No abstract

Since the latest downturn of the Oil and Gas (O&G) sector, productivity growth has become the priority, which led to the adoption of new key technologies for better performance and lower cost. As part of the 4th Industrial Revolution, these technologies are taking the O&G by storm, forcing hardware manufactures to innovate quickly and create devices with more capabilities. What was impossible a few short years ago is reality today. Managed pressure drilling (MPD) systems are taking advantage of this development by implementing advanced data analytics and hydraulic modeling in programable automation controllers. This approach has several benefits, reliability and full real-time deterministic operation being the major ones. Previously, controlling bottomhole pressure (BHP), the required modeling and analysis were executed in a networked computer at a different physical location. Now it is executed locally in a fully deterministic control system. This includes data acquisition (DAQ), signal conditioning, well bore modeling, analytics and data visualization. This approach mitigates the risk of connectivity issues and makes the system more robust in case of network failure. Furthermore, it simplifies the deployment of the system, as all critical functions are at a single physical location. Another major benefit is an increased resolution of the model in areas of specific interest. For example, a zoom in on the openhole section and a lower resolution in the cased hole section. The flexibility of this approach manifests itself on the deployment side. This system can be installed in several different scenarios, for example as a full system with controls and manifolds, or as a control system only, and integrated into existing rig MPD manifolds and drilling controls. This level of integration creates robust and flexible systems while reducing downtime during operations. This new approach includes the integration of cyber secure communications, an upcoming standard in the O&G industry. The system communicates via the open platform communications unified architecture (OPC UA) protocol – recommended by the Cyber Security community. This new platform integrates advanced data analytics and hydraulic modeling in real-time, coupled with cyber security features via OPC UA protocol directed to automation controllers for managed pressure drilling applications.

A good primary cementation requires careful selection of centralizers and their placement on the string. The centralizer placement algorithm described in the API-10D was corrected and put into a computer program. The standoff value is calculated based on the actual borehole geometry, string data and centralizer performance. The model was enhanced by a newly developed drag force simulation taking the centralizer running force into account. Additionally, the prediction of the expected torque values for rotating liner applications is included. Introduction The key factor for a successful cementation job is the replacement of the mud in the wellbore by the cement slurry. Hydraulic considerations call for the need of a good centralization of the string for all sections in which a good cementation is required. Centralizers have been used for decades to fulfill this job. Throughout the past couple of years, more and more designs of highly inclined, including horizontal, wells incorporate cemented production casing and liner sections. In these cases, the optimum placement of centralizers is achieved by balancing between a high standoff ratio and low drag forces. A mathematical simulation model is used to calculate the optimum spacing of centralizers to obtain the best standoff at a given borehole location. This model takes into account relevant factors, such as:–the lateral force at any given location based on borehole geometry, buoyed string weights and tension forces–the centralizer's reaction to these forces, based on test data for each pipe size/hole size combination–the sag between centralizers based on the elasticity of the pipe and a three-dimensional vector analysis of the weight and tension components. This mathematical model is associated with a torque and drag analysis, utilizing the known running forces of the centralizers and the friction factors that depend on the mud type. This analysis is important in order to evaluate whether the desired centralizer spacing can be run or rotated without creating problems due to high drag forces, or damage to the pipe connections. The equations upon which these models are based and the computer algorithms used are described in this paper. 2 Maximize Standoff and Minimize Drag Various models have been described in the literature to calculate the centralizer placement. The criteria to select a centralizer pattern should be not only the achieved centralization but, especially in highly inclined wells, the ability to move the string. Thus, drag and torque calculation should be a part of the centralizer placement calculation. It is important to understand that all mathematical equations and relationships regarding centralizer placement describe a model situation only. The actual standoff in a borehole depends on many different factors. There is no method of actually looking into the well and no tool like a "Standoff-Logging-Tool" to provide this information directly. P. 153

The optimum placement of centralizers does not only incorporate the achievable standoff. In highly inclined and horizontal well situations the consideration of friction forces is relevant to assure, that a string can be run safely to bottom. The set of equations described in SPE 21282 and the API Specification 10D of January 1, 1995 were revised and corrected. The enhanced equations and a newly developed algorithms to calculate torque and drag were described in IADC/SPE 36382. A new software based on these mathematical models was developed and installed strategically in over 100 locations worldwide. The software was programmed in "C" and runs under Windows TM. It allows the interactive simulation of different centralizer pattern and its effect on the expected hook load and torque. Simulated data compared with real field situations showed excellent compliance. Introduction Careful planning and engineering are critical for the success of the cementing job with proper centralization necessary for good mud removal (Fig. 1). The optimum placement of centralizers does not only incorporate the achievable standoff. In highly inclined and horizontal well situations the consideration of friction forces is relevant to assure, that a string can be run safely to bottom. A new mathematical simulation model based on revised API-10D formulas is used to calculate the optimum spacing of centralizers to obtain the best standoff at a given borehole location. The revised mathematical model was described in [3]. It is associated with a torque and drag analysis, taking into account the known running forces of the centralizers as well as the friction factors depending on the mud type. This analysis is important in order to evaluate whether the desired centralizer spacing can be run or rotated without delays, problems due to high drag forces, or without damaging the pipe connections. A powerful computer program (CentraPro PlusTM) has been developed which runs under a Windows operating system. It makes use of the standard conventions defined for these systems. Users familiar with a Windows environment are able to run the software intuitively based on those conventions. 2 Technical and Mathematical Background Centralizers are used to optimize the hydraulic flow during the cementation process and to minimize drag forces caused for example by differential sticking. Two types of centralizers are known: bow-type centralizers and rigid centralizers. The bow-type centralizers are made of two collars and a number of metal springs attached to them. The characteristics of the metal springs is known, so that the deflection of a centralizer in a given load situation can be determined. Rigid centralizers have a fixed outer diameter which is basically independent of lateral forces. The decision which type of centralizer to use depends on several factors such as the borehole condition (see table 1). During the cementation process, the mud used to drill the well has to be replaced by cement.

Managed Pressure Drilling (MPD) usage is in increasing demand in both offshore and onshore market – especially for wells with narrow drilling pressure margins. Maintaining desired constant Bottom-Hole Pressure (BHP) within given range during MPD operation requires accurate hydraulic modeling for frictional pressure losses. Surface BackPressure (SBP) and StandPipe Pressure (SPP) data, measured in real-time, allows the calibration of frictional pressure losses and an estimation of BHP more accurately. An estimation of SPP is calculated based on integration from the measured SBP back to the inlet in the hydraulics model. The pressure loss increment in the hydraulics model is calculated based on a difference between the measured and estimated SPP value. The system parameters are monitored during drilling and could be adjusted based on the hydraulics model corrected for the pressure loss. The frictional pressure loss due to the drilling pipe rotation could be calibrated by analyzing a difference between the measured and estimated SPP for two operations: without rotation and with rotation. This work underscores the development and usage of advanced hydraulics modeling to accurately calculate BHP during MPD operations. In current hydraulics, mathematical models have been developed to account for various operating parameters, wellbore geometry and fluid properties. This is an enabler to estimate pressure, density, rheology, temperature, and cuttings distributions within the wellbore, drill string and surface equipment more precisely. The frictional pressure loss model includes rotational pressure losses, local pressure losses at special tools and tool joints. This model is based on empirical sub-models which require experimental verification for accuracy. The results of the BHP, calculated with a calibrated hydraulic model, are compared with corresponding results calculated with the uncalibrated model. The use of the pressure calibration for the hydraulics model allows estimating BHP with less error. Recommendations are presented for the accurate mathematical modeling and their applicability and limitations for successful MPD operations.