CN-121981048-A - Short-circuit resistance evaluation method for three-dimensional coiled core transformer

CN121981048ACN 121981048 ACN121981048 ACN 121981048ACN-121981048-A

Abstract

The invention provides a short-circuit resistance evaluation method of a three-dimensional coiled iron core transformer, which belongs to the technical field of transformers, and comprises the steps of calculating leakage magnetic field distribution and winding axial short-circuit line force under a short-circuit condition by establishing a three-dimensional equivalent magnetic network model, constructing an end insulation structure parameter space, establishing a three-dimensional mechanical model, calculating winding equivalent stress distribution and winding axial displacement, classifying and identifying short-circuit current time sequence data by adopting an anti-short-circuit evaluation model based on a gate control circulation unit, simultaneously capturing transient impact characteristics and accumulated effect characteristics of short-circuit current by updating an information selective transmission mechanism of a gate and a reset gate, converting the slight difference of current waveforms in the process of multiple short-circuit impact into a judging basis of an end insulation assembly state, outputting a short-circuit resistance evaluation result, and solving the technical problem that the short-circuit failure state of the end insulation assembly of the three-dimensional coiled iron core transformer is difficult to accurately identify.

Inventors

XU JIANWEN
ZANG YING
GAO TENG
DU YINGHUI
JI WEITAO
ZHANG ZHIQIANG
ZHANG CHENXI
CHEN DEZHI
Xin Ziyuan

Assignees

山东电工电气集团科学技术研究有限公司
沈阳工业大学

Dates

Publication Date: 20260505
Application Date: 20251230

Claims (10)

1. A three-dimensional equivalent magnetic network model is established, leakage magnetic field distribution under a short-circuit working condition is calculated through an ultra-relaxation iterative magnetic field solving algorithm, an instantaneous maximum value of short-circuit current is determined according to an effective value of voltage at a transformer end, a short-circuit impedance percentage and a short-circuit current impact coefficient, axial short-circuit line force of windings and accumulated axial force of the windings are solved, an end insulation structure parameter space is constructed, a three-dimensional mechanical model is established, a short-circuit electromagnetic force boundary condition is applied to calculate equivalent stress distribution of the windings and axial displacement of the windings, a triangular baffle compression deformation limit value is determined, an end insulation cushion net distance threshold value, end insulation cushion end coverage rate and a triangular baffle thickness threshold value are set, short-circuit current time sequence data are classified and identified by adopting an anti-short-circuit evaluation model based on a gate control circulating unit, information flow is controlled in sequence processing steps through an update gate and a reset gate, correlation is maintained, and an anti-short-circuit capability evaluation result is output according to the end insulation assembly state category and structural parameter coincidence degree.
2. The method of claim 1, wherein the building of the three-dimensional equivalent magnetic network model disperses the core region, the oil passage region, and the winding region into a combined network of rectangular flux guiding units and annular flux guiding units.
3. The method of claim 2, wherein the rectangular flux guide unit equivalent flux guide is calculated as the product of vacuum magnetic permeability and material relative magnetic permeability multiplied by the ratio of rectangular width to rectangular length divided by rectangular height.
4. A method according to claim 3, wherein a super-relaxation iterative magnetic field solution algorithm introduces a relaxation factor to weight average the new permeability and the old permeability to accelerate convergence and loop iteration until the difference between the calculated magnetic fluxes of two adjacent times is less than a set threshold.
5. The method of claim 4, wherein the instantaneous maximum value of the short circuit current is calculated as root 2 times the transformer terminal voltage effective value divided by the short circuit impedance and then times the short circuit current surge coefficient.
6. The method of claim 5, wherein the construction of the parameter space of the end insulation structure is performed by constructing a parameter variation range set from three design variables of end insulation pad net spacing, end insulation pad end coverage and triangular baffle thickness.
7. The method of claim 6, wherein the three-dimensional mechanical model is built comprising a finite element structural analysis model of the end insulating spacer, the triangular baffle, the upper clamp, the lower clamp, the high voltage winding and the low voltage winding.
8. The method of claim 7, wherein the determination of the limit value of the compressive deformation of the triangular baffle plate is performed by determining the maximum allowable thickness variation of the triangular baffle plate under the action of short-circuit impact according to a stress-strain fitting curve of the wedman laminated wood insulating plate.
9. The method of claim 8 wherein the end spacer clear spacing threshold is set to a maximum allowable end spacer clear spacing that does not allow for axial displacement of the winding beyond an allowable displacement value.
10. The method of claim 9, wherein the input layer receives short circuit current time series data, the gated loop cell layer comprises 256 hidden cells, the fully connected layer maps the output to a probability distribution of 4 classes, and the output layer activates the function output insulation component state class by softmax.

