光纤激光网络故障大数据自动分类方法

Automatic classification method of fiber laser network fault big data

光纤激光网络通信受噪声影响,导分类效果较差,提出了光纤激光网络故障大数据自动分类方法。对数据进行噪处理,将源域中的样本和目标领域的样本配对处理,采用非线性变换提供故障特征,通过K-means算法和最近邻算法分析数据扰动性,将、征输入到机器学习分类器中分类处理,最终完成故障大数据自动分类处理。经测试证明,所提方法对于不同类型的故障大数据分类时间低于20 s,并且分类正确率、召回率以及F值高于80%、90%和95%,可以快速准确完成故障大数据自动分类处理。

Fiber laser network communication is affected by noisy data,and the effect of guide classification is poor.An automatic classification method for fiber laser network fault big data is proposed.The noisy data is processed,the samples in the source domain and the samples in the target domain are paired and processed,the nonlinear transformation is used to provide fault features,the data disturbance is analyzed by K-means algorithm and nearest neighbor algorithm,and the input and levy are classified and processed into the machine learning classifier,and finally the automatic classification processing of fault big data is completed.The test proves that the classification time of the proposed method for different types of fault big data is less than 20 s,and the classification accuracy,recall rate and F value are higher than 80%,90%and 95%,which can quickly and accurately complete the automatic classification of fault big data.