基于MOPSO的输变电设备状态评估管理研究

Condition Estimate and Management of Power Transmission and Transformation Equipment based on MOPSO

针对电压等级的不断提高和输变电设备使用年限的增加,发生电网故障的风险也逐渐增加的问题,提出了一种基于MOPSO的输变电设备状态评估方法。首先,训练得到一系列基SVM分类器;然后,使用MOPSO算法选择一部分精度和多样性较高的SVM构建决策分类器,并将输变电设备分为多种不同的运行状态;最后,使用变压器油中的溶解气体数据进行仿真实验来评估变压器的状态,并与传统的SVM、IEC三比值法和BP神经网络算法进行比较。实验结果表明,所提出的算法能准确、有效地评估变压器的状态,相比传统的诊断方法其具有更高的准确率。

In order to solve the problem that the risk of power grid failure gradually increases due to continuous improvement of the voltage level and increase of the service life of the power transmission and transformation equipment,this paper proposes a method of condition estimate for the power transmission and transformation equipment based on MOPSO.First,a series of SVM classifiers is obtained through training.Second,the MOPSO algorithm is used to select a part of SVM with high precision and diversity to construct the decision classifier,and then classify the power transmission and distribution equipment into different running states. Finally,the proposed algorithm is used to estimate the state of power transformer,and compared with the traditional SVM,IEC method and BP neural network algorithms. The experimental results show that the proposed algorithm can accurately and effectively evaluate the transformer state,and its diagnostic accuracy is higher than the traditional diagnostic method.