融合改进教与学优化算法的接地网腐蚀检测
Grounding network corrosion detection based on improved teaching-learning optimization algorithm
摘要
[目的]接地网作为电力系统的关键安全设施,长期埋设于复杂土壤环境中,易发生腐蚀,导致导体截面积减小、泄流能力下降,威胁电网稳定运行.[方法]针对传统检测方法(如电网络分析法、电化学分析法)需停电操作、流程复杂或无法精确定位的问题,该文提出一种基于改进教与学优化算法的瞬变电磁法接地网腐蚀检测技术.首先基于频率域磁场信号特征构建瞬变电磁一维正演模型,随后提出一种改进的教与学优化算法用于瞬变电磁反演,进一步对标准算法进行三方面改进:引入 Hammersley序列和改进反向学习策略提升初始种群多样性与质量;设计基于优势学科保留的更新机制增强局部搜索能力;采用自适应t分布变异策略平衡全局与局部搜索精度.[结果]通过四层地电模型和低阻异常体模型仿真验证,改进算法在浅层电阻率反演中误差满足反演精度要求.[结论]实验结果表明,该方法可准确定位腐蚀断裂位置并量化腐蚀程度,为接地网不开挖、高精度腐蚀检测提供有效解决方案.
[Objective]As a key safety facility of the power system,the grounding network is easy to corrode when buried in the complex soil environment for a long time,which leads to the decrease of conductor cross-sectional area and current discharge capacity,and threatens the stable operation of the power grid.[Methods]In view of the traditional detection methods(such as electrical network analysis,electrochemical analysis)need to power outage operation,process complexity or can not be accurately located,this paper proposes a transient electromagnetic method based on improved teaching and learning optimization algorithms grounding network corrosion detection technology.Firstly,a transient electromagnetic one-dimensional orthogonal model is constructed based on the magnetic field signal characteristics in the frequency domain,and then an improved teaching and learning optimization algorithm is proposed for the transient electromagnetic inversion,and the standard algorithm is further improved in three aspects:introducing the Hammersley sequence and improving the inverse learning strategy to enhance the diversity and quality of the initial populations;designing the updating mechanism based on the retention of dominant disciplines to enhance the local searching ability;and adopting the adaptive t-distribution mutation strategy to balance the local searching ability.[Results]The adaptive t-distribution variation strategy is used to balance the global and local search accuracy.Through the simulation of the four-layer geoelectric model and low-resistance anomaly model,it is verified that the improved algorithm can meet the requirements of inversion accuracy in shallow resistivity inversion.[Conclusion]The experimental results show that the method can accurately locate the corrosion fracture position and quantify the corrosion degree,which provides an effective solution for high-precision corrosion detection without excavation of the grounding network.
年,卷(期):
2026, 52(9)
分类号:
TB9
TM83
机标分类号:
TP301.6
TP183
TM731
在线出版日期:
2026年10月14日
基金项目:
GXKJXM20240121:广西电网科技项目,