Short Circuit Current Zero Point Prediction Technology for Synthetic Test of Circuit Breaker Based on CNN-LSTM

WANG Zhi; MAO Xuefei; FENG Ying; LI Zheyuan; LI Yuchun

doi:10.13296/j.1001-1609.hva.2025.09.005

您当前的位置：

首页 >

文章列表页 >

Short Circuit Current Zero Point Prediction Technology for Synthetic Test of Circuit Breaker Based on CNN-LSTM

Research & Analysis | 更新时间：2025-09-05

- Short Circuit Current Zero Point Prediction Technology for Synthetic Test of Circuit Breaker Based on CNN-LSTM
- High Voltage Apparatus Vol. 61, Issue 9, Pages: 35-42(2025)
- 作者机构：
  
  1.北京理工大学，北京　100081
  2.北京航空航天大学，北京　100191
  3.中国电力科学研究院有限公司，北京　100192
- 作者简介：
- 基金信息：
  
  Science and Technology Project of SGCC (Research on Key Technologies for Enhancing the Quality of Distribution Switchgear Equipment)
- DOI：10.13296/j.1001-1609.hva.2025.09.005
  CLC：
- Received：13 January 2025，
  
  Revised：05 April 2025，
  
  Published：16 September 2025
- 稿件说明：
移动端阅览
WANG Zhi, MAO Xuefei, FENG Ying, et al. Short Circuit Current Zero Point Prediction Technology for Synthetic Test of Circuit Breaker Based on CNN-LSTM[J]. High voltage apparatus, 2025, 61(9): 35-42.
DOI：

WANG Zhi, MAO Xuefei, FENG Ying, et al. Short Circuit Current Zero Point Prediction Technology for Synthetic Test of Circuit Breaker Based on CNN-LSTM[J]. High voltage apparatus, 2025, 61(9): 35-42. DOI： 10.13296/j.1001-1609.hva.2025.09.005.

摘要

合成试验是检验大容量断路器短路开断能力的重要试验方法，文中针对合成试验中短路电流预测精度的要求，首先分析了威尔合成回路同步控制的原理，将过零点预测问题等效为时间序列预测问题，据此提出建立CNN-LSTM预测模型拟合短路电流序列，预测过零点。其次，通过某次试验中的短路电流情况建立预测模型，并比较了CNN-LSTM混合网络与单一LSTM网络的预测效果，在此基础之上，将该CNN-LSTM预测模型应用到不同短路电流下T100s及T100a试验中，实验结果显示该模型对于过零时刻的预测误差均≤4 μs，具备良好的泛化能力，研究结果对合成试验同步控制的准确实现具有实际意义。

Abstract

Synthetic test is an important test method for verifying the short-circuit breaking capacity of high power circuit breakers. As for the requirement to short-circuit current prediction accuracy in the synthetic test in this paper

the synchronous control principle of Weir synthetic circuit is firstly analyzed

and the problem of zero-crossing prediction is equivalently regarded as a time sequence prediction problem. Hereby

it is proposed to set up a CNNLSTM prediction model to fit the short-circuit current sequence and predict the current zero-crossing point. Then

a prediction model is set up based on the short-circuit current situation in a certain test

and the prediction effect of the CNN-LSTM hybrid network and that of the single LSTM network is compared. On this basis

the CNN-LSTM prediction model is applied to T100s and T100a tests under different short-circuit currents. The experiment results show that the prediction error of the model for the zero-crossing time is less than or equal to 4 μs and has good generalization ability. The research results have practical significance for the accurate realization of the synchronous control of synthetic test.

关键词

Keywords

references

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Method for Forecasting Current Zero Point in Synthetic Short Circuit Test

Synchronous Control System for Synthetic Test of Circuit Breaker

Fault Classification Method of Circuit Breaker Based on Improved Phase Trajectory Features of Coil Currents

Analysis on Oil Cylinder Cracking of Energy Storage Mechanism for 500 kV Circuit Breaker

Study on Electromagnetic Transient Characteristics During Switching Operation of 750 kV GIS Substation

Related Author

LI Peng

GAO Fei

LIU Pu

LIU Ping

HE Jian

JIANG Yuan

WU Jianwen

WANG Hefei

Related Institution

Xi’an High Voltage Apparatus Research Institute Co.,Ltd.

Xi’an Qinli Set of Electric Equipment Co.,Ltd.

School of Automation Science and Electrical Engineering, Beihang University

Institute of Electric Power Science, State Grid Jiangxi Electric Power Co., Ltd.

School of Information Engineering, Nanchang Hangkong University

⁰