Multidomain petrophysically constrained inversion and geology differentiation using guided fuzzy c-means clustering

Abstract

Geophysical inversion methods are used as part of an interpretation process that seeks to differentiate geologic units. To improve the reliability of geologic differentiation based on recovered images from geophysical inversions, we have developed a multidomain clustering inversion algorithm that can incorporate statistical petrophysical data into a deterministic geophysical inversion framework through the use of the fuzzy c-means clustering technique. Petrophysical data are treated in the parameter domain in the same way that geophysical data are treated in the spatial domain, and these two different types of data are simultaneously inverted in their respective domains through the minimization of a single common objective function. The resulting physical property model honors the geophysical and petrophysical data and therefore can represent the earth better than geophysical inversion models that solely honor the geophysical data. Geophysical inversion and geology differentiation are generally treated as two independent procedures and applied in a serial manner. In our inversion method, we integrate these two components into a unified scheme, within which they interact with each other and are mutually enhanced. We produce a geologic map showing the distribution of geologic units as a direct and integral part of the geophysical inversion. We tested the algorithm using two synthetic examples and a field data example.

Heuristic Clustering Based on Centroid Learning and Cognitive Feature Capturing

“…If is extremely small, much more clusters will be obtained because of emphasis on detailed structures of data set, while extremely large will form only one big cluster because of ignoring of details. Extensive experiments show that ∈ [5,30] for medium data sets, and is set as 2% of the data set scale for large data sets. Other two parameters are significance testing parameter and small constant .…”

“…As unsupervised learning, clustering is greatly influenced by similarity measurement and is closely related with priors in application fields. Clustering is extensively applied in fields such as biology [1,2], computer vision [3,4], geological exploration [5][6][7], and information retrieval [8], because it shows excellent advantages in automatic grouping.…”

“…Agglomerative clustering initially sees all data as one group and then conducts the fusion of clusters based on certain principles, thus forming proper data groups. Partitional clustering redistributes the existing data groups and forms proper clusters, such as k-means [9] and fuzzy c-means [5][6][7].…”

Fuzzy Sets in Earth and Space Sciences

“…The model providing satisfactory results highlighted that fuzzy based inference model was found as a useful tool for estimation of internal waves in ocean engineering. Sun and Li [33] developed a multidomain clustering inversion algorithm and used fuzzy c-means clustering technique for converting statistical petrophysical data into a deterministic geophysical inversion framework. They tested the algorithm with examples of synthetic and a field data.…”

Inversion of vector-acoustic data in a local domain

“…The high intensity areas of the gradient correspond with the locations of the Gaussian anomalies in the true model. There are also high intensities near the local domain borders, which influence the inversion, but those can be compensated for by incorporating an additional model objective function term (Sun and Li, 2015), as opposed to only using a data objective function in this example.…”