A training image evaluation and selection method based on minimum data event distance for multiple-point geostatistics

Feng, Wenjie; Wu, Shenghe; Yin, Yanshu; Zhang, Jiajia; Zhang, Ke

doi:10.1016/j.cageo.2017.04.004

Cited by 29 publications

(14 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…(2014) and also used in the work of Feng et al. (2017). In this setup, there are three training images with different features: ellipsoids, sine waves, and vertical stripes (Figure 4).…”

Section: Case Studiesmentioning

confidence: 99%

“…The methods of Pérez et al. (2014) and Feng et al (2017) are the only ones allowing to identify the training image. However, they assume that the simulation is stationary, while this is rarely the case for practical applications in which there are different trends, for example, in orientations, proportions of the facies, or even types of patterns.…”

Section: Introductionmentioning

confidence: 99%

“…Feng et al. (2017) used a minimal distance between the data event found in the TI and in the conditioning data. Based on any of these metrics, one can derive automated parameter selection methods (Baninajar et al., 2019; Dagasan et al., 2018).…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Framework for the Cross‐Validation of Categorical Geostatistical Simulations

Juda

Renard

Straubhaar

2020

Earth and Space Science

View full text Add to dashboard Cite

The mapping of subsurface parameters and the quantification of spatial uncertainty requires selecting adequate models and their parameters. Cross‐validation techniques have been widely used for geostatistical model selection for continuous variables, but the situation is different for categorical variables. In these cases, cross‐validation is seldom applied, and there is no clear consensus on which method to employ. Therefore, this paper proposes a systematic framework for the cross‐validation of geostatistical simulations of categorical variables such as geological facies. The method is based on K‐fold cross‐validation combined with a proper scoring rule. It can be applied whenever an observation data set is available. At each cross‐validation iteration, the training set becomes conditioning data for the tested geostatistical model, and the ensemble of simulations is compared to true values. The proposed framework is generic. Its application is illustrated with two examples using multiple‐point statistics simulations. In the first test case, the aim is to identify a training image from a given data set. In the second test case, the aim is to identify the parameters in a situation including nonstationarity for a coastal alluvial aquifer in the south of France. Cross‐validation scores are used as metrics of model performance and quadratic scoring rule, zero‐one score, and balanced linear score are compared. The study shows that the proposed fivefold stratified cross‐validation with the quadratic scoring rule allows ranking the geostatistical models and helps to identify the proper parameters.

show abstract

“…(2014) and also used in the work of Feng et al. (2017). In this setup, there are three training images with different features: ellipsoids, sine waves, and vertical stripes (Figure 4).…”

Section: Case Studiesmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

A Framework for the Cross‐Validation of Categorical Geostatistical Simulations

Juda

Renard

Straubhaar

2020

Earth and Space Science

View full text Add to dashboard Cite

show abstract

“…The maximum search range in the test was set to 31 × 31 × 1, and the number of upper limit condition points was to 35. The absolute compatibility and the relative compatibility were calculated respectively for the number of repetitions when searching for 5,10,15,20,25,30,35 condition points within the search range ( Figure 6). It can be seen that as the condition points increased, the relative compatibility of the training images close to the original geological model tended to increase, while the absolute compatibility was higher than that of other training images.…”

Section: Two-dimensional Testmentioning

confidence: 99%

“…It has become a problem that modelers have to face. Yet, the optimal selection methods for training images are very limited, which include the optimal selection method based on variogram, the method based on conditional probability [3] [6] [19], and the method based on similar distance [20].…”

mentioning

confidence: 99%

A New Method to Select Training Images in Multi-Point Geostatistics

Wang¹,

Yin²,

Feng³

2018

OJOGas

Self Cite

View full text Add to dashboard Cite

Training images, as an important modeling parameter in the multi-point geostatistics, directly determine the effect of modeling. It's necessary to evaluate and select the candidate training image before using the multi-point geostatistical modeling. The overall repetition probability is not sufficient to describe the relationship of single data events in the training image. Based on the understanding, a new method was presented in this paper to select the training image. As is shown in the basic idea, the repetition probability distribution of a single data event was used to characterize the type and stationarity of the sedimentary pattern in the training image. The repetition probability mean value and deviation of single data event reflected the stationarity of the geological model of the training image; the rate of data event mismatching reflected the diversity of geological patterns in training images. The selection of optimal training image was achieved by combining the probability of repeated events and the probability of overall repetition of single data events. It's illustrated in the simulation tests that a good training image has the advantages of high repetition probability compatibility, stable distribution of repeated probability of single data event, low probability mean value, low probability deviation and low rate of mismatching. The method can quickly select the training image and provide the basic guarantee for multi-point geostatistical simulations.

show abstract