基于BOTDR监测数据的光纤复合海底电缆状态预测

Prediction of the state of optical fiber composite submarine cable based on BOTDR monitoring data

为了判断光纤复合海底电缆状态的发展趋势,及时发出故障预警信号,提出了基于加权最小二乘支持向量机(WLS-SVM)的海缆BOTDR监测数据多步预测模型。利用Birge-Massart策略计算实测数据小波分解后不同尺度上的阈值,使用软阈值法消除随机噪声对预测准确性的影响;在混沌序列分析的基础上,采用G-P算法进行相空间重构,确定最佳嵌入维数,同时验证频移时间序列的混沌特性;将重构相空间中的相点馈入到WLSSVM模型完成递归多步预测;最后对海缆两个典型位置处测点进行了频移6步预测。结果表明,递归6步预测的最大平均相对误差为1.80%,具有比标准支持向量机预测结果更高的预测精度和更好的适用性。

In order to judge the development trend of the state of optical fiber composite submarine cable and timely send the fault warning signals,the multi step forecast model of the submarine cable BOTDR monitoring data based on weighted least squares support vector machine（ WLS-SVM） is proposed in this paper. Birge-Massart method is used to calculate the threshold on different scales which is derived from wavelet decomposition of the measured data. And the soft threshold method is used to eliminate the influence of random noise on the accuracy of prediction. The phase space is reconstructed using G-P algorithm based on chaotic sequence analysis,then the optimal embedding dimension can be determined and the chaotic characteristic of time series of frequency shift is verified. The phase point of reconstructed phase space is fed to the WLS-SVM model to proceed recursive multi-step prediction. Finally,6 step prediction of frequency shift in two typical position of cable is carried out. The results show that the maximum average relative error of recursive 6 step prediction is 1. 80%,which has higher prediction accuracy and better applicability compared with the standard support vector machine method.