Real-time porosity prediction using gas-while-drilling data and machine learning with reservoir associated gas: Case study for Hassi Messaoud field, Algeria

Ameur-Zaimeche, Ouafi; Kechiched, Rabah; Heddam, Salim; Wood, David A.

doi:10.1016/j.marpetgeo.2022.105631

Cited by 7 publications

(3 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These studies demonstrate that machine learning can effectively identify the properties of rock and soil based on drilling parameters using simple models and less noisy indoor drilling experiment data. Ameur-Zaimeche et al used different machine-learning methods to predict porosity with reservoir-associated gas using gas-while-drilling data [22]. Kadkhodaie-Ilkhchi et al based on the fuzzy logic (FL) algorithm [23], Fang et [24][25][26], applied machine learning methods in practical engineering with a large amount of noisy data.…”

Section: Introductionmentioning

confidence: 99%

BILSTM-Based Deep Neural Network for Rock-Mass Classification Prediction Using Depth-Sequence MWD Data: A Case Study of a Tunnel in Yunnan, China

Tang

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

Measurement while drilling (MWD) data reflect the drilling rig–rock mass interaction; they are crucial for accurately classifying the rock mass ahead of the tunnel face. Although machine-learning methods can learn the relationship between MWD data and rock mechanics parameters to support rock classification, most current models do not consider the impact of the continuous drilling-sequence process, thereby leading to rock-classification errors, while small and unbalanced field datasets result in poor model performance. We propose a novel deep neural network model based on Bi-directional Long Short-Term Memory (BILSTM) to extract information-related sequences in MWD data and improve the accuracy of the rock-mass classification. Two optimization modules were designed to improve the model’s generalization performance. Stratified K-fold cross-validation was used for model optimization in small and unbalanced datasets. Model validation is based on the MWD dataset of a highway tunnel in Yunnan, China. Multiple metrics show that the prediction ability of the network is significantly better than those of a multilayer perceptron (MLP) and a support-vector machine (SVM), while the model exhibits an improved generalization performance. The accuracy of the network can reach 90%, which is 13% and 15% higher than the MLP and SVM, respectively.

show abstract

Section: Introductionmentioning

confidence: 99%

BILSTM-Based Deep Neural Network for Rock-Mass Classification Prediction Using Depth-Sequence MWD Data: A Case Study of a Tunnel in Yunnan, China

Tang

et al. 2023

Applied Sciences

View full text Add to dashboard Cite

show abstract

“…During oil and gas drilling, with the increase of formation depth, the rock strength increases, and the wellbore trajectory becomes more difficult to control [1]. In the process of drilling deep and ultra-deep wells, downhole measuring devices are used to obtain real-time downhole data, such as an instrument for measuring while drilling [2][3][4].…”

Section: Introductionmentioning

confidence: 99%

Flow field and erosion characteristics of the valve head of a vertical well inclinometer

Wang

Zhang

et al. 2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

During the drilling, the flow field characteristics of the vertical well inclinometer are complex. In a harsh underground environment, the valve head is prone to erode and wear. To overcome these disadvantages, the three-dimensional (3D) model of the valve head of a vertical well inclinometer was developed, and the flow field and erosion characteristics were simulated by using a numerical program. Results show that the maximum velocity and the maximum turbulence effect occur at the throttling inlet through which the drilling fluid passes the guide sleeve. By analyzing the particle trajectory of the discrete phase model, it is found that the particles above the valve head surface have obvious accumulation and rebound phenomena and that the erosion effect appears to some extent on the valve head surface. Analysis of the shear stress on the surface of the valve head shows that the stress concentration occurs in the valve head directly impacted by the drilling fluid and in the groove of the flow drilling fluid, respectively. The feasibility and accuracy of the simulation are verified by comparing the theoretical results with that of the field test. This investigation can be used to explain the failure of the valve head of the vertical well inclinometer and provide the design and optimization of the valve head with scientific support.

show abstract

“…In the process of drilling complex deep wells, passive drilling techniques based on post-accident improvement techniques and equipment can easily lead to problems such as low drilling efficiency and poor wellbore trajectory [2]. Measurement while drilling enables active-controlled drilling by measuring downhole data in real-time, which is important for obtaining undisturbed formation information and improving drilling efficiency [3][4][5].…”

Section: Introductionmentioning

confidence: 99%

Design and Application of Downhole Drilling Mud Pulse Signal Transmission System

Zhang,

Chen,

et al. 2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

Time-efficient measurement of undisturbed formation through measurement while drilling technology is an effective means to improve drilling efficiency in deep formations and increase the drilling-encounter ratio of reservoirs. Based on the application and development trend of existing downhole signal transmission technology, a pulse signal transmission system of downhole mud while drilling is designed. The system pulse generation device adopts the structure of a positive pulse generator, which is through the internal driving circuit and hydraulic circuit of the device to drive the valve head to move to generate a mud pulse signal while using the design of a limit switch to optimize the expansion spacing of the valve head so that the generated pulse signal amplitude is close. The system uses combined encoding as the encoding method, while the ground part of the system is designed to monitor and process the collected signals. Finally, the transmission capability and accuracy of the system are tested through field experiments, and the field results show that the system can work smoothly in the downhole environment to transmit downhole data to the wellhead with small errors, which meets the design requirements and field requirements. The field applications are carried out in several wells, such as the Shixi 105 well. The results show that the system has good downhole pulse signal transmission functions such as borehole inclination measurement, which helps to promote the development of existing drilling measurement technology.

show abstract

Real-time porosity prediction using gas-while-drilling data and machine learning with reservoir associated gas: Case study for Hassi Messaoud field, Algeria

Cited by 7 publications

References 42 publications

BILSTM-Based Deep Neural Network for Rock-Mass Classification Prediction Using Depth-Sequence MWD Data: A Case Study of a Tunnel in Yunnan, China

BILSTM-Based Deep Neural Network for Rock-Mass Classification Prediction Using Depth-Sequence MWD Data: A Case Study of a Tunnel in Yunnan, China

Flow field and erosion characteristics of the valve head of a vertical well inclinometer

Design and Application of Downhole Drilling Mud Pulse Signal Transmission System

Contact Info

Product

Resources

About