| 引用本文: | 曾勇斌,王星华,彭显刚,等.输电线路缺陷风险建模及其预测方法研究[J].电力系统保护与控制,2020,48(10):91-98.[点击复制] |
| ZENG Yongbin,WANG Xinghua,PENG Xiangang,et al.Research on risk modeling and forecasting method of transmission line defects[J].Power System Protection and Control,2020,48(10):91-98[点击复制] |
|
| 摘要: |
| 输电线路缺陷发展状况的预测评估工作对安排检修和制定运维策略有着指导性意义。针对目前输电线路整体缺陷状况的评价指标较为模糊和单一、不同内外环境下输电线路缺陷发展状况不一致的问题，提出了输电线路缺陷风险建模及其预测的方法。该方法首先根据输电线路自身的特点将其细分成若干个部件。然后对输电线路各部件的缺陷严重程度进行量化，并根据输电线路的缺陷历史数据，通过隶属度分析进而定义得到输电线路整体的缺陷风险值。最后研究了各种随机因素分别对线路各部件缺陷风险值的影响。建立基于支持向量机的输电线路缺陷风险值预测模型并对线路未来时段内的缺陷风险值进行预测。通过实例证明了该方法对输电线路缺陷风险值预测的可行性。 |
| 关键词: 输电线路 缺陷风险值 相关性分析 影响因素 支持向量机 |
| DOI：10.19783/j.cnki.pspc.190935 |
| 投稿时间：2019-08-02修订日期：2019-11-23 |
| 基金项目:国家自然科学基金项目资助(61903091) |
|
| Research on risk modeling and forecasting method of transmission line defects |
| ZENG Yongbin,WANG Xinghua,PENG Xiangang,HUANG Jinglin,JIAN Shengchao,LU Di |
| (School of Automation, Guangdong University of Technology, Guangzhou 510006, China) |
| Abstract: |
| The forecasting and evaluation of the development of transmission line defects has guiding significance for arranging maintenance and formulating operation and maintenance strategies. Considering the current fuzzy and single evaluation indicators of the overall defect status of transmission lines, and the inconsistent development of transmission line defects in different internal and external environments, a transmission line defect risk modeling and prediction method is proposed. The method first subdivides the transmission line into several parts according to the characteristics of the transmission line itself; then quantifies the defect severity of each part of the transmission line. The defect risk value of the whole transmission line is defined by a membership degree analysis based on the defect history data of the transmission line. Finally, the effects of various random factors on the risk value of each component of the line are studied. The support vector machine-based transmission line defect risk value prediction model is established and the defect risk value in the future time of the line is predicted. The feasibility of the method for predicting the risk value of transmission line defects is proved by an example. This work is supported by National Natural Science Foundation of China (No. 61903091). |
| Key words: transmission line defect risk value correlation analysis influencing factor support vector machine |
