摘要
针对低压交流系统的短路故障诊断问题,提出一种基于黑洞粒子群优化算法(BHPSO)和多层级SVM的低压交流系统短路故障类型辨识方法.首先,对故障前后0.5 ms电流信号进行小波变换分解,采用小波细节分量标准差构建故障特征向量.其次,以黑洞粒子群算法对SVM的核参数和惩罚因子进行参数优化来构建多层级SVM分类器,实现低压交流系统短路故障类型辨识.最后,采用TMS320F28335 DSP硬件化技术实现故障类型辨识决策模型,通过低压交流系统短路实验证实本方法准确率高,且在噪声干扰、负荷电流变化等工况下均有较好的鲁棒性.
Aiming at the problem of short-circuit fault diagnosis in low-voltage AC systems,this paper proposes a short-circuit fault type identification method for low-voltage AC systems based on black hole particle swarm optimization(BHPSO)and multi-level SVM.First,based on the wavelet transform decomposition of the 0.5 ms current signal before and after the fault,the standard deviation of the wavelet detail components is used to construct the fault feature vector.Secondly,the black hole particle swarm algorithm is used to optimize the parameters of kernel parameter and penalty factor of the SVM to construct a multi-level SVM classifier to realize the short-circuit fault type identification of the low-voltage AC system.Finally,the decision model of fault type identification is implemented by hardware technology based on TMS320F28335 DSP.The high accuracy of the method is confirmed by the short circuit experiment of the low voltage AC system.And it has better robustness under working conditions such as noise interference and load current changes.
作者
郭冰云
黄炳华
张美锋
陈荣全
缪希仁
李文院
GUO Bingyun;HUANG Binghua;ZHANG Meifeng;CHEN Rongquan;MIAO Xiren;LI Wenyuan(Fujian Huadian Electric Power Engineering Co.,Ltd.,Fuzhou,Fujian 350012,China;Fujian Huadian Kemen Power Generation Co.,Ltd.,Fuzhou,Fujian 350012,China;College of Electrical Engineering and Automation,Fuzhou University,Fuzhou,Fujian 350108,China)
出处
《福州大学学报（自然科学版）》
CAS
北大核心
2022年第5期627-634,共8页
Journal of Fuzhou University(Natural Science Edition)
基金
福建省高校产学合作资助项目(2019H6009)。
关键词
低压交流系统
短路故障类型辨识
黑洞粒子群
多层级SVM
low voltage AC system
short circuit fault type identification
black hole particle swarm
multilevel SVM
