一二次融合的智能集成开关在配电网上应用的研究

Research on the Application of Primary and Secondary Fusion Intelligent Integrated Switch in Distribution Network

针对传统技术中一二次融合智能集成开关故障诊断不灵敏,应用效率不高,提出了新的应用方案。构建了EMD-ICA融合算法和BP神经网络算法模型,通过EMD-ICA算法,快速、准确地提取一二次融合智能集成开关故障信息的输出基函数,从混合信号中提取时域、频域和能量特征参数,将一二次融合智能集成开关故障信息数字化表示,便于用户更直观地分析故障信息、运行情况等。利用BP神经网络计算模型,采用反向传播算法将误差数据反向传播,以获取更高的学习效率。通过一二次融合智能集成开关的失效概率评价模型,及时、有效、全面地衡量系统是否正常运行。试验表明,这一研究方法引用误差低于±0.1%,对比误差在2%以下,总误差精度不超过5%。

Aiming at the insensitive fault diagnosis of the primary and secondary fusion intelligent integrated switches in the traditional technology and the low application efficiency,a new application scheme is proposed.We construct the EMD-ICA fusion algorithm and the BP neural network algorithm model.Through the EMD-ICA algorithm,the output basis function of the primary and secondary fusion intelligent integrated switch fault information is quickly and accurately extracted,and the time domain and frequency domain parameters are extracted from the mixed signal.And energy characteristic parameters,the digital representation of the fault information are obtained from integrated once and twice switches,so that users can be convenient to analyze the fault information and operating conditions intuitively.Using the BP neural network calculation model,the error back propagation algorithm is used to obtain higher learning efficiency.Through the primary and secondary integration of the failure probability evaluation model of the intelligent integrated switch,it can timely,effectively and comprehensively measure whether the system operates normally.Experiments show that the citation error of the method proposed in this paper is less than±0.1%,the comparison error is less than 2%,and the total error accuracy does not exceed 5%.