Method for transformer fault graphic early warning
A transformer fault and transformer technology, applied in the direction of testing dielectric strength, etc., can solve the problem of slow real-time response of the transformer fault graphic early warning system, and achieve the effects of fast calculation speed, simple modeling, and guaranteed projection relationship.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Publication Date
- 2015-01-21
Smart Images
Figure 1 Figure 2 Figure 3
Abstract
Description
technical field
[0001] The invention relates to a method for using a three-dimensional wire frame model to carry out graphic early warning of transformer faults. Background technique
[0002] The transformer is under the action of working voltage for a long time, and partial discharge is prone to occur in weak insulation. Generator partial discharge on-line monitoring is an effective means to prevent generator insulation damage accidents. The current technologies mainly include: two fractal features (fractal latitude and missing items) are extracted from the three-dimensional features of partial discharge, and the neural network is used to identify the discharge pattern; the three-dimensional discharge pattern is recognized based on the extended neural network technology; The model analyzes the statistical characteristics of partial discharge and uses three-dimensional graphs to display; analyzes the statistical characteristics of partial discharge and simulates partial dis...
Examples
Embodiment Construction
[0019] The present invention will be described in further detail below in conjunction with the accompanying drawings.
[0020] The three-dimensional wireframe model that the present invention adopts is defined as: in:
[0021] (1) f(t) is the analog signal of transformer partial discharge collected by the embedded system;
[0022] (2) is a wavelet function, through the wavelet transform function After signal processing, get the phase angle-discharge amount-discharge times parameter;
[0023] (3) P is a vector obtained after feature extraction is performed on the discharge sample;
[0024] (4) T 3 D = a 11 a 12 a 13 ...