformer potential fault diagnosis method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach

[0050] The present invention is mainly used for potential fault diagnosis of power transformers, and the specific implementation methods are as follows:

[0051] Step 1. Determination of the fault type of the power transformer. Let Ω(Θ)={f 1 , f 2 ,..., f n} is a collection of transformer fault types. Practice shows that: transformer internal faults can be divided into overheating faults and discharge faults according to their nature. The statistical results of relevant literature on the types of oil-filled electrical equipment failures are shown in Table 2:

[0052] Table 2 Statistics of failure types of oil-filled electrical equipment

[0053] Fault type

Taiwan

ratio / (%)

overheating failure

226

53.0

High Energy Discharge Fault

65

18.1

Overheating and high energy discharge fault

36

10.0

spark discharge failure

25

7.0

Moisture or partial discharge failure

7

1.9

[0054] Therefore, the t...

example 1

[0076] Table 2 Example 1 Diagnosis Results

[0077]

[0078]

[0079] It can be seen from Table 2 that the mass value of overheating fault in the first-level diagnosis is the largest, and the mass value of low-temperature overheating in the second-level diagnosis is the largest, so this kind of fault is considered to be low-temperature overheating (<300°C). From the on-site actual diagnosis, through the on-site hanging core inspection, it was found that the transformer core had a multi-point ground fault, which belonged to the low-temperature overheating type.

example 2

[0081] On June 2, 2006, the 1# main transformer of 35KV Chengnan Substation was put into operation.

[0082] h 2 =38.5 μL / L; CH 4 = 19.7 μL / L; C 2 h 6 = 48.2 μL / L; C 2 h 4 = 4.83 μL / L; C 2 h 2 = 0.99 μL / L.

[0083] The three ratio codes are 1, 0, 0, and the preliminary diagnosis is arc discharge. However, using the method of this paper for diagnosis, the results are shown in Table 3:

[0084] Table 3 Example 2 Diagnosis Results

[0085]

[0086] It can be seen from Table 3 that the mass value of other faults is the largest, so this type of fault is considered to be other faults. During the actual maintenance, no abnormality was found in the conventional electrical test of the transformer. After the oil was drained, it was found that one of the three fixing screws of the shielding cover at the end of the bushing was loose, but there was no obvious discharge trace. After tightening, the oil was degassed. Chromatographic analysis of the main transformer commissionin...

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 provides a power transformer potential fault diagnosis method. The method comprises the steps of: for a data sample to be detected, employing a codeless four-ratio method and a characteristic gas method to perform preprocessing of the data sample to be detected, and generating two sets of corresponding characteristic vectors; employing a 'one-to-one' multi-class support vector machine (SVM) posterior probability estimation to provide a basic probability allocation (BPA) for an evidence theory; and finally, constructing a multi-stage comprehensive diagnosis system to perform diagnosis of potential faults of a power transformer. A model with a high generalization ability is established in a condition of ensuring limited fault samples, the defect of hard decision of theSVM is overcome, the probability output is provided, and the objectivity of the subsequent BPA assignment is ensured; and moreover, the analyzability of the system is increased based on evidence fusion of the posterior probability SVM, and the identification ability of a fault mode is effectively improved.

Description

technical field [0001] The invention relates to a method for latent fault diagnosis of a power transformer, relates to a posterior probability support vector machine and an evidence theory algorithm, and belongs to the technical field of electric engineering. Background technique [0002] The power transformer is an important equipment in the power system. Once a failure occurs, it will cause huge economic losses. Therefore, the fault diagnosis of power transformer plays a very important role in the safe operation of power system. Due to the complex structure of power transformers, there are many uncertain factors involved in faults, and the faults reflected by traditional characteristic signals (such as oil chromatography, dielectric loss, iron core ground current, partial discharge, etc.) have great limitations. Therefore, it is urgent to study some new intelligent algorithms applied to the fault diagnosis of power transformers. [0003] For many years, the dissolved gas...

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): G06K9/62G01R31/06
CPCG01R31/72G06F18/2411
Inventor 朱震
Owner 朱震
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products