| 引用本文: | 邹 阳,俞豪奕,金 涛.融合模糊K近邻及证据理论的变压器油纸绝缘状态评估方法[J].电力系统保护与控制,2023,51(14):55-63.[点击复制] |
| ZOU Yang,YU Haoyi,JIN Tao.Evaluation method of the oil-paper insulation condition of a transformer based on fuzzy K nearest neighbor and evidence theory[J].Power System Protection and Control,2023,51(14):55-63[点击复制] |
|
| 摘要: |
| 为实现“双碳”目标，构建新型电力系统已成为电网发展的必然趋势。在此背景下，保障电力变压器的可靠运行具有重要意义。鉴于此，提出融合模糊K近邻(fuzzy K-nearest neighbor, FKNN)及证据理论的变压器油纸绝缘状态评估方法。首先，构建基于回复电压法的多特征参量数据库，并基于数据库提出证据的基本概率分配方法。而后，采用组合赋权法综合特征参量的主观权重及客观权重，同时藉由证据折扣因子对证据基本概率进行再分配，避免D-S证据理论的冲突问题。最终，对各证据进行融合推理，获得绝缘状态命题的置信水平。利用提出的方法对变压器实测数据进行验证。结果表明，绝缘状态的置信分布式结果不仅能够准确反映变压器油纸绝缘状态，也能表征出变压器油纸绝缘的劣化趋势，为电力变压器检修策略制定提供了指导。 |
| 关键词: 油纸绝缘 模糊K近邻 D-S证据理论 回复电压 状态综合评估 |
| DOI：10.19783/j.cnki.pspc.221755 |
| 投稿时间：2022-11-05修订日期：2023-03-14 |
| 基金项目:国家自然科学基金项目资助(51977039)；福建省自然科学基金项目资助(2019J01248) |
|
| Evaluation method of the oil-paper insulation condition of a transformer based on fuzzy K nearest neighbor and evidence theory |
| ZOU Yang1,2,YU Haoyi1,JIN Tao1 |
| (1. School of Electrical Engineering and Automation, Fuzhou University, Fuzhou 350108, China;
2. Key Laboratory of Fujian Universities for New Energy Equipment Testing, Putian 351100, China) |
| Abstract: |
| To achieve the goal of "carbon peaking and carbon neutrality", the construction of a new power system has become an inevitable trend of power grid development. Given this, it is of great significance to ensure the reliable operation of power transformers. Thus an evaluation method of the oil-paper insulation condition of a transformer based on fuzzy K nearest neighbor and evidence theory is proposed. First, a multi-characteristic parameter database is constructed, and the basic probability allocation method of evidence is proposed based on the database. Then, a combination weighting method is used to synthesize the subjective and objective weights of the characteristic parameters, and an evidence discount factor is used to redistribute the basic probability of evidence to avoid the conflict of D-S evidence theory. Finally, the evidence is fused and inferred from. The confidence level of the insulation state proposition is obtained. The proposed method is used to verify the measured data of the transformer, and the results show that the confidence distributed result of the insulation state not only accurately reflects the oil-paper insulation state of the transformer, but also characterizes its deterioration trend. It provides guidance for the formulation of power transformer maintenance strategy. |
| Key words: oil-paper insulation fuzzy K-nearest neighbor D-S evidence theory recovery voltage comprehensive evaluation of state |
