Method and system for determining permeability and lost circulation

A method, comprising:

• acquiring, with a Magnetic Resonance (MR) scanner or a pulse Nuclear Magnetic Resonance (NMR) logging tool that generates a magnetic field and measures a presence of protons in the magnetic field, first NMR data and permeability data regarding a geological region of interest;
• wherein the first NMR data is organized according to one or more T2 signal parameters,
• obtaining a plurality of ratio values of free fluids volume index (FFI) over bound fluids volumes (BFV) regarding the plurality of training wells;
• wherein the BFV data is determined based on a sum of a BFI value and a value for clay bound water;
• wherein the plurality of ratio values comprises bulk volume irreducible (BVI) values and FFI values from training wells and a target well;
• obtaining, using a computer processor, the first nuclear magnetic resonance (NMR) data for the target well and second NMR data for a plurality of training wells;
• obtaining, using the computer processor and the plurality of ratio values, first acquired permeability data for the plurality of training wells, wherein the first acquired permeability data is acquired using core samples from the plurality of training wells;
• determining, using the computer processor, a neural network, and the second NMR data, first predicted permeability data for the target well;
• updating, using the computer processor, the neural network based on a difference between the first predicted permeability data and the first acquired permeability data,
• wherein the neural network is updated to produce a trained neural network by adjusting one or more T2 cutoff values regarding the first NMR data and the second NMR data for different formations;
• transmitting, by the computer processor, a command to a well system that triggers a well operation using the predetermined type of LCM,
• wherein the well operation is a drilling operation that supplies a drilling fluid with a predetermined LCM material into a wellbore coupled to the well system, and wherein the predetermined LCM material corresponds to the predetermined type of LCM,
• determining a training filter based on one or more geological parameters, wherein the one or more geological parameters correspond to a predetermined grain density, a predetermined core porosity, a predetermined core permeability, or a predetermined NMR porosity;
• wherein the training filter comprises one or more filter ranges configured to filter NMR data from training wells and the target well, and
• determining filtered NMR data using the training filter and the first NMR data,
• where the neural network is trained using the filtered NMR data.

A system, comprising:

• a logging system coupled to a Magnetic Resonance (MR) scanner or a pulse Nuclear Magnetic Resonance (NMR) logging tool that is configured to generate a magnetic field and measure a presence of protons in the magnetic field;
• a well system coupled to the logging system and a wellbore;
• and
• a reservoir simulator comprising a computer processor, wherein the reservoir simulator is coupled to the logging system and the well system, the reservoir simulator comprising functionality for:
• obtaining, by the MR scanner or the NMR logging tool, first NMR data regarding a geological region of interest;
• wherein the first NMR data is organized according to one or more T2 signal parameters,
• obtaining a plurality of ratio values of free fluids volume index (FFI) over bound fluids volumes (BFV) regarding the geological region of interest;
• wherein the BFV data is determined based on a sum of a BFI value and a value for clay bound water;
• determining, by the computer processor using the plurality of ratio values, and using a neural network and the first NMR data, first predicted permeability data regarding a predetermined formation within the geological region of interest, wherein the neural network is trained using second NMR data and acquired permeability data regarding the geological region of interest;
• wherein the plurality of ratio values comprises bulk volume irreducible (BVI) values and FFI values from training wells and a target well;
• determining a predetermined fracture size within the predetermined formation based on the first predicted permeability data;
• determining a predetermined type of lost circulation material (LCM) based on the predetermined fracture size;
• and
• transmitting a command to the well system that triggers a well operation using the predetermined type of LCM,
• wherein the well operation is a drilling operation that supplies a drilling fluid with a predetermined LCM material into a wellbore coupled to the well system, and
• wherein the predetermined LCM material corresponds to the predetermined type of LCM,
• determining a training filter based on one or more geological parameters, wherein the one or more geological parameters correspond to a predetermined grain density, a predetermined core porosity, a predetermined core permeability, or a predetermined NMR porosity;
• wherein the training filter comprises one or more filter ranges configured to filter NMR data from training wells and the target well, and
• determining filtered NMR data using the training filter and the first NMR data,
• where the neural network is trained using the filtered NMR data.