###
DOI:
本文二维码信息
基于多源数据融合的变电站设备健康状态多维度预警模型研究
张历, 张俊杰, 李鑫卓
(贵州电网有限责任公司电力科学研究院)
Research on Multi dimensional Early Warning Model of Substation Equipment Health Status Based on Multi source Data Fusion
zhangli, Zhang Junjie, Li Xinzhuo
(Guizhou Electric Power Research Institute of Guizhou Power Grid Co., Ltd)
摘要
相似文献
本文已被:浏览 3次   下载 0
投稿时间:2024-11-05    修订日期:2024-11-15
中文摘要: 在变电站设备健康状态预警中,由于影响设备健康状态的因素较多,使得设备健康状态的分析过程较为复杂,从而导致结果准确度较低。为了缓解这一问题,提出了基于多源数据融合的变电站设备健康状态多维度预警模型的研究。采集多维度的变电站设备数据,将重采样后的数据样本进行多源数据的融合处理,利用融合数据对变电站设备的健康状态进行评估。在此基础上,以XGBoost 算法为核心建立多维度的预警模型,并根据设置评估体系作出相应的预警处理。经过实验测试可知,该模型在实践应用中表现出了较高的准确度水平,相较于其他方法,对于设备健康状态分析的ROC曲线更为准确,有着更高的预警准确度水平,在变电站设备的实际运维管理工作中,具备良好的应用前景。
Abstract:In the health status warning of substation equipment, due to the many factors that affect the health status of equipment, the analysis process of equipment health status is complex, resulting in low accuracy of the results. To alleviate this problem, a multi-dimensional warning model for the health status of substation equipment based on multi-source data fusion has been proposed. Collect multi-dimensional substation equipment data and expand the data samples through resampling processing. Perform multi-source data fusion processing on the resampled data samples to evaluate the health status of substation equipment using the fused data. On this basis, a multidimensional warning model is established with XGBoost algorithm as the core, and corresponding warning processing is made according to the set evaluation system. Through experimental testing, it is known that the model has demonstrated a high level of accuracy in practical applications. Compared with other methods, the model has more accurate analysis of equipment health status, better ROC curve, and higher early warning accuracy, which has a good application prospect in the actual operation and maintenance management of substation equipment.
文章编号:20241105001     中图分类号:    文献标志码:
基金项目:
引用文本: