Author Name | Affiliation | Li Bian | College of Electronic and Information Engineering, Guangdong Ocean University, Zhanjiang 524088,Guangdong, China | Hui He | College of Electrical and Control Engineering, Heilongjiang University of Science and Technology, Harbin 150022, China | Hongna Sun | College of Electrical and Control Engineering, Heilongjiang University of Science and Technology, Harbin 150022, China | Wenjing Liu | Handan Power Supply Company, Handan 056002, Hebei,China |
|
Abstract: |
The original fault data of oil immersed transformer often contains a large number of unnecessary attributes, which greatly increases the elapsed time of the algorithm and reduces the classification accuracy, leading to the rise of the diagnosis error rate. Therefore, in order to obtain high quality oil immersed transformer fault attribute data sets, an improved imperialist competitive algorithm was proposed to optimize the rough set to discretize the original fault data set and the attribute reduction. The feasibility of the proposed algorithm was verified by experiments and compared with other intelligent algorithms. Results show that the algorithm was stable at the 27th iteration with a reduction rate of 56.25% and a reduction accuracy of 98%. By using BP neural network to classify the reduction results, the accuracy was 86.25%, and the overall effect was better than those of the original data and other algorithms. Hence, the proposed method is effective for fault attribute reduction of oil immersed transformer. |
Key words: transformer fault improved imperialist competitive algorithm rough set attribute reduction BP neural network |
DOI:10.11916/j.issn.1005-9113.19030 |
Clc Number:TM407 |
Fund: |
|
Descriptions in Chinese: |
基于改进型帝国竞争算法的变压器故障属性约简 边莉1,何辉2,孙洪娜2,刘文静3 (1.广东海洋大学 电子与信息工程学院,广东 湛江 524088;2.黑龙江科技大学 电气与控制工程学院,哈尔滨 150022;3.国网邯郸供电公司,河北 邯郸 056002) 创新点说明:通过利用改进型帝国竞争算法与粗糙集和神经网络相结合的方式,对油浸式变压器的故障数据集进行了优化,并验证了该方法具有较好的性能。 研究目的: 电力系统的运行与人们的工作生活和工业生产有着非常密切的联系,其安全稳定的运行具有重要的意义,而油浸式变压器作为电网运行的重要组成部分,它的安全可靠运行对整个电力系统具有重要的影响。由于原始的变压器故障数据具有相当大的冗余,这就对进一步的故障判断增加了难度,其结果就是运行速度慢,且诊断正确率较低。所以本文针对原始数据集过于繁杂的问题,优化原始数据,避免大量无意义的计算,并提高其准确率。 研究方法: 利用改进型帝国竞争算法对粗糙集进行优化后,对离散化的油浸式变压器原始故障数据进行属性约简,得到最终决策表,为了验证该方法得到的决策表是否具有优越性,将其带入神经网络中进行验证,并与遗传算法、粒子群算法和模拟退火算法进行了对比。 结果: 改进型帝国竞争算法在第27次迭代时已经趋于稳定,约简率为56.25%,约简精度为98%。采用BP神经网络对故障数据集最终决策表进行诊断验证,准确率为86.25%,总体效果优于原始数据和其他算法。因此,对油浸式变压器的故障属性进行约简是非常有效的。 结论: 1) 本文将改进型帝国竞争算法应用到油浸式变压器故障属性约简问题中,通过对算法进行介绍,并建模、仿真,最后与其他智能算法进行比较分析,得出该算法具有较强的可行性和适用性。 2)该方法实现了变压器故障属性的最优约简。与遗传算法、粒子群算法和模拟退火算法相比,帝国竞争算法优化粗糙集属性约简具有迭代次数少、约简率高、精度高等优点,降低了对数据存储的要求,提高了分类精度。 3)在保持判别关系不变的前提下,去掉一些无意义的属性,可大大降低后续操作的难度。当样本集数据量较大时,宜采用改进的帝国竞争算法优化粗糙集的方法来进行属性约简。 关键词:变压器故障;改进型帝国竞争算法;粗糙集;属性约简;BP神经网络 |