Goto

Layer | Fill | Outline |
---|

Theme | Visible | Selectable | Appearance | Zoom Range (now: 0) |
---|

Fill | Stroke |
---|---|

**Abstract**

It is critical to obtain the rock strength along the wellbore to control drilling problems such as pipe sticking, tight hole, collapse and sand production. The purpose of this research is to predict the uniaxial compressive strength based on data of sonic travel time, formation porosity, density and penetration rate. For prediction of UCS, artificial neural networks were developed between UCS and input data resulting a practical correlation. In this research, a long well segment possessing complete and continuous data coverage has been analysed, and collected data of the wellbore are used to correlate data of the four mentioned input parameters of artificial neural networks with uniaxial compressive strength data as network targets. Selection of input parameters is based on a vast literature review in this area. Due to the fact that standard experimental test methods based on established standards require costly equipment and that the methods for sample preparation is difficult and time-consuming, indirect methods are more favourable. Using these methods, the UCS values are predicted in a simpler, faster and more economical way. In this study, it is concluded that artificial neural networks are a good predictor of rock strength, and can reduce drilling costs significantly. It is observed in this paper that UCS predicted values by neural networks are very close with lab and field data, which is concluded by analysis of network performance results including mean squared error and correlation coefficient. It is also concluded in this study that input parameters which are chosen in this study, have deep effects in UCS prediction studies, and should be considered in other scientific studies. Conclusions show that using artificial neural networks to predict UCS of formation rocks in petroleum fields around the world, would ease UCS estimation, optimize drilling plans and decrease costs.

**1. Introduction**

A geomechanical model requires a great deal of input information including measurements of magnitude of vertical and minimum stresses, pore pressure, rock mechanics properties and drilling experiences, all oriented to determine the magnitude of maximum horizontal stress. To conduct a geomechanical reservoir characterization, it is essential to have the knowledge of the in-situ stress magnitudes and rock mechanical properties [17].

adel asadi procedia engineering 191, application, Artificial Intelligence, artificial neural network, asadi procedia engineering 191, compressive strength, correlation, drilling data, estimation, input parameter, machine learning, neural network, porosity, prediction, Reservoir Characterization, reservoir geomechanics, rock strength, sonic travel time, strength, unconfined compressive strength, uniaxial compressive strength, Upstream Oil & Gas

Country:

- Europe (1.00)
- Asia (0.94)
- North America > United States (0.93)

Oilfield Places:

- North America > Canada > Alberta > Western Canada Sedimentary Basin > Alberta Basin > Deep Basin > Willesden Green Field > Cardium Formation (0.99)
- Europe > United Kingdom > North Sea Basin (0.98)
- Europe > United Kingdom > Atlantic Margin > West of Shetland > Faroe-Shetland Basin > Rona Ridge > Block 206/5a > Fulla Prospect (0.98)
- (4 more...)

SPE Disciplines:

Technology: Information Technology > Artificial Intelligence > Machine Learning > Neural Networks (1.00)

Abstract Newly developed software has illustrated that the safe mudweight window to avoid borehole collapse and fracturing can be obtained from the direct use of drilling and/or the use of sonic data. Instantaneous or post analysis using inverted drilling operational data like WOB, RPM, flowrate, and mud properties can be done by applying an inverted ROP model to predict rock compressive strength. This rock compressive strength is then used in conjunction with survey, in-situ stresses and rock failure criteria to predict the safe mud weight window. The sonic logs are also used to obtain the rock compressional properties through published correlations. The rock compressive strength is then used the same way as the drilling data approach. This paper gives a solid field example from a North Sea well where both methods are applied and compared. The field case provides proof that the new software is a very good tool for predicting a safe mud weight window to avoid borehole collapse and fracturing while drilling. Introduction During drilling operations it is crucial to maintain the hydrostatic mud weight pressure between the fracture and collapse pressure at any depth to ensure trouble free drilling. The casing program is also dictated mainly from the fracture and collapse gradient. Therefore, the knowledge and use of the correct fracture and collapse gradients can save the operating companies large sums of money. When estimating the safe operational mud weight window two important parameters arethe strength of the rock, and the stress conditions the rock is subjected to. The question is; how can these parameters be monitored while drilling a well. Today there is no way of measuring compressive rock strength or stresses while drilling. The best method of determining compressive rock strength is by triaxial testing of core samples, but the problem is that these data will then only be available after the well is completed. These triaxial data are then used in conjunction with logs through correlations when planning the next well. This paper describes a decision support system for well planning, and follow-up of well integrity (safe operating mud weight window) during drilling operations. The system, is a Windows based modular system, using Internet as the communication network for data exchange and technological support. One module in the system is the inverted ROP module which back calculates the compressive strength of the formation while drilling. This in-situ compressive strength will be used dynamically to calibrate the in-situ stresses which in conjunction with knowledge of the overburden and pore pressure, and the wellbore survey dictates the collapse and fracture pressure for the wellbore instantaneously. If the well has already been drilled and the above information is available, the true vertical depth collapse and fracture gradient can be predicted as part of the post analysis or for pre-planning the next well in the area. Another approach to obtain the collapse and fracture gradient is from log data, where the rock compressive strength of the rock is estimated from the sonic travel times. After the compressive strengths as a function of depth are estimated the same approach as for the use of drilling data to obtain in-situ stresses is applied. Using such a decision support software system, data may also be more readily available for display instantaneously or for use in future planning, thus providing a more complete set of data for design prognosis. P. 253

