Application of artificial bee colony-based neural network in bottom hole pressure prediction in underbalanced drilling

2011 ◽  
Vol 78 (1) ◽  
pp. 6-12 ◽  
Author(s):  
Rasoul Irani ◽  
Reza Nasimi
Keyword(s):  
Neural Network ◽  
Artificial Bee Colony ◽  
Hole Pressure ◽  
Bottom Hole Pressure ◽  
Bee Colony ◽  
Underbalanced Drilling ◽  
Bottom Hole

Neural Networks for Flow Bottom Hole Pressure Prediction

2016 ◽  
Vol 6 (4) ◽  
pp. 1839 ◽  
Author(s):  
Medhat Awadalla ◽  
Hassan Yousef
Keyword(s):  
Neural Network ◽  
Mean Square Error ◽  
Correlation Coefficient ◽  
Data Sets ◽  
Mean Square ◽  
Layer Model ◽  
Hole Pressure ◽  
Bottom Hole Pressure ◽  
Bottom Hole ◽  
Hidden Layer

Installation of down-hole gauges in oil wells to determine Flowing Bottom-Hole Pressure (FBHP) is a dominant process especially in wells lifted with electrical submersible pumps.  However, intervening a well occasionally is an exhaustive task, associated with production risk, and interruption. The previous empirical correlations and mechanistic models failed to provide a satisfactory and reliable tool for estimating pressure drop in multiphase flowing wells. This paper aims to find the optimum parameters of Feed-Forward Neural Network (FFNN) with back-propagation algorithm to predict the flowing bottom-hole pressure in vertical oil wells.  The developed neural network models rely on a large amount of available historical data measured from actual different oil fields. The unsurpassed number of neural network layers, the number of neurons per layer, and the number of trained samples required to get an outstanding performance have been obtained. Intensive experiments have been conducted and for the sake of qualitative comparison, Radial Basis neural and network and the empirical modes have been developed. The paper showed that the accuracy of FBHP estimation using FFNN with two hidden layer model is better than FFNN with single hidden layer model, Radial Basis neural network, and the empirical model in terms of data set used, mean square error, and the correlation coefficient error. With best results of 1.4 root mean square error (RMSE), 1.4 standard deviation of relative error (STD), correlation coefficient (R) 1.0 and 99.4% of the test data sets achieved less than 5% error. The minimum sufficient number of data sets used in training ANN model can be low as 12.5% of the total data sets to give 3.4 RMSE and 97% of the test data achieved 90% accuracy.

Managed Pressure Drilling and Underbalanced Drilling Compounded Role in Solving the Exploration and Appraisal Challenges of Biogenetic Gas Resources in the United Arab Emirates -Case Study

2021 ◽  
Author(s):  
Ahmed Al Mutawa ◽  
Ibrahim Hamdy ◽  
Eias Daban Al Shamisi ◽  
Bassem El Yossef ◽  
Mohamed Sameer Amin ◽  
...  
Keyword(s):  
United Arab Emirates ◽  
Drilling Fluid ◽  
Productivity Index ◽  
Biogenic Gas ◽  
Hole Pressure ◽  
Bottom Hole Pressure ◽  
Underbalanced Drilling ◽  
Bottom Hole ◽  
Novel Approach ◽  
Managed Pressure Drilling

Abstract Biogenic gas resources have gathered importance recently due to its widespread availability, occurrence at geologically predictable circumstances, and existence at shallow depths. It is estimated that biogenic gas forms more than 20% of the global discovered reserves. However, the exploration and development of these unconventional resources come with numerous drilling and reservoir challenges. This paper showcases a novel approach used in the United Arab Emirates to overcome these challenges using managed pressure and underbalanced drilling. To tackle both reservoir and drilling challenges, a hybrid solution combining Underbalanced (UBD) and Managed Pressure Drilling (MPD) was applied. UBD was used to characterize the reservoir in terms of pressure and productivity index to ultimately enhance productivity by eliminating formation damage. MPD was used next to continue drilling through the problematic zone which had high instability due to the presence of highly sensitive salt, in addition to the presence of high pressure and loss zones. The fit for purpose hybrid application design allowed the operator to immediately switch between UBD and MPD conditions, as the well required with the same equipment. Three of the four targeted formations were in the 8 ½″ hole section, UBD was selected to drill the first reservoir formation which allowed pore pressure verification and avoided using excessive mud weight that was the culprit of many challenges like slow ROP, drilling fluid losses, bit balling, and fracking the formations. UBD has proved that mud weight can be reduced by 20%-30% comparing to conventional drilling. The second formation was a salt formation that has caused previously hole collapse and losses-kicks problems as heavy mud used to drill this salty formation. MPD used successfully drill this section by constant bottom hole pressure and lower mud weight as it was found from analyzing offset wells reports that hole collapse occurred at connections and pump off events. Constant Bottom Hole Pressure (CBHP) also eliminated tight spots and excessive reaming resulting in optimized drilling. The third formation used MPD as well to minimize overbalance pressure over previous sections while the fourth formation was drilled by UBD as it had a separate 6″ hole section as it formed an independent reservoir. The combined MPD and UBD approach eliminated most the NPT encountered in offset wells, enhanced Rate of Penetration (ROP) by 200% to 300% and slashed the well drilling time by 27 days.

scholarly journals Neural Networks for Flow Bottom Hole Pressure Prediction

2016 ◽  
Vol 6 (4) ◽  
pp. 1839
Author(s):  
Medhat Awadalla ◽  
Hassan Yousef
Keyword(s):  
Neural Network ◽  
Mean Square Error ◽  
Correlation Coefficient ◽  
Data Sets ◽  
Mean Square ◽  
Layer Model ◽  
Hole Pressure ◽  
Bottom Hole Pressure ◽  
Bottom Hole ◽  
Hidden Layer

Installation of down-hole gauges in oil wells to determine Flowing Bottom-Hole Pressure (FBHP) is a dominant process especially in wells lifted with electrical submersible pumps.  However, intervening a well occasionally is an exhaustive task, associated with production risk, and interruption. The previous empirical correlations and mechanistic models failed to provide a satisfactory and reliable tool for estimating pressure drop in multiphase flowing wells. This paper aims to find the optimum parameters of Feed-Forward Neural Network (FFNN) with back-propagation algorithm to predict the flowing bottom-hole pressure in vertical oil wells.  The developed neural network models rely on a large amount of available historical data measured from actual different oil fields. The unsurpassed number of neural network layers, the number of neurons per layer, and the number of trained samples required to get an outstanding performance have been obtained. Intensive experiments have been conducted and for the sake of qualitative comparison, Radial Basis neural and network and the empirical modes have been developed. The paper showed that the accuracy of FBHP estimation using FFNN with two hidden layer model is better than FFNN with single hidden layer model, Radial Basis neural network, and the empirical model in terms of data set used, mean square error, and the correlation coefficient error. With best results of 1.4 root mean square error (RMSE), 1.4 standard deviation of relative error (STD), correlation coefficient (R) 1.0 and 99.4% of the test data sets achieved less than 5% error. The minimum sufficient number of data sets used in training ANN model can be low as 12.5% of the total data sets to give 3.4 RMSE and 97% of the test data achieved 90% accuracy.

Sign in / Sign up

Export Citation Format

Share Document