	US 12,352,913 B2
	Storage and retrieval of subsurface rock physical property prediction models using seismic interpretation
Leonardo Silveira de Albuquerque Martins, Campinas (BR); Dario Sergio Cersosimo, Lisbon (PT); Leonardo Pondian Tizzei, Sao Paulo (BR); Emilio Ashton Vital Brazil, Rio de Janeiro (BR); Rodrigo da Silva Ferreira, Rio de Janeiro (BR); Renato Fontoura de Gusmao Cerqueira, Rio de Janeiro (BR); Joana de Noronha Ribeiro de Almeida, Lisbon (PT); Astrid de Jesus Torres Fernandez, Lisbon (PT); and Ulisses Thibes Mello, Blauvelt, NY (US)
Assigned to International Business Machines Corporation, Armonk, NY (US); and Petrogal Brasil S.A., Rio de Janeiro (BR)
Filed by International Business Machines Corporation, Armonk, NY (US); and Petrogal Brasil S.A., Rio de Janeiro (BR)
Filed on Mar. 8, 2021, as Appl. No. 17/194,349.
Prior Publication US 2022/0291405 A1, Sep. 15, 2022
Int. Cl. G01V 1/28 (2006.01); G01V 1/50 (2006.01); G06F 30/20 (2020.01)

CPC G01V 1/282 (2013.01) [G01V 1/50 (2013.01); G06F 30/20 (2020.01); G01V 2210/61 (2013.01)]

17 Claims

1. A computer implemented method for storing and retrieving subsurface rock physical property models for well drilling, comprising:

generating a plurality of estimated subsurface rock physical property models, each estimated subsurface rock physical property model being based on a respective set of seismic data;

determining a stratigraphic classification based on seismic interpretation data for each of the plurality of estimated subsurface rock physical property models;

storing the plurality of estimated subsurface rock physical property models associated with its respective stratigraphic classification in a database;

determining characteristics of the seismic data used to generate each of the estimated subsurface rock physical property models;

selecting from the database a subset of the estimated subsurface rock physical property models based on a similarity of the stratigraphic classification;

determining measures of similarity between the characteristics of the seismic data used to generate each of subset of estimated subsurface rock physical property models and a new set of seismic data;

retrieving at least one estimated subsurface rock physical property model from the subset of estimated subsurface rock physical property model based on the similarity measures;

processing a given a seismic image into image texture descriptors for serving as a horizontal filter in determining the measures of similarity;

performing, based on the at least one estimated subsurface rock physical property model, drilling exploration of geolocation associated with the new set of seismic data;

inserting a new well data into the database;

notifying that the new well data is successfully inserted into the database;

searching the database and finding an estimated subsurface rock physical property model associated with seismic data of the new well;

requesting and receiving permission to update the estimated subsurface rock physical property model; and

updating the estimated subsurface rock physical property model based on the new well data;

wherein determining measures of similarity and retrieving at least one estimated subsurface rock physical property model, comprises:

providing a list of available seismic data interpretations based a current set of seismic data from a second area;

selecting a subset of the available seismic data interpretations;

measuring a statistical distribution of the current set of seismic data and its derivatives;

searching the database for models associated with seismic data interpretations having stratigraphic classifications that match those of the selected subset of seismic data interpretations;

comparing a goodness of fit distance of a distribution of a model for each model whose stratigraphic classifications match those of the selected subset of seismic data interpretations; and

retrieving the at least one estimated subsurface rock physical property model based on the comparison.