Artificial Intelligence, bit model, collapse gradient, compressive strength, decision support system, drilling fluid chemistry, drilling fluid formulation, drilling fluid property, drilling fluid selection and formulation, drilling fluids and materials, equation, fluid loss control, gradient, information, instantaneous, North Sea Well, post analysis software, Reservoir Characterization, reservoir geomechanics, risk management, rock compressive strength, rock strength, safe mud weight predictor, safe mud weight window predictor, strength, strength value, structural geology, triaxial data, Upstream Oil & Gas, Wellbore Design, wellbore integrity

Country:

- North America > United States (0.68)
- Europe > United Kingdom > North Sea (0.35)
- Europe > Norway > North Sea (0.35)
- (2 more...)

SPE Disciplines:

Technology:

Herrick 2 states: "Temperature variation within ordinary limits does not have much effect on yield points." Lewis, Squires and Thompsons investigated one clay and found no marked change with temperature in either yield point or mobility. Reed 4 found that potassium bentonite showed a great increase of gelation with increase in temperature and qualitatively there occurred a marked increase in viscosity upon heating. The apparatus used in these experiments consisted of a Stormer viscosimeter,. Marsh funnel viscosimeter, rectangular and cylindrical shearometers, Mudwate hydrometer, and other minor laboratory equip-' ment. The various methods of measuring viscosity and gel strength have been extensively described by Evans and Reid. 5. EFFECT OF TEMPERATURE ON GEL STRENGTH At the time these experiments were performed, the Stormer viscosimeter was considered the most practical and accurate instrument for measuring gel strengths and therefore was employed.

chemical treatment, drilling fluid chemistry, drilling fluid formulation, drilling fluid property, drilling fluid selection and formulation, drilling fluids and materials, drilling mud, experiment, fluid loss control, Funnel Viscosity, gel strength, Gulf Coast Drilling Mud, initial gel strength, initial shear strength, instrument, low shear strength, operation, relative gel strength, shear strength, shearometer, Stormer viscosimeter, strength, untreated mud, Upstream Oil & Gas, viscosity

SPE Disciplines: Well Drilling > Drilling Fluids and Materials > Drilling fluid selection and formulation (chemistry, properties) (1.00)

Xu, Chengyuan (State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation, Southwest Petroleum University) | Kang, Yili (State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation, Southwest Petroleum University) | Li, Daqi (Sinopec Research Institute of Petroleum Engineering) | You, Zhenjiang (Australian School of Petroleum, University of Adelaide) | Luo, Yaohua (Geomathematics Key Laboratory of Sichuan Province, Chengdu University of Technology)

Abstract Drill-in fluid loss is the most important cause of formation damage during drill-in process in fractured tight reservoirs. Lost circulation material (LCM) addition into drill-in fluid is the most popular technique for loss control. However, traditional LCM selection is mainly performed by trial-and-error method, due to lack of mathematical models. The present work aims at filling this gap, by developing a new mathematical model to characterize the performance of drill-in fluid loss control using LCM during drill-in process of fractured tight reservoirs. Plugging zone strength and fracture propagation pressure are the two main factors affecting drill-in fluid loss control. The developed mathematical model consists of two sub-models, i.e., the plugging zone strength model and the fracture propagation pressure model. Explicit formulae are obtained for LCM selection based on the proposed model, in order to control drill-in fluid loss and prevent formation damage. Laboratory tests on loss control effect by different types and concentrations of LCMs are performed. Plugging pressure and total loss volume are measured and compared with modeling results. Effects of LCM mechanical and geometric properties on loss control performance are analyzed, for optimal fracture plugging and propagation control. Different combinations of acid-soluble rigid particles, fibers and elastic particles are tested in order to generate a synergy effect for drill-in fluid loss control. The derived model is validated by laboratory data.

