Presented is a new methodology for selecting rotary drilling bits in an oil or gas well. Currently, bits are selected based on the performance of similar bits at offset wells. Parameters affecting a bit performance have a complex pattern. The relationship between formation properties, drilling fluid characteristics, bit design, and operational parameters in these patterns are not easily understood. For a given field, studied were variables such as bit size, weight on bit, rotary speed, pump rate, drilled interval, and bit type. A three-layer artificial neural network was designed and trained with field data. This method incorporates computational intelligence to define the relationship between the variables. Further, it can be used to estimate other drilling parameters. The results indicate that the back propagation achitecture with two hidden slabs is the most effective neural network design for predicting the optimum bit type. With the given data sets, this new model successfully predicted the bit types for several fields. For different data sets used in this study, the correlation coefficients for the predicted and field used bit types ranged between 0.857 and 0.975. Introduction Drilling engineers deal with many challenges before and during drilling a new well even in a known area. There are many parameters related to hardware and daily operations that are planned and also modified as the drilling progress. Bit selection is one of the important parameters for planning and designing a new oil or gas well. The selection of a proper bit is a difficult task since the factors affecting the bit performance are complex relationships between formation properties, bit hardware design, and operational parameters. Formation properties such as hardness and drillability are based on the interval drilled and they can not be changed. Additionally, they are not measured prior to drilling and they have to be estimated from geophysical surveys or from offset well records. Several investigators conducted research to estimate the bit behavior based on operational parameters and data recorded at offset wells.1–10 Different models were developed by these researchers with assumptions limiting the applicability of their model. In general, the data from offset wells are used by the engineer to select a proper bit. The bit type with the highest rate of penetration or minimum cost per foot are the two commonly used criterias for selecting the bit for the next interval. Additional factors such as hydraulics, formation hardness, bit design, and operational parameters are also considered in the selection process. Due to the number of variables considered, the selection process is a trial and error procedure. In many cases, this approach can ignore some of the important parameters affecting the bit performance and can not guarantee the selection of the optimum bit type. When sufficient data exists, the use of neural networks are demonstrated to identify complex relationships.11–13 The increased availability of down hole measuring tools resulted in data sets with many recorded variables related to the drilling process. As a result, different neural networks can be designed for a region or a field to predict unknown parameters. Approach In this study, a new methodology is introduced to select rotary drilling bits. This approach uses a three-layer feed forward neural network to select bit types. Several neural network models were used to determine the complex relationship between formation and bit properties together with operating parameters.

In this study, a new method is presented to select rotary drilling bits in an oil or gas well. Planning a new oil or gas well requires selection of hardware components such as casings and bits. Additional information needed in the design and cost estimation of the new well is the performance of bits under varying operating conditions. The bit selection is an important task for the drilling engineer. The offset well information can provide valuable information. However, the data from the offset well contains a complex relationship between operational parameters, formation properties, drilling fluid properties and hydraulics, and bit design. The bits are selected based on the performance of similar bits in offset wells. Parameters affecting a bit performance have a complex pattern. The relationships between formation properties, drilling fluid characteristics, bit design, and operational parameters in these patterns are not easily understood. In this approach, different variables such as bit type, rotary speed, weight on bit, pump circulation rate, drilled interval, and penetration rate are studied over a range of values. Two data sets from different fields located in the Middle East were used. Several neural networks were designed with three-layer back propagation architecture to select the bits for the next drilling interval. Additional neural networks were designed for predicting the cost of bits. With the given data sets, these new models successfully predicted the bit types and cost-per-foot values for several fields. For the different data sets used in this study, the correlation coefficients for the predicted and field observed values ranged between 0.831 and 0.968. Introduction Drilling engineers deal with many challenges before and during drilling a new well. There are many parameters related to hardware and daily operations that are planned and modified as the drilling progress. Bit selection is one of the most important parameters for planning and designing a new oil or gas well. The selection of a proper bit is a difficult task since the factors affecting the bit performance are complex relationships between formation properties, bit hardware design, and operational parameters. Formation properties such as hardness and drillability are based on the interval drilled and they cannot be changed. Additionally, they are not measured prior to drilling and they have to be estimated from geophysical surveys or from offset well records. Several investigators conducted research to estimate the bit behavior based on operational parameters and data recorded at offset wells.1–10 Different models were developed by these researchers with assumptions limiting the applicability of their model. The conventional approach in bit selection is to use performance data from offset wells. The two commonly used criteria for selecting the bit for the next interval is the bit type with the highest rate of penetration or the bit with minimum cost per foot. In addition, factors such as hydraulics, formation hardness, bit design, and operational parameters are considered in the selection process. Due to the number of variables considered, the selection process is a trial and error procedure. In many cases, this approach can ignore some of the important parameters affecting the bit performance and cannot guarantee selection of the optimum bit type. Recently, the neural networks are used to identify complex relationships when sufficient data exists.11–14 The increased availability of down hole measuring tools has resulted in data sets with many recorded variables related to drilling process. As a result, different neural networks can be designed for a region or a field to predict unknown parameters.

