Inversion of Rayleigh wave dispersion curves via adaptive GA and nested DLS

Lei, Yuhang; Hu, Shen; Li, Xinxin; Wang, Xin; Li, Qingchun

doi:10.1093/gji/ggz171

Cited by 22 publications

(11 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Dispersion curve inversion is the core content and key step of the MASW technology. Traditional localsearching linear inversion methods, such as the damped least squares method, are highly dependent on the selection of an initial model and the calculation of partial derivatives, so it is easy to converge to the local minimum (Lei et al, 2018(Lei et al, , 2019Moro et al, 2007a;Xia et al, 2002aXia et al, , 2002b. In fact, Rayleigh wave dispersion curve inversion is a typical non-linear, multi-parameter and multi-extremum inversion problem.…”

Section: Introductionmentioning

confidence: 99%

“…However, traditional global-searching non-linear inversion algorithms still have some defects, such as the failure to achieve an effective balance between global search and local exploitation, and being prone to premature convergence and falling into local optimal solutions. As a result, solution accuracy, operational efficiency and inversion stability are difficult to achieve (Lei et al, 2018(Lei et al, , 2019Wang et al, 2021). In addition, when there are high-velocity or low-velocity interlayers in a stratum, in addition to fundamental mode Rayleigh waves, highorder mode Rayleigh waves are usually developed in the seismic records.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Multi‐mode non‐linear inversion of Rayleigh wave dispersion curves with grey wolf optimization and cuckoo search algorithm

Che,

Shen,

et al. 2024

Near Surface Geophysics

Self Cite

View full text Add to dashboard Cite

Dispersion curve inversion is one of the core contents of Rayleigh wave data processing. However, the dispersion curve inversion has the characteristics of multi‐parameter, multi‐extremum as well as nonlinearity. In the face of Rayleigh wave data processing under complex seismic‐geological conditions, it is difficult to reconstruct an underground structure quickly and accurately apply a single global‐searching non‐linear inversion algorithm. For this reason, we proposed a strategy to invert multi‐order mode Rayleigh wave dispersion curves by combining with grey wolf optimization (GWO) and cuckoo search (CS) algorithms. On the basis of introducing the mechanism of iterative chaotic map with infinite collapses (ICMIC) and the strategy of dimension learning–based hunting (DLH), an improved GWO was developed that was called IDGWO (ICMIC and DLH GWO). After searching the near‐optimal region through IDGWO, the CS with a variable step‐size Lévy flight search mechanism was switched adaptively to complete the final inversion. The correctness of our method was verified by the multi‐order mode dispersion curve inversion of a six‐layer high‐velocity interlayer model. Then it was further applied to the processing of real seismic datasets. The research results show that our method fully utilizes the advantages of each of the two global‐searching non‐linear algorithms after integrating IDGWO and CS, while effectively balancing the ability between global search and local exploitation, further improving the convergence speed and inversion accuracy and having good anti‐noise performance.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Multi‐mode non‐linear inversion of Rayleigh wave dispersion curves with grey wolf optimization and cuckoo search algorithm

Che,

Shen,

et al. 2024

Near Surface Geophysics

Self Cite

View full text Add to dashboard Cite

show abstract

“…For population-based inversion algorithms, many populations of multilayer forward modeling and many generations of iterations are essential to search the optimal site parameters (shear wave velocity, overburden thickness, etc.). It is difficult to model site structures that have many layers and apply population-based inversion algorithms (Picozzi and Albarello, 2007;Lu et al, 2016;Poormirzaee, 2016;Lei et al, 2019;Poormirzaee and Fister, 2021). Therefore, it is necessary to develop a fast, accurate, stable dispersion curve construction method.…”

Section: Introductionmentioning

confidence: 99%

Rapid construction of Rayleigh wave dispersion curve based on deep learning

2023

View full text Add to dashboard Cite

Introduction: The dispersion curve of the Rayleigh-wave phase velocity (VR) is widely utilized to determine site shear-wave velocity (Vs) structures from a distance of a few metres to hundreds of metres, even on a ten-kilometre crustal scale. However, the traditional theoretical-analytical methods for calculating VRs of a wide frequency range are time-consuming because numerous extensive matrix multiplications, transfer matrix iterations and the root searching of the secular dispersion equation are involved. It is very difficult to model site structures with many layers and apply them to a population-based inversion algorithm for which many populations of multilayers forward modelling and many generations of iterations are essential.Method: In this study, we propose a deep learning method for constructing the VR dispersion curve in a horizontally layered site with great efficiency. A deep neural network (DNN) based on the fully connected dense neural network is designed and trained to directly learn the relationships between Vs structures and dispersion curves. First, the training and validation sets are generated randomly according to a truncated Gaussian distribution, in which the mean and variance of the Vs models are statistically analysed from different regions’ empirical relationships between soil Vs and its depth. To be the supervised dataset, the corresponding VRs are calculated by the generalized reflection-transmission (R/T) coefficient method. Then, the Bayesian optimization (BO) is designed and trained to seek the optimal architecture of the deep neural network, such as the number of neurons and hidden layers and their combinations. Once the network is trained, the dispersion curve of VR can be constructed instantaneously without building and solving the secular equation.Results and Discussion: The results show that the DNN-BO achieves a coefficient of determination (R2) and MAE for the training and validation sets of 0.98 and 8.30 and 0.97 and 8.94, respectively, which suggests that the rapid method has satisfactory generalizability and stability. The DNN-BO method accelerates the dispersion curve calculation by at least 400 times, and there is almost no increase in computation expense with an increase in soil layers.

show abstract

“…The general linear inversion method's effect depends on the early model's choice and the accuracy of the partial derivative matrix. Some nonlinear and global optimization algorithms include particle swarms [4], genetics [5], simulated annealing [6], and multiple global optimization methods [7]. However, they all have problems such as slow convergence speed and low accuracy.…”

Section: Introductionmentioning

confidence: 99%

Application of Deep Learning in Rayleigh Dispersion Curves Inversion

Ye,

Xie,

Zhang

et al. 2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

The inversion of the Rayleigh wave dispersion curve is a fundamental step in determining the shear wave velocity profile. However, existing inversion methods, such as linear and nonlinear approaches, have inherent limitations. In order to overcome these challenges and enhance the speed and accuracy of the inversion, this paper proposes a novel deep learning-based approach. We establish a method to construct sample data that ensures ergodicity and evolutive orderliness of the near-surface velocity profile using a constrained Markov decision process to address these limitations. We then design a deep learning model to capture the nonlinear mapping relationship between the dispersion sequence data and velocity structure. Training the model on the generated sample data accurately predicts the subsurface velocity profile based on the dispersion curve. The proposed method is evaluated through theoretical model experiments, demonstrating high computational accuracy.

show abstract

Inversion of Rayleigh wave dispersion curves via adaptive GA and nested DLS

Cited by 22 publications

References 30 publications

Multi‐mode non‐linear inversion of Rayleigh wave dispersion curves with grey wolf optimization and cuckoo search algorithm

Multi‐mode non‐linear inversion of Rayleigh wave dispersion curves with grey wolf optimization and cuckoo search algorithm

Rapid construction of Rayleigh wave dispersion curve based on deep learning

Application of Deep Learning in Rayleigh Dispersion Curves Inversion

Contact Info

Product

Resources

About