| 摘要: |
| 基于回复电压测试数据，运用粗糙集理论评估变压器绝缘状态，并建立变压器绝缘状态评估系统。首先选取回复电压法特征量和扩展德拜等效弛豫电路特征量作为评估特征量。其次利用模糊C均值聚类算法对评估特征量进行模糊划分并指标化，生成绝缘状态决策表。接着利用差别矩阵对绝缘状态决策表进行属性约简。最后，利用盲目删除值约简算法对已约简决策表进行属性值约简，得到完全约简的决策表，并在此基础上建立变压器绝缘状态评估的规则表。通过实例分析证明提出的评估方法能够有效地评估变压器内部绝缘状态。 |
| 关键词: 粗糙集 回复电压法 模糊C均值聚类 差别矩阵 盲目删除值约简算法 绝缘状态评估 |
| DOI：10.7667/PSPC180504 |
| 投稿时间：2018-05-02修订日期：2018-06-07 |
| 基金项目:国家自然科学基金项目资助(61174117) |
|
| Evaluation of transformer oil-paper insulation based on rough set theory |
| LIN Xiaoning,CAI Jinding |
| (College of Electrical Engineering and Automation, Fuzhou University, Fuzhou 350108, China) |
| Abstract: |
| Based on the recovery voltage test datum, the rough set theory is used to evaluate the insulation status of the transformer and a transformer insulation status evaluation system is established. Firstly, the characteristics of the recovered voltage method and the extended Debye equivalent relaxation circuit are selected as the evaluation features. Then, the fuzzy C-means clustering algorithm is used to divide the quantities of evaluation features into fuzzy indicators and generate the insulation status decision table. The discrepancy matrix performs attribute reduction on the insulation status decision table. Finally, the blind reduction value reduction algorithm is used to reduce the attribute value of the reduced decision table, and a complete reduction decision table is obtained. On this basis, the transformer insulation status evaluation rule table is established. The example analysis proves that the proposed evaluation method can effectively evaluate the internal insulation status of the transformer. This work is supported by National Natural Science Foundation of China (No. 61174117). |
| Key words: rough set recovery voltage method fuzzy C-means clustering discrepancy matrix blind deletion value reduction algorithm insulation status assessment |