Description

Short-circuit resistance evaluation method for three-dimensional coiled core transformer Technical Field The invention belongs to the technical field of transformers, and particularly relates to a short circuit resistance evaluation method of a three-dimensional coiled iron core transformer. Background The three-dimensional wound core transformer is widely applied to power systems, and the evaluation of the short circuit resistance of the three-dimensional wound core transformer depends on the design rationality and the state monitoring accuracy of an end insulation structure. According to the traditional evaluation method, the stress distribution and displacement of windings are calculated through finite element simulation, and whether an end insulating cushion block and a triangular baffle are invalid or not is judged by combining a material allowable stress threshold. In the current transformer state monitoring, as the characteristics of the tiny displacement of the end insulating cushion block and the gradual compression deformation of the triangular baffle plate are not obvious in single short-circuit impact measurement, the traditional evaluation method based on single stress strain measurement is difficult to capture the gradual failure process of the end insulating assembly caused by the accumulation of multiple short-circuit impacts, so that the failure state identification accuracy is insufficient. That is, the technical problem that the short-circuit failure state of the insulating component of the end of the three-dimensional coiled core transformer is difficult to accurately identify exists in the prior art. Disclosure of Invention In view of the above, the invention provides a method for evaluating the short-circuit resistance of a three-dimensional wound core transformer, which can solve the technical problem that the short-circuit failure state of an insulating component at the end of the three-dimensional wound core transformer is difficult to accurately identify in the prior art. The invention provides a three-dimensional roll iron core transformer anti-short circuit capacity assessment method, which is characterized in that a three-dimensional equivalent magnetic network model is established, leakage magnetic field distribution under a short circuit working condition is calculated through an ultra-relaxation iterative magnetic field solving algorithm, an instantaneous maximum value of short circuit current is determined according to an effective value of terminal voltage, a short circuit impedance percentage and a short circuit current impact coefficient of the transformer, an axial short circuit line force of a winding and an accumulated axial force of the winding are solved, an end insulation structure parameter space is constructed, a three-dimensional mechanical model is established, a short circuit electromagnetic force boundary condition is applied, winding equivalent stress distribution and an axial displacement of the winding are calculated, a triangular baffle compression deformation limit value is determined, a net spacing threshold value of an end insulation cushion block, an end coverage rate of the end insulation cushion block and a triangular baffle thickness threshold value are set, a gate control loop unit-based anti-short circuit assessment model is adopted to classify and identify state categories of an insulation component at an output end, information flow is controlled in a sequence processing step through an update gate and a reset gate, correlation is maintained, and an anti-short circuit capacity assessment result is output according to the state categories of the insulation component and the structural parameter coincidence degree. The method comprises the steps of establishing a three-dimensional equivalent magnetic network model, and dispersing an iron core area, an oil duct area and a winding area into a combined network of rectangular magnetic conduction units and annular magnetic conduction units. Wherein, the calculation of the equivalent flux guide of the rectangular flux guide unit is equal to the product of the vacuum magnetic permeability and the material relative magnetic permeability multiplied by the ratio of the rectangular width to the rectangular length divided by the rectangular height. The method comprises the steps of introducing a relaxation factor into a super-relaxation iterative magnetic field solving algorithm to perform weighted average on new magnetic permeability and old magnetic permeability so as to accelerate convergence and loop iteration until a magnetic flux difference value calculated in two adjacent times is smaller than a set threshold value. The calculation of the instantaneous maximum value of the short-circuit current is equal to the root number 2 multiplied by the effective value of the voltage of the transformer terminal divided by the short-circuit impedance and then multiplied by the short-circuit current impact coefficie