Machine-learning based rig-site on-demand drilling mud characterization, property prediction, and optimization
A computer-implemented method for machine learning based rig-site on-demand drilling mud characterization, property prediction and optimization is described. The method includes predicting rheological properties of drilling mud from compositional data based on a first empirical relationship across the compositional data. The method also includes predicting sensor based rheological performance of drilling mud based on a second relationship between the predicted rheological properties of the drilling mud and sensor responses from one or more sensors. Additionally, the method includes providing feedback to the one or more sensors, wherein the feedback comprises the sensor based rheological performance, sensor responses, and a formulation associated with the drilling mud, wherein the feedback enables updating the formulation associated with the drilling mud.
1 . A computer-implemented method for machine learning based rig-site on-demand drilling mud characterization, property prediction and optimization, the method comprising:
predicting, with one or more hardware processors, rheological properties of drilling mud from compositional data based on a first empirical relationship across the compositional data, wherein the compositional data comprises multiple material mud properties;
predicting, with the one or more hardware processors, sensor based rheological performance of drilling mud based on a second relationship between the predicted rheological properties of the drilling mud and sensor responses from one or more sensors, wherein sensor-based rheological performance comprises a behavior of the drilling mud at the sensor responses; and
providing, with the one or more hardware processors, feedback to the one or more sensors, wherein the feedback comprises the sensor based rheological performance, sensor responses, and a formulation associated with the drilling mud, wherein the feedback enables updating the formulation associated with the drilling mud.
2 . The computer implemented method of claim 1 , comprising transmitting the updated formulation to a mixing tank and applying the formulation at the mixing tank.
3 . The computer-implemented method of claim 1 , wherein predicting sensor based rheological performance of drilling mud from compositional data comprises
extracting and interpolating rheological behavior of the drilling mud from the compositional data;
converting the rheological behavior into a single vector using time-temperature superposition; and
predicting the sensor based rheological performance using a gradient boosted decision tree, wherein the sensor based rheological performance comprises behavior of the drilling mud at the response of a sensor, wherein the response is temperature.
4 . The computer-implemented method of claim 1 , wherein predicting sensor based rheological performance of drilling mud from compositional data comprises:
converting the rheological behavior of the drilling mud from the compositional data to a matrix;
predicting the sensor based rheological performance using a fully connected neural network, wherein the sensor based rheological performance comprises behavior of the drilling mud at the response of a sensor, wherein the response is temperature.
5 . The computer-implemented method of claim 1 , wherein predicting sensor based rheological performance of drilling mud from compositional data comprises:
converting the rheological behavior of the drilling mud from the compositional data to a matrix;
predicting the sensor based rheological performance using a convolutional neural network, wherein the sensor based rheological performance comprises behavior of the drilling mud at the response of a sensor, wherein the response is temperature.
6 . The computer implemented method of claim 1 , wherein the multiple materials mud properties comprise plastic viscosity, shear yield stress, yield strain, flow point, yield point, low shear yield point, gel strength, fluid loss control and weight percentage, or any combinations thereof.
7 . The computer implemented method of claim 1 , wherein the sensor-based rheological performance comprises on-demand drilling mud characterization, mud property prediction, and optimization of drill mud formulation.
8 . An apparatus comprising a non-transitory, computer readable, storage medium that stores instructions that, when executed by at least one processor, cause the at least one processor to perform operations comprising:
predicting rheological properties of drilling mud from compositional data based on a first empirical relationship across the compositional data, wherein the compositional data comprises multiple material mud properties;
predicting sensor based rheological performance of drilling mud based on a second relationship between the predicted rheological properties of the drilling mud and sensor responses from one or more sensors, wherein sensor-based rheological performance comprises a behavior of the drilling mud at the sensor responses; and
providing feedback to the one or more sensors, wherein the feedback comprises the sensor based rheological performance, sensor responses, and a formulation associated with the drilling mud, wherein the feedback enables updating the formulation associated with the drilling mud.
