基于主元分析算法的电费抄核收数据异常诊断方法

Abnormal Diagnosis Method of Electricity Charges Based on Principal Component Analysis Algorithm

由于当前已有方法未能考虑电费抄核收数据去噪问题,导致电费抄核收数据异常诊断准确性下降,诊断延时增加。为此,提出一种基于主元分析算法的电费抄核收数据异常诊断方法。将采集的电费抄核收数据作为研究样本数据,通过小波变换对其进行去噪预处理。以此为基础,采用相关系数矩阵求解数据主成分,确定数据贡献率的主元个数,同时利用主元分析表达式建立数据异常诊断模型,通过模型实现电费抄核收数据异常诊断。实验结果表明,通过对数据进行去噪处理,促使所提方法电费抄核收数据异常诊断结果准确性得到有效提升,诊断延时得到明显降低。

Because the current existing methods fail to consider the problem of denoising the data collected by the electricity bill, the accuracy of the diagnosis of abnormal data collected by the electricity bill decreases, and the diagnosis delay increases. To this end, a method for diagnosing abnormalities in the data collected by electricity bills based on principal component analysis algorithm is proposed. Take the collected electricity bill check data as the research sample data, and perform denoising preprocessing on it through wavelet transform. Based on this, the correlation coefficient matrix is used to solve the principal components of the data,and the number of principal components of the data contribution rate is determined. At the same time, the principal component analysis expression is used to establish a data abnormality diagnosis model, and the abnormal diagnosis of electricity bill checking data is realized through the model. The experimental results show that by denoising the data, the accuracy of the abnormal diagnosis result of the electricity bill checking data collected by the proposed method is effectively improved, and the diagnosis delay is significantly reduced.