Partial discharge signal comprehensive entropy value-based electrical equipment dangerous discharge discrimination method

Abstract

The invention discloses a partial discharge signal comprehensive entropy value-based electrical equipment dangerous discharge discrimination method, which relates to a discharge discrimination method.Multiple dangerous signals are hidden in a data sequence and can not be found, and thus, equipment failure is caused, and the failure can not be warned in advance. The method comprises steps: 1) a discharge state feature library of insulation defects under long-term voltage is built, wherein the library building comprises the following sub steps of building of a test system, building of a transformer defect test model, boosting test and data analysis; 2) neural network learning for solid insulation surface defect discharge signal data in a sample library is carried out; 3) the neural networkis used to carry out real-time monitoring on the data; and 4) according to the detection result, diagnosis and early warning are carried out on solid insulation surface defect discharge failure of high voltage power equipment. Effective evaluation and early warning on the risks of breakdown or flashover failure caused by insulation defect discharge under field engineering conditions can be realized.

Description

technical field [0001] The invention relates to a method for discriminating discharge, in particular to a method for discriminating dangerous discharge of electrical equipment based on the comprehensive entropy value of partial discharge signals. Background technique [0002] Partial discharge is one of the important parameters to reflect the insulation state of large-scale power equipment, and it is also an important means to discover potential defects of high-voltage insulation equipment. It has the characteristics of high sensitivity, timely and effective. The evaluation method of existing transformers and GIS equipment is relatively simple, that is, blind evaluation is carried out based on the historical data or maintenance records and operating experience based on the size of the discharge amplitude. However, the amount of detected data is often huge, and many dangerous signals are hidden in the data sequence and cannot be found, resulting in equipment failures that can...

AI Technical Summary

Problems solved by technology

The evaluation method of existing transformers and GIS equipment is relatively simple, that is, through the size of the discharge amplitude, and then conduct blind evaluation based on the historical data or maintenance records and operating experience.

However, the amount of detected data is often huge, and many dangerous signals are hidden in the data sequence and cannot be found, resulting in equipment failure and cannot be warned in advance

Images