基于PSO-SVM算法的电压越限成因诊断方法

Diagnosis Method for Voltage Overlimit Based on PSO-SVM Algorithm

电压越限问题会对市民生活用电、国民经济发展等产生影响,因此设定合理的电压越限诊断模型,对于制定其相应的电压调控对策有着很关键的意义。首先采用Canopy算法对K-means算法进行改进,得到电压越限成因的聚类结果。然后,输入基于PSO的支持向量机算法进行自适应训练,实现电压越限成因的在线诊断。最后通过实例仿真分析表明,改进后的算法提高了电压越限成因诊断的准确性,可满足实际的电压越限诊断要求。

The voltage overlimit problem will have a significant negative impact on the normal life of citizens and the healthy development of the national economy.Therefore,setting a reasonable voltage overrun diagnostic model is of great significance for formulating corresponding voltage regulation and control countermeasures.Firstly,the Canopy algorithm is used to improve the K-means algorithm,and the clustering results of the cause of voltage exceeding the limit are obtained.Then,the PSO based support vector machine algorithm is input for adaptive training to realize online diagnosis of the cause of voltage overlimit.Finally,through case simulation analysis,the improved algorithm improves the accuracy of the diagnosis of the cause of voltage overlimit,which can meet the actual requirements of voltage overrun diagnosis.