基于多参数自适应VMD的GNSS形变监测序列分解

Decomposition of GNSS deformation monitoring sequences based on multi-parameter adaptive VMD

针对结构健康监测场景下,全球导航卫星系统(GNSS)形变监测序列中各类特征相互混叠,难以进行特征提取与独立分析的问题,提出一种基于多参数自适应变分模态分解(MA-VMD)的时间序列分解算法。首先对变分模态分解(VMD)算法中多项参数对分解结果的影响进行了综合分析;然后从原始序列以及分解结果的频域特性出发,自适应调整分解模态数、惩罚因子、初始中心频率及拉格朗日乘子四组参数,建立MA-VMD算法。仿真序列实验表明,MA-VMD算法的序列分解结果与真实值之间的互相关系数为98.77%、均方根误差为0.1365 mm,均接近全局最优,并显著优于经验模态分解、奇异谱分析、改进变分模态分解等算法。最后基于实测GNSS变形监测数据验证了所提算法在工程应用上的有效性。

Aiming at the problem that various types of features in global navigation satellite system(GNSS)deformation monitoring sequences are intermingled with each other and difficult to extract features and analyze them independently in the structural health monitoring scenario,a time series decomposition algorithm based on multi-parameter adaptive variational mode decomposition(MA-VMD)is proposed.Firstly,the effects of multiple parameters in the variational mode decomposition(VMD)algorithm on the decomposition results are analyzed thoroughly.Then,according to the frequency domain characteristics of the original sequence and decomposition results,the MA-VMD algorithm is established by adaptively adjusting the decomposition mode number,penalty factor,initial center frequency and Lagrange multiplier.Experiments on the simulated sequences show that the correlation between the decomposition results of MA-VMD algorithm and real values is 98.77%,and the root mean square error is 0.1365 mm,which is close to the global optimum and significantly better than other decomposition algorithms like empirical mode decomposition,singular spectrum analysis and improved variational mode decomposition.Finally,based on the actual GNSS deformation monitoring data,the effectiveness of the proposed algorithm in engineering application is verified.