电工技术学报2024，Vol.39Issue(23)：7628-7638,11.DOI:10.19595/j.cnki.1000-6753.tces.231839
虚实融合的变压器内绝缘状态孪生方法
State Twinning Method of Transformer Internal Insulation by Virtual-Real Fusion
摘要
Abstract
The national development plan calls for the digital and intellectual transformation of power enterprises to support the construction of new power systems.Power transformer as the core equipment of the power grid,its operational reliability is related to the safety of the entire power grid,therefore,the development of digital twin transformer is of great significance to support the intelligent,digital,visualized operation and maintenance of high-end power equipment.However,there are still significant challenges for transformers due to the lack of corresponding digital twin capabilities,such as insufficient sensors,unknown parametric mechanism,and lack of post-processing techniques.For this reason,this paper carries out a research on a state twinning method for transformer internal insulation by virtual-real fusion. Firstly,the internal temperature field distribution of the transformer is the data basis for calculating the state parameter distribution.Considering the complex multi-physical field coupling effect of the transformer,the calculated losses are used as the heat source of the temperature field simulation for the multi-physical field coupling calculation,and the temperature field distribution corresponding to the real-time load change is derived based on the load rate monitored in the field.Secondly,the temperature field distribution of virtual calculation and the moisture in oil data of physical monitoring are combined to analyze the distribution law of moisture in paper at each local location inside the transformer.Based on the relationship between the total moisture content in the paper of multiple transformers in the field and the operation time,an empirical formula is established,and the oil-paper equilibrium equation and the law of moisture diffusion are utilized to determine the content and distribution.Thirdly,considering the effect of moisture distribution on the moisture translational factor,a dynamic derivation model of degree of polymerization distribution based on the joint action of time-temperature-moisture distribution is established,and the results of the degree of polymerization distribution of the insulating cardboard inside the transformer are obtained.Finally,due to the fact that the multi-physics field simulation calculation results and the dynamic deduction numerical results are discrete point data,this paper adopts the point-by-point insertion method of Delaunay triangular mesh for the continuity of the discrete data,and the state parameter data obtained by virtual-realistic fusion calculation are visualized in this paper. The results show that the moisture distribution calculated by virtual-real fusion is mainly concentrated in the laminated cardboard on both sides and the bottom of the solid insulation,with the highest moisture content of about 3.2%in the 48th year,which is greatly influenced by the temperature field distribution and environmental factors.The rule of the degree of polymerization distribution in the windings is in general agreement with the measurements reported by CIGRE,and the lowest value is located in the hot spot temperature region of the low-voltage windings,with a value of 252.15 in the 48th year,which verifies the validity and reasonableness of the methodology.This study realizes the assessment of the parametric distribution for the aging state parameter inside the transformer insulation,and lays the foundation for the subsequent comprehensive construction of the digital twin of the transformer.关键词
虚实融合/变压器/数字孪生/状态参量/分布监测Key words
Virtual-real fusion/transformer/digital twin/state parameter/distributed monitoring分类
信息技术与安全科学引用本文复制引用
罗豪,成立,杨丽君,赵学童,张永泽..虚实融合的变压器内绝缘状态孪生方法[J].电工技术学报,2024,39(23):7628-7638,11.基金项目
国家重点研发计划资助项目(2021YFB2401700). （2021YFB2401700）