9 . The apparatus of claim 8 , the operations comprising transmitting the updated formulation to a mixing tank and applying the formulation at the mixing tank.
10 . The apparatus of claim 8 , wherein predicting sensor based rheological performance of drilling mud from compositional data comprises:
extracting and interpolating rheological behavior of the drilling mud from the compositional data;
converting the rheological behavior into a single vector using time-temperature superposition; and
predicting the sensor based rheological performance using a gradient boosted decision tree, wherein the sensor based rheological performance comprises behavior of the drilling mud at the response of a sensor, wherein the response is temperature.
11 . The apparatus of claim 8 , wherein predicting sensor based rheological performance of drilling mud from compositional data comprises:
converting the rheological behavior of the drilling mud from the compositional data to a matrix;
predicting the sensor based rheological performance using a fully connected neural network, wherein the sensor based rheological performance comprises behavior of the drilling mud at the response of a sensor, wherein the response is temperature.
12 . The apparatus of claim 8 , wherein predicting sensor based rheological performance of drilling mud from compositional data comprises:
converting the rheological behavior of the drilling mud from the compositional data to a matrix;
predicting the sensor based rheological performance using a convolutional neural network, wherein the sensor based rheological performance comprises behavior of the drilling mud at the response of a sensor, wherein the response is temperature.
13 . The apparatus of claim 8 , wherein the multiple materials mud properties comprise plastic viscosity, shear yield stress, yield strain, flow point, yield point, low shear yield point, gel strength, fluid loss control and weight percentage, or any combinations thereof.
14 . The apparatus of claim 8 , wherein the sensor-based rheological performance comprises on-demand drilling mud characterization, mud property prediction, and optimization of drill mud formulation.
15 . A system, comprising:
one or more memory modules;
one or more hardware processors communicably coupled to the one or more memory modules, the one or more hardware processors configured to execute instructions stored on the one or more memory models to perform operations comprising:
predicting rheological properties of drilling mud from compositional data based on a first empirical relationship across the compositional data, wherein the compositional data comprises multiple material mud properties;
predicting sensor based rheological performance of drilling mud based on a second relationship between the predicted rheological properties of the drilling mud and sensor responses from one or more sensors, wherein sensor-based rheological performance comprises a behavior of the drilling mud at the sensor responses; and
providing feedback to the one or more sensors, wherein the feedback comprises the sensor based rheological performance, sensor responses, and a formulation associated with the drilling mud, wherein the feedback enables updating the formulation associated with the drilling mud.
16 . The system of claim 15 , the operations comprising transmitting the updated formulation to a mixing tank and applying the formulation at the mixing tank.
17 . The system of claim 15 , wherein predicting sensor based rheological performance of drilling mud from compositional data comprises:
extracting and interpolating rheological behavior of the drilling mud from the compositional data;
converting the rheological behavior into a single vector using time-temperature superposition; and
predicting the sensor based rheological performance using a gradient boosted decision tree, wherein the sensor based rheological performance comprises behavior of the drilling mud at the response of a sensor, wherein the response is temperature.
18 . The system of claim 15 , wherein predicting sensor based rheological performance of drilling mud from compositional data comprises:
converting the rheological behavior of the drilling mud from the compositional data to a matrix;
predicting the sensor based rheological performance using a fully connected neural network, wherein the sensor based rheological performance comprises behavior of the drilling mud at the response of a sensor, wherein the response is temperature.
19 . The system of claim 15 , wherein predicting sensor based rheological performance of drilling mud from compositional data comprises:
converting the rheological behavior of the drilling mud from the compositional data to a matrix;
predicting the sensor based rheological performance using a convolutional neural network, wherein the sensor based rheological performance comprises behavior of the drilling mud at the response of a sensor, wherein the response is temperature.
20 . The system of claim 15 , wherein the multiple materials mud properties comprise plastic viscosity, shear yield stress, yield strain, flow point, yield point, low shear yield point, gel strength, fluid loss control and weight percentage, or any combinations thereof.