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

Generating-probability-model-based transformer fault detection method

A transformer fault and detection method technology, applied in the direction of instruments, character and pattern recognition, computer components, etc.

Active Publication Date: 2016-02-17
ELECTRIC POWER RESEARCH INSTITUTE, CHINA SOUTHERN POWER GRID CO LTD
View PDF3 Cites 7 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

These methods do not use continuous values ​​to describe the characteristics of transformer parameters, because the transformer parameters contain data of different unit categories such as voltage, temperature, and gas concentration, and it is difficult to directly describe them in a feature vector

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
  • Generating-probability-model-based transformer fault detection method
  • Generating-probability-model-based transformer fault detection method
  • Generating-probability-model-based transformer fault detection method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0032] The transformer fault detection method based on generation probability model of the present embodiment, it comprises the following steps:

[0033] A. Obtain the transformer state sample data; it specifically includes the following steps:

[0034] A1. There are two possible states for defining the transformer, which are normal state and abnormal state respectively. Use S 0 Indicates normal state, use S 1 Indicates an abnormal state, then N=1. ;

[0035] A2. For the two states S of the transformer 0 and S 1 Collect 1000 training data samples respectively, then M 0 =1000, M 1 = 1000, vector and vector Respectively represent the state S 0 and S 1 A training sample data of i∈[1,M n ], n∈{0,1}, the parameters of the transformer are voltage V and current I, then C=2, a sample data can be used express.

[0036] B. Transformer state model learning based on mixed Gaussian model; it specifically includes the following steps:

[0037] B1, use the 1000 training data...

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 provides a generating-probability-model-based transformer fault detection method. The method comprises the following steps: A, obtaining transformer state sample data; B, carrying out Gaussian-mixture-model-based transformer state model learning; C, carrying out Gaussian-mixture-model-based transformer state feature extraction; and D, carrying out support-vector-machine-based transformer fault detection. According to the invention, problems that robustness of the transformer fault detection to the environment change is low and sensitivity to the parameter measurement error is low according to the traditional discrete transformer parameter characteristic extraction method based on manual threshold value setting can be solved and unified characteristic description of continuous numerical values by different unit category parameters is realized.

Description

technical field [0001] The invention relates to a power device fault detection technology, in particular to a transformer fault detection method based on a generation probability model. Background technique [0002] Transformer fault automatic detection algorithm based on transformer parameters plays a vital role in smart grid. The traditional transformer fault detection algorithm is based on human judgment to extract features of discrete values, such as discretization and partition representation of parameters such as temperature and air pressure by artificially setting thresholds. This feature extraction method of artificial discretization has many disadvantages. On the one hand, artificially setting the threshold is very sensitive to environmental changes in the detection results. On the other hand, the discrete numerical features make the detection results very sensitive to errors. [0003] Most of the existing related research is devoted to eliminating the artificial t...

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): G06K9/62
CPCG06F18/214
Inventor 黄文琦许爱东郭晓斌陈华军蒋屹新袁小凯张福铮杨航陈富汉蒙家晓关泽武陈立明黄建理吴争荣
Owner ELECTRIC POWER RESEARCH INSTITUTE, CHINA SOUTHERN POWER GRID CO LTD