doi: 10.2118/182266-MS

SPE-182266-MS

annular pressure drilling, Artificial Intelligence, drilling fluid chemistry, drilling fluid formulation, drilling fluid management & disposal, drilling fluid property, drilling fluid selection and formulation, drilling fluids and materials, elastic particle, Engineering, fiber, fluid loss control, fracture, fracture propagation pressure, fracture surface, friction angle, frictional strength, logic & formal reasoning, logic programming, particle, permeability, shear strength, strength, stress-intensity factor, Upstream Oil & Gas, well control, zone permeability, zone porosity, zone strength, zone strength failure

Country:

- Asia > China (0.94)
- North America (0.67)

SPE Disciplines:

Technology: Information Technology > Artificial Intelligence > Representation & Reasoning > Logic & Formal Reasoning (1.00)

Summary

A computer program was developed for the simultaneous selection of a roller-cutter bit, bit bearing, weight on bit (WOB), and drillstring rotation that minimizes drilling cost per foot for a single bit run. Two drilling models were tested with data from five wells located offshore Alagoas, Brazil. Results show that the rate of penetration (ROP) of the fifth well can be predicted with coefficients calculated from the four previous wells, resulting in cost savings. previous wells, resulting in cost savings. Introduction

Several authors have studied optimization of single bit runs to develop techniques to predict the best combination of rotary speed and WOB to minimize cost per foot drilled. Minimizing cost per foot drilled is of particular interest to offshore operations because rig time is very costly and savings can be significant. Such cost savings can occur only if the drilling model correctly predicts bit performance throughout the bit's entire life. Thus, before any performance throughout the bit's entire life. Thus, before any optimization is performed, bit-performance predictions must be checked and the reasons for prediction failures must be studied. Some drilling models work by predicting the effects of various drilling parameters on ROP and back-calculate their "constants" whenever history data are available. Therefore, good data recording is essential for meaningful predictions. The objective of this study was to check the predictions of two drilling models with field data and to develop a practical method that can be applied in the field for selecting a roller-cutter bit, bit bearing, WOB, and drillstring rotation to minimize drilling costs (within certain constraints).

Drilling Models

Two drilling models were considered in this study.

Bourgoyne and Young's Model. Bourgoyne and Young considered the effect on ROP of the formation strength, compaction, differential pressure, WOB, rotary speed, tooth wear, and bit hydraulics and pressure, WOB, rotary speed, tooth wear, and bit hydraulics and used a multiple-regression technique to calculate the constants of the model. Their model is given by

R = exp ,......................(1) D

where a1 = effect of formation strength and a2x2 and a3x3 = effects of compaction, with

x2 = 10,000 -D....................................(2)

0.69 and x3 = D (gp-9.0)................................(3)

a4x4 = effect of pressure differential, with

x4 = D(gp -pc),...................................(4)

and a5x5 = effect of WOB and bit diameter, with

x5 = 1n{[(W/d)-(W/d)t]/[4.0-(W/d)t]}..............(5)

where W/d=WOB per inch of bit diameter, 1,000 lbf/in., and (W/d)t=threshold WOB at which the bit begins to drill, 1,000 lbf/in. Note that Eq. 5 has not been verified by published experimental data. a6x6 = effect of rotary speed, with

x6 = 1n(v/100);,..................................(6)

a7x7 = effect of tooth wear, with

x7 = -h,..........................................(7)

and a8x8 = effect of bit hydraulics, with

x8 = 1n(Fj/1,000).................................(8)

Tooth wear rate can be calculated by

H3 v H1 (W/d)max- 1+(H2/2) Rh = ,..(9) H 100 (W/d)max-(W/d) 1+H2h

where (W/d)max = weight diameter ratio for instantaneous tooth failure, 1,000 lbf/in. Bearing wear rate can be calculated by

1 v b1 W b2 Rb = ....................(10) B 100 4d

Authors' Model. This model was developed by using dimensional analysis and by considering the effects on ROP of rock compressive strength, formation strength, compaction, differential pressure, WOB, rotary speed, tooth wear (the same as Bourgoyne pressure, WOB, rotary speed, tooth wear (the same as Bourgoyne and Young's model), and bit hydraulics.

Artificial Intelligence, bit run, bit selection, compressive strength, computerized selection, diameter, Drilling Cost, drilling model, exponent, lithology, prediction, Reservoir Characterization, ROP, rotary speed, rotation, shear travel time, SPE Drilling Engineering, strength, structural geology, Upstream Oil & Gas, WOB

SPE Disciplines:

Thank you!