Spaces:
Runtime error
Runtime error
Commit
β’
bc2eec8
1
Parent(s):
9514984
Update Project Description
Browse files- Project Description +20 -15
Project Description
CHANGED
@@ -1,8 +1,10 @@
|
|
1 |
Oil Flow Predictor (Barrels Per Day) Using Machine Learning Algorithms β Random Forest and Gradient Boosting
|
2 |
-
Machine. Executives in the Oil & Gas industry are always seeking ways to optimize the production of oil and
|
|
|
3 |
wells and make informed decisions about drilling and exploration. Machine learning can be used to effectively
|
4 |
resolve this optimization problem, but would usually involve extensive or complex coding that the typical
|
5 |
-
executive or non-technical staff would not be able to undertake. For example, the code behind this simple
|
|
|
6 |
bot does thinngs the typical oil executive knows nothing about in that it loads the data from a CSV file and
|
7 |
performs data scaling, PCA dimensionality reduction, DBSCAN clustering, and random forest regression and
|
8 |
gradient boosting regression for predicting output variables related to oil and gas production rates β not
|
@@ -11,17 +13,20 @@ insights into the factors that affect oil and gas production rates and to predic
|
|
11 |
the input variables. Executives are then able to use this knowledge to improve the efficiency of their
|
12 |
operations.
|
13 |
|
14 |
-
This Gradio chatbot reads in a CSV file containing well rate data, performs some preprocessing on the data,
|
15 |
-
|
|
|
|
|
16 |
the preprocessed data. The input data is preprocessed using Principal Component Analysis (PCA) to reduce the
|
17 |
-
dimensionality of the input features, and then the DBSCAN clustering algorithm is applied to the
|
|
|
18 |
data to filter out outlier data points. The remaining data points are used for training and testing the
|
19 |
regression models. The RF and GBM models are both trained using the preprocessed input features and
|
20 |
-
corresponding well rates, and their performance is evaluated using the R-squared metric. Finally, the
|
21 |
-
well rates from the RF and GBM models are plotted against the actual well rates for visualization.
|
22 |
|
23 |
-
For example, the following inputs are required for predicting the oil flow rate (Qoil) based on input
|
24 |
-
(BHP, WHP, WHT, Tsep, Psep, and Choke_in):
|
25 |
|
26 |
β’ BHP: Bottom Hole Pressure
|
27 |
β’ WHP: Well Head Pressure
|
@@ -30,10 +35,10 @@ For example, the following inputs are required for predicting the oil flow rate
|
|
30 |
β’ Psep: Separator Pressure
|
31 |
β’ Choke_in: Choke Size
|
32 |
|
33 |
-
These features (BHP, WHP, WHT, Tsep, Psep, and Choke_in) are used as inputs to the machine learning models
|
34 |
-
were trained in the notebook to predict the oil flow rate (Qoil). We suspect and propose that the same
|
35 |
-
features (BHP, WHP, WHT, Tsep, Psep, and Choke_in) are provided to the trained model which would
|
36 |
-
enabled to predict Qoil or the oil flow rate (number of barrels per day) via the Gradio input
|
37 |
-
The oil executive is able to learn this without having to know machine learning or the
|
38 |
-
training the model himself.
|
39 |
|
|
|
1 |
Oil Flow Predictor (Barrels Per Day) Using Machine Learning Algorithms β Random Forest and Gradient Boosting
|
2 |
+
Machine. Executives in the Oil & Gas industry are always seeking ways to optimize the production of oil and
|
3 |
+
gas
|
4 |
wells and make informed decisions about drilling and exploration. Machine learning can be used to effectively
|
5 |
resolve this optimization problem, but would usually involve extensive or complex coding that the typical
|
6 |
+
executive or non-technical staff would not be able to undertake. For example, the code behind this simple
|
7 |
+
Gradio
|
8 |
bot does thinngs the typical oil executive knows nothing about in that it loads the data from a CSV file and
|
9 |
performs data scaling, PCA dimensionality reduction, DBSCAN clustering, and random forest regression and
|
10 |
gradient boosting regression for predicting output variables related to oil and gas production rates β not
|
|
|
13 |
the input variables. Executives are then able to use this knowledge to improve the efficiency of their
|
14 |
operations.
|
15 |
|
16 |
+
This Gradio chatbot reads in a CSV file containing well rate data, performs some preprocessing on the data,
|
17 |
+
and
|
18 |
+
then trains and evaluates two regression models, a Random Forest (RF) and a Gradient Boosting Machine (GBM),
|
19 |
+
on
|
20 |
the preprocessed data. The input data is preprocessed using Principal Component Analysis (PCA) to reduce the
|
21 |
+
dimensionality of the input features, and then the DBSCAN clustering algorithm is applied to the
|
22 |
+
PCA-transformed
|
23 |
data to filter out outlier data points. The remaining data points are used for training and testing the
|
24 |
regression models. The RF and GBM models are both trained using the preprocessed input features and
|
25 |
+
corresponding well rates, and their performance is evaluated using the R-squared metric. Finally, the
|
26 |
+
predicted well rates from the RF and GBM models are plotted against the actual well rates for visualization.
|
27 |
|
28 |
+
For example, the following inputs are required for predicting the oil flow rate (Qoil) based on input
|
29 |
+
features (BHP, WHP, WHT, Tsep, Psep, and Choke_in):
|
30 |
|
31 |
β’ BHP: Bottom Hole Pressure
|
32 |
β’ WHP: Well Head Pressure
|
|
|
35 |
β’ Psep: Separator Pressure
|
36 |
β’ Choke_in: Choke Size
|
37 |
|
38 |
+
These features (BHP, WHP, WHT, Tsep, Psep, and Choke_in) are used as inputs to the machine learning models
|
39 |
+
that were trained in the notebook to predict the oil flow rate (Qoil). We suspect and propose that the same
|
40 |
+
input features (BHP, WHP, WHT, Tsep, Psep, and Choke_in) are provided to the trained model which would
|
41 |
+
then be enabled to predict Qoil or the oil flow rate (number of barrels per day) via the Gradio input
|
42 |
+
interface. The oil executive is able to learn this without having to know machine learning or the
|
43 |
+
complexities of training the model himself.
|
44 |
|