Constructing the 12¼" direction hole section through approximately 3000ft of difficult interbedded lithologies (Mutriba-Lower Burgan) in northern Kuwait presents a number of distinct challenges. In the upper portion of the hole section, a PDC bit must penetrate medium to hard carbonate and shale formations with compressive strength ranging between 9-12 kpsi with peaks up to 30kpsi. Next, a challenging abrasive sand with compressive strength between 6-9kpsi requires an RSS/PDC assembly to reach TD. The operator experimented with several different bit designs attempting to efficiently achieve directional objectives and improve borehole quality with limited success. Issues with baseline designs included lack of cutting structure durability and low ROP.To accomplish the operator's objectives, the engineering team analyzed all relevant offset data and concluded an existing 12¼" six-bladed bit with 16-mm cutters would serve as the starting point for an optimization effort. The bit's design data was fed into an FEA-based modeling system. Formation characterization software was then utilized to select the appropriate rock samples to simulate the field formations in the laboratory. Multiple simulations were run to determine the best combination of technologies to achieve the objectives. A new 12¼" directional design (616-type) would include premium cutters that can withstand impact in the interbedded carbonate/shale section and remain sharp while drilling the lower sand formations to TD. The bit also features a torque limiting feature in the blade top and TSP inserts in gauge to ensure hole quality. Next, a series of simulations were preformed to observe how different RPM and WOB values would affect vibration and torque levels. The results were plotted to create a smooth drilling parameter window to maximize the new bit's ROP potential.The new bit design was run on RSS with PDM assist and set a new ROP record of 46 ft/hr, 39% faster than the best offset of 33 ft/hr and 68% higher than the five-well offset average (27.3 ft/hr). The bit met all directional objectives (5-6°DLS) and was pulled in excellent dull condition (0-1-WT). The authors will discuss the bit design and selection process in addition to the HTHP cutter technology which saved the operator 2.5 days of rig-time and associated costs.

Greater Burgan Field has been producing from more than 70 years and brings many challenges with the maturing field. During the last decade almost all the new wells drilled were completed as dual completions to simultaneously produce from two different formations in Burgan. As the workover rig activity tremendously increased with the increased number of new wells per year, it has been a constant endeavor to carry out maximum well intervention jobs possibly by rig less operations to reduce rig related OPEX and well production down time during rig work over. One of the biggest long term challenges was to carry out well interventions in dual completions short tubing, which was used to be considered inaccessible due to presence of tubing long in the well bore. In view of above, most of the intended well interventions in tubing short were normally used to be proclaimed as work over candidate thereby increasing load on limited work over rigs. First time in Burgan a dual completion well was carefully identified to carry out water shut off job by chemical application and adding perforation by oriented perforation gun in short tubing. The tubing short was producing from BGSU reservoir and became ceased to flow due to lateral water encroachment which was evident from the recent PNC log. The formation layer below BGSM was still having oil potential and was intended to produce after shutting off the upper layer BGSU. The Tubing Long was producing from deeper reservoir BGSL around 1200 BLPD with 50% W/C. This paper elaborates the procedures adopted during the execution of the job and selection process of the chemical squeeze technology. The best suitable chemical technology was selected after carrying out a careful & detailed approach towards different vendors. The step by step detailed program was prepared to squeeze the chemical in to existing perforations, cleaning to the desired depth, testing the squeezed perforations and adding new perforations below with oriented perforations gun. The results were quite encouraging with sustainable oil gain of 1400 BOPD and the detailed cost/pay back analysis indicated the payback period of only 1.5 days of production. This successful job has opened a new era in the history of Burgan field and many other water shut off jobs may be successfully carried out without the need of rig work over. This not only will reduce the work load on rig work overs and save rig cost significantly but will also allow to increase the well productivity with a very short period of well down time.

In North Kuwait, formation evaluation in horizontal/highly deviated wells typically requires the use of Logging While Drilling (LWD) technology. In this paper, we will discuss how for the first time in Kuwait a state-of-the-art wireline open hole tractor has been successfully used to convey an advanced wireline pressure measurement in a horizontal well. Two wells will be discussed in this paper, the first was a short radius horizontal side track in the tight carbonate formation while the second is highly deviated well across sand/shale layers. Traditionally in horizontal wells, the pressure measurement is either run on drill pipe or LWD. Moreover, the formation tightness posed another challenge, as stabilized formation pressures can be difficult to achieve. To address the challenge of formation tightness and save rig time, a fast wireline pretest measurement tool allowing dynamic control of the pretest system would be conveyed on wireline using Open Hole tractor. A job simulation was conducted, based on the friction force and tool weight, to ensure the ability of conveying the tractor to the required depth in addition to the ultimate tractor drives number with tandem1 configuration. After gaining confidence and experience from the first well tractoring across tight consolidated carbonate, it was decided to go for open hole tractoring across less consolidated sand/shale layers in a highly deviated well (maximum deviation of 85deg). Both jobs were successfully executed as per plan. The tool was conveyed smoothly across the tight carbonate reservoir to the target depth of 10,030 ft MD at an average tractoring speed of 1800 ft/hr. The job was concluded with a cumulative tractoring footage of 3200 ft and an operating time of 12.5 hours, which resulted in more than 30 hours of rig time savings compared to other alternatives. The requested pressure program was achieved. Due to the pressure tool's low rate pretest capability, in addition to the flexible volume and pretest time options; stabilized formation pressure data could be acquired2,3. The precise depth control resulting from this conveyance method enabled accurate pressure profiling. Similarly, the second job was conveyed smoothly to the target depth. The combination of the advanced pressure measurement with a state-of-the-art open hole tractor1 conveyance method has proved to be an efficient and effective alternative to LWD and drill pipe in a challenging horizontal and highly deviated environments (the first job of its kind in the region). It resulted in acquiring accurate pressure data with significant rig time saving, and proved to be a cost effective solution contributing to an overall reduction in the well cost per barrel.