LIFT PROCESS

FIELD OF INVENTION

The invention relates to a method and apparatus for optimal tracking control of an artificial gas lift process. The method and apparatus is comprised of an algorithm and an apparatus for generation of an optimal trajectory defined in terms of set points for specifically selected process variables.

BACKGROUND OF THE INVENTION

Artificial gas lift is a widely-used technique in oil production, which helps to increase oil flow rates when the pressure in the reservoir is not high enough to support acceptable flow rate.

However, this process is often faced with instability characterized by regular or irregular cyclic variations in mass, pressure and flow rates, which results in production loss and sometimes chaotic unstable production behavior. This oscillation is often referred as casing-heading phenomenon, which has been a challenging problem for a long time.

To solve this problem, currently available techniques include reducing the injection valve port size, wellhead choking and increasing gas injection. These remediation methods are widely used in practical situations with advantage of easy operation.

However, the main drawbacks of these manual control approaches are economic cost and loss of optimality in oil production. Also, it seems impossible to assure stability for a gas lift well in the whole production life by manual control, because it does not have adaptability of original design to new situations if reservoir condition changes.

OBJECT OF THE INVENTION

It is an object of the current invention to address the aforementioned problems, at least to a degree, by providing a method and apparatus for automatic control, which may be easily incorporated into gas lift processes. It is a further object of the invention to provide a cost- effective alternative, to a degree, to existing gas lift processes by opening both the gas injection choke and the oil production choke to stabilize the casing-heading instability without losing optimality by using feedback control.

SUMMARY OF THE INVENTION

In accordance with the invention there is provided an optimal trajectory generation algorithm for driving an automatic gas lift system from any unstable operating mode to an optimal stable operating mode which includes; a tracking controller for set-point tracking and a nonlinear state estimation method for process variable estimations.

In accordance with the invention there is a method provided for achieving transition of an initial operating mode of an artificial gas lift process to an optimal operating mode and maintaining said optimal operating mode under variable parameters of the process, comprising a mathematical dynamic model of a well; an optimal trajectory generation algorithm; a model- based tracking control algorithm, and a nonlinear state estimation method, characterized in that both a gas injection choke opening and an oil production choke opening are manipulated per said model-based tracking control algorithm, wherein said model-based tracking control algorithm is designed based on a third order dynamic model, where process variables used in control design are estimated by said nonlinear state estimation method.

The invention provides further in that said transition of an initial operating mode of the artificial gas lift process to an optimal operating mode is automatically generated by the optimal trajectory generation algorithm.

The invention provides further in that said optimal trajectory generation algorithm is based on stability analysis of the artificial gas lift process.

The invention provides further in that optimal trajectory of said optimal trajectory generation algorithm is plotted on a stability map.

The invention provides further in that optimal trajectory of said optimal trajectory generation algorithm is evaluated based on forecasted process transition time.

The invention provides further still in that optimal trajectory of said optimal trajectory generation algorithm is produced in terms of control commands, which are commands for gas injection choke and oil production choke.7. The method according to claim 6, characterized in that set- point values of down-hole controlled variables are produced from said control commands.

The invention provides further in that said down-hole control variables are any combination of the following three variables: down-hole tubing pressure, down-hole annulus pressure, down hole injection gas flow rate.

The invention provides further in that the stability map is developed based on the study of open- loop simulation for artificial gas lift process by manipulating both gas injection choke and oil production choke.

The invention provides even further still in that the stability map is produced as a function of openings of both gas injection choke and oil production choke manipulated continuously between 0% to 100%.

The invention provides further still in that the stability of the artificial gas lift process is related to the openings of both gas injection choke and oil production choke, where three different regions can be divided according to different choke openings.

The invention provides further in that the model-based tracking control algorithm is raised based on ramp limits of set-point values.

The invention provides further in that control strategy comprises controlling the estimation of bottom-hole pressures or injection gas flow rate