基于改进VMD的岩石破裂声发射信号滤波算法

The Filtering Algorithm for Rock Fracture Signal Based on Improved VMD

在岩体损伤演化的过程中会释放出带有大量信息的声发射(Acoustic Emission,AE)信号,针对AE信号的非平稳性和非线性等特点以及在采用变分模态分解(Variational Mode Decomposition,VMD)算法对岩石AE信号去噪时分解参数K和惩罚因子α难以确定的问题,将VMD算法应用于岩石破裂AE信号滤波中,对岩体AE信号进行预处理,利用Tent混沌优化的麻雀搜索算法(Sparrow Search Algorithm,SSA)对VMD算法的参数K和α进行优化搜索,其适应度函数为包络熵最小值。之后根据搜索结果来确定VMD的分解参数K和惩罚因子α,并利用优化后的VMD对岩石破裂AE信号进行滤波。经仿真和实验分析可知,Tent-SSA的优化速度和优化精度较为优越,Tent-SSA-VMD的滤波效果优于常规滤波算法。

In the process of rock damage evolution,large amount of acoustic emission(AE)signals will be released.However,the AE signal is nonlinear and non-stationary,and using the variational mode decomposition(VMD)algorithm is hardly to determine the decomposition parameter K and penalty factorαwhen denoising the rock acoustic emission signal.In order to solve the problem,in this paper,VMD algorithm is applied to the AE signal filtering of rock fracture as the preprocessing of the rock mass AE signal.The sparrow search algorithm(SSA)of Tent chaotic optimization is used to optimize the searching process of the parameter K and the penalty factorαof the VMD algorithm with the fitness function being the minimum envelope entropy.After that,the decomposition parameter K and penalty factorαof VMD are determined.The optimized VMD is used to denoise the AE signal of rock fracture.The simulation and experimental analysis show that the optimization speed and accuracy of the Tent-SSA algorithm are superior,and the filtering effect of Tent-SSAVMD is better than conventional filtering algorithm.