| 引用本文: | 谭风雷,陈昊,何嘉弘.基于通径分析和相似时刻的特高压变压器顶层油温预测[J].电力自动化设备,2021,41(11): |
| TAN Fenglei,CHEN Hao,HE Jiahong.Top oil temperature forecasting of UHV transformer based on path analysis and similar time[J].Electric Power Automation Equipment,2021,41(11): |
|
| 摘要: |
| 提出了一种基于通径分析和相似时刻的特高压变压器顶层油温预测方法，该方法通过动态优化相似时刻数量，来获取最优预测效果。首先基于顶层油温大小模糊排序方法量化处理时间因子后，利用通径分析方法计算各影响因素的简单相关系数并分析其与特高压变压器顶层油温的相关性。然后在基于逼近理想解排序(TOPSIS)法和时间“距离远相关性小，距离近相关性大”原则，利用气象因素相关度、时间因子相关度和负荷因子相关度线性加权得到综合因素相关度。最后详细分析了相似时刻选择和顶层油温预测的流程，并将其应用到华东地区某特高压变压器顶层油温预测算例中。结果表明所提方法的平均预测误差为1.90%，平均标准差为0.013 3，预测精度高，误差波动小，验证了该方法的有效性与可行性。 |
| 关键词: 特高压变压器 顶层油温预测 相似时刻 动态优化 模糊排序 通径分析 TOPSIS法 线性加权 |
| DOI：10.16081/j.epae.202109026 |
| 分类号:TM41 |
| 基金项目:国家自然科学基金资助项目(51807028)；国网江苏省电力有限公司科技项目(J2018014) |
|
| Top oil temperature forecasting of UHV transformer based on path analysis and similar time |
|
TAN Fenglei1, CHEN Hao1, HE Jiahong2
|
|
1.Maintenance Branch Company of State Grid Jiangsu Electric Power Co.,Ltd.,Nanjing 211102, China;2.Department of Electrical Engineering, Southeast University, Nanjing 210096, China
|
| Abstract: |
| A method of top oil temperature forecasting for UHV(Ultra High Voltage) transformer based on path analysis and similar time is proposed, which can obtain the best forecasting effect by dynamically optimizing the amount of similar time. Firstly, after quantifying the time factor based on the fuzzy ranking method of top oil temperature, the path analysis is used to calculate the simple correlation coefficients of influencing factors and analyze their correlation with top oil temperature of UHV transformer. Then based on the TOPSIS(Technique for Order Preference by Similarity to Ideal Solution) method and the time principle of “long distance with small correlation, near distance with big correlation”,comprehensive factor correlation is obtained by using the linear weighting of meteorological factor correlation, time factor correlation and load factor correlation. Finally, the process of similar time selection and top oil temperature forecasting is analyzed in detail, and it is applied to an example of top oil temperature forecasting for an UHV transformer in East China. The results show that the average error of the proposed method is 1.90% and the average standard deviation is 0.013 3, which shows high forecasting accuracy and small error fluctuation of the proposed method and verifies its feasibility and validity. |
| Key words: UHV transformer top oil temperature forecasting similar time dynamic optimization fuzzy ranking path analysis TOPSIS method linear weighting |
