SAGD Subcool Control with Smart Injection Wells

Abstract

Steam-Assisted Gravity Drainage (SAGD) is a widely used in situ recovery process for heavy oil and bitumen reservoirs. The performance of the SAGD process is tied with growth of the steam chamber which in turn depends on uniform steam delivery along well length and the underlying geology and fluid properties in the near wellbore region. If the reservoir has poor injectivity due to poor underlying geology oil production suffers. This can be avoided in by examining the interwell subcool. The subcool is the temperature difference between the injected steam and produced fluids. In this study, Proportional-Integral-Derivative (PID) feedback control has been employed to control inflow control valves settings to promote subcool to a target value. This control strategy is examined by using a PID algorithm to control SAGD in a detailed three-dimensional heterogeneous reservoir model with properties typical of an Athabasca bitumen reservoir. Specifically, the SAGD injector is divided into six intervals each with its own steam injection pressure. The interwell subcool is calculated and the PID feedback control algorithm is used to direct the subcool to a target value by changing the steam injection pressure in each well interval. The results show that this control method can be used to enhance uniform steam chamber growth and ultimately more oil production with less steam injection. The key benefit of dynamic well control is that the injection strategy is adjusted dynamically to fit the underlying geological and fluid compositional heterogeneity to obtain improved steam conformance along the wellpair. This implies that potentially a priori detailed knowledge of the geological and fluid compositional heterogeneity may not be as critical for well placement for uniform steam conformance. Introduction There are two main objectives of a thermal recovery process for heavy oil and bitumen reservoirs: first, mobilize the oil - this is often done by using high pressure steam, and second, move the mobilized oil to a production well. If one of these two objectives is not met, then the recovery process will not be technically successful. For a steam-based thermal recovery process, the first objective equates to delivering the latent heat of the steam to the oil phase in the oil sand. In general, the more efficient the heat transfer from steam to oil, the more productive and economic is the recovery process. Thus, a key goal of a steam-based recovery process is to control energy delivery within the reservoir. However, thermal production from heavy oil and bitumen reservoirs is difficult because subsurface processes such as steam flow and heat transfer in a system with both geological and fluid compositional heterogeneity tend to be difficult to control. For example, in steam-based recovery technologies such as Steam-Assisted Gravity Drainage (SAGD), displayed in cross-section in Figure 1, the steam chamber that develops in the reservoir may not be uniform along the length of the wells. For example, Figure 2 displays interpreted seismic data of the heated zones in Clearwater bitumen reservoir at the end of the steam injection period in the first three cycles of horizontal well CSS (Imperial Oil, 2006). The images reveal that steam injection is not uniform along the length of the wells. Thermocouple data also reveal that heat transfer in the reservoir is non-uniform and is controlled by the heterogeneity of the underlying geology and fluid composition (ConocoPhillips, 2008). This implies that despite the desire for uniform steam delivery within the reservoir, this is most likely not achieved in the majority of steam injection wells. Recently, an analysis by Zhang et al. (2005) of 4D and crosswell seismic and production data have shown that steam chamber growth and oil recovery are strongly influenced by reservoir geology. Steam chamber growth is especially affected by the presence of low permeability facies in the vicinity of the SAGD wellpair. Furthermore, Gotawala and Gates (2009) have demonstrated that there is a direct link between permeability heterogeneity and the evolution of a SAGD steam chamber.

Practical Control of SAGD Wells With Dual-Tubing Strings

“…Weaver et al (2005) also looked at both the installation and application of permanent downhole optical pressure and distributed-temperature-sensing (DTS) gauges in a deepwater Gulf of Mexico environment where the temperatures (approximately 52 C) were not as high as those encountered in a typical SAGD process (approximately 250 C). Gotawala and Gates (2009) looked at the use of a PID feedback control to control inflow-control-valve settings to promote subcool to be within a specified value. They applied a stochastic model of shale distribution and examined two flow regions, near-well and above-well, to decouple the effects of heterogeneity.…”

Benefits of Passive Inflow Control Devices in a SAGD Completion

“…In practice, this phenomenon has been partially controlled by running an inner tubing string to the toe of the well, creating a second, independent steam injection point. With injection at the heel and the toe of the well, more HO-B is mobilized effectively, but a 'barbell' steam chamber develops, reflecting how this well design has little capability to control fluid injection into targeted regions between these two fixed points (Gotawala and Gates 2009). …”

Model-Predictive-Control (MPC) of Steam Trap Subcool in Steam-Assisted Gravity Drainage (SAGD)

“…Therefore, unconventional techniques of oil recovery, such as thermal recovery of bitumen, have proved to be the most efficient. Thermal recovery techniques pertaining to heavy oil recovery can broadly be said to consist of two steps; the first involves heating the bitumen such that its viscosity falls from millions of cP to around 20 cP, thus mobilizing it and the second involves directing the mobile bitumen towards a production well (Gotawala et al (2009) and Gotawala et al (2012)). The most current and widely used technologies for the fluidization of bitumen are Cyclic Steam Stimulation (CSS) and Steam-Assisted Gravity Drainage (SAGD) (Butler (1994)), both of which use steam.…”

“…Although preliminary research and some field implementations have suggested that PID control based techniques is an improvement over the open loop manual control (Gotawala et al (2009) and Gotawala et al (2012)), there still remains the inability of PID controllers to handle constraints and calculate optimal control gains for the MV. The process of achieving the desired subcool by increasing T P through injection of superheated steam and mobilizing the bitumen has a high time constant (τ ).…”