Controlling resistance in well system

Abstract

A method for determining a new factor vector for a first hydraulic fracture treatment in a wellbore. The method may include determining, based on one or more first pressure pulses, a first current resistance in the well during the first hydraulic fracture treatment; determining a second factor vector based, at least in part, on a first factor vector and the first current resistance and a target resistance; implementing the second factor vector in the wellbore; determining, based on one or more second pressure pulses, a second current resistance in the well during the first hydraulic fracture treatment; determining whether the second current resistance is above a resistance threshold; and in response to the second current resistance being above the resistance threshold, determining a third factor vector.

What is claimed is: 
     
         1 . A method for determining a new factor vector for a first hydraulic fracture treatment in a wellbore, the method comprising:
 determining, based on one or more first pressure pulses, a first current resistance in the well during the first hydraulic fracture treatment;   determining a second factor vector based, at least in part, on a first factor vector and the first current resistance and a target resistance;   implementing the second factor vector in the wellbore;   determining, based on one or more second pressure pulses, a second current resistance in the well during the first hydraulic fracture treatment;   determining whether the second current resistance is above a resistance threshold; and   in response to the second current resistance being above the resistance threshold, determining a third factor vector.   
     
     
         2 . The method of  claim 1  further comprising:
 in response to the second current resistance being below a threshold resistance, continuing with the first hydraulic fracture treatment in the wellbore. 
 
     
     
         3 . The method of  claim 1  further comprising:
 determining a fourth factor vector for a second hydraulic fracture treatment based on a third current resistance that was determined during the first hydraulic fracture treatment, a fourth current resistance that was determined during the first hydraulic fracture treatment, and a target resistance for the second hydraulic fracture treatment. 
 
     
     
         4 . The method of  claim 1  further comprising:
 obtaining training samples each including a first current resistant sample, a first factor vector sample, a second current resistance sample, and a second factor vector sample; 
 training, with the training samples, a machine learning model to output a predicted factor vector that matches the second factor vector sample based on the first current resistance sample, the first factor vector sample, and the second current resistance sample. 
 
     
     
         5 . The method of  claim 4 , wherein the determining the second factor vector occurs after the training and includes an inversion of the machine learning model by which the machine learning model outputs the second factor vector based on the first factor vector, first current resistance, and the second current resistance. 
     
     
         6 . The method of  claim 1 , wherein the determining whether the second current resistance is above a resistance threshold includes:
 determining a difference between the second current resistance and the target resistance;   in response to the difference being above a given resistance value, determining the second current resistance is above the resistance threshold; and   in response to the difference being below a given resistance value, determining the second current resistance is below the resistance threshold.   
     
     
         7 . The method of  claim 6 , wherein the target resistance is based, at least in part, on a design perforation resistance. 
     
     
         8 . The method of  claim 1  further comprising:
 implementing the first factor vector during the first hydraulic fracture treatment in the wellbore. 
 
     
     
         9 . One or more non-transitory computer-readable mediums including instructions that, when executed by a processor, perform operations for determining a new factor vector for a first hydraulic fracture treatment in a wellbore, the instructions comprising:
 instructions to determine, based on one or more first pressure pulses, a first current resistance in the well during the first hydraulic fracture treatment;   instructions to determine a second factor vector based, at least in part, on a first factor vector and the first current resistance and a target resistance;   instructions to implement the second factor vector in the wellbore;   instructions to determine, based on one or more second pressure pulses, a second current resistance in the well during the first hydraulic fracture treatment;   instructions to determine whether the second current resistance is above a resistance threshold; and   instructions to, in response to the second current resistance being above the resistance threshold, determine a third factor vector.   
     
     
         10 . The one or more computer-readable mediums of  claim 9  further comprising:
 instructions to, in response to the second current resistance being below a threshold resistance, continue with the first hydraulic fracture treatment in the wellbore. 
 
     
     
         11 . The one or more computer-readable mediums of  claim 9 , the instructions further comprising:
 instructions to determine a fourth factor vector for a second hydraulic fracture treatment based on a third current resistance that was determined during the first hydraulic fracture treatment, a fourth current resistance that was determined during the first hydraulic fracture treatment, and a target resistance for the second hydraulic fracture treatment.   
     
     
         12 . The one or more computer-readable mediums of  claim 9 , the instructions further comprising:
 instructions to obtain training samples each including a first current resistant sample, a first factor vector sample, a second current resistance sample, and a second factor vector sample;   instructions to train, with the training samples, a machine learning model to output a predicted factor vector that matches the second factor vector sample based on the first current resistance sample, the first factor vector sample, and the second current resistance sample.   
     
     
         13 . The one or more computer-readable mediums of  claim 12 , wherein the determination of the second factor vector to occur after the instructions to train and include an inversion of the machine learning model by which the machine learning model to output the second factor vector based on the first factor vector, first current resistance, and the second current resistance. 
     
     
         14 . The one or more computer-readable mediums of  claim 9 , wherein the instructions to determine whether the second current resistance is above a resistance threshold includes:
 instructions to determine a difference between the second current resistance and the target resistance;   instructions to, in response to the difference being above a given resistance value, determine the second current resistance is above the resistance threshold; and   instructions to, in response to the difference being below a given resistance value, determine the second current resistance is below the resistance threshold.   
     
     
         15 . The one or more computer-readable mediums of  claim 14 , wherein the target resistance is based, at least in part, on a design perforation resistance. 
     
     
         16 . The one or more computer-readable mediums of  claim 9 , the instructions further comprising:
 instructions to implement the first factor vector during the first hydraulic fracture treatment in the wellbore.   
     
     
         17 . A computer system comprising:
 a processor;   one or more non-transitory computer-readable mediums including instructions that, when executed by the processor, perform operations for determining a new factor vector for a first hydraulic fracture treatment in a wellbore, the instructions comprising:
 instructions to determine, based on one or more first pressure pulses, a first current resistance in the well during the first hydraulic fracture treatment; 
 instructions to determine a second factor vector based, at least in part, on a first factor vector and the first current resistance and a target resistance; 
 instructions to implement the second factor vector in the wellbore; 
 instructions to determine, based on one or more second pressure pulses, a second current resistance in the well during the first hydraulic fracture treatment; 
 instructions to determine whether the second current resistance is above a resistance threshold; and 
 instructions to, in response to the second current resistance being above the resistance threshold, determine a third factor vector. 
   
     
     
         18 . The computer system of  claim 17 , the instructions further comprising:
 instructions to, in response to the second current resistance being below a threshold resistance, continue with the first hydraulic fracture treatment in the wellbore.   
     
     
         19 . The computer system of  claim 17 , the instructions further comprising:
 instructions to determine a fourth factor vector for a second hydraulic fracture treatment based on a third current resistance that was determined during the first hydraulic fracture treatment, a fourth current resistance that was determined during the first hydraulic fracture treatment, and a target resistance for the second hydraulic fracture treatment.   
     
     
         20 . The computer system of  claim 17 , the instructions further comprising:
 instructions to obtain training samples each including a first current resistant sample, a first factor vector sample, a second current resistance sample, and a second factor vector sample;   instructions to train, with the training samples, a machine learning model to output a predicted factor vector that matches the second factor vector sample based on the first current resistance sample, the first factor vector sample, and the second current resistance sample.