Method of Acoustic Time Difference Logging Curve Reconstruction Based on VMD-SSA-LSTM
In order to solve the problem that some logging data is distorted or missing due to the influence of geolog-ical,engineering and other factors in the actual logging process,a reconstruction method of acoustic time difference logging curve based on VMD-SSA-LSTM(variational mode decomposition-Sparrow search algorithm-long and short time memory neural network)is constructed.Firstly,Whale algorithm is applied to optimize VMD(variation-al mode decomposition)and the original data is decomposed and denoised.Then,the improved SSA(Sparrow search algorithm)is applied to optimize the selection of LSTM(short-duration memory neural network)parame-ters.Finally,the values of each decomposition sequence are superimposed to obtain the final values.The experi-mental results show that the combined model of VMD-SSA-LSTM has higher curve reconstruction accuracy,dem-onstrate better stability and generalization ability and can generate curves that are more in line with the actual situa-tion.
reconstruction of logging curvewhale algorithmvariational mode decompositionsparrow search algo-rithmlong and short term memory neural network
国家科技期刊平台
NSTL
万方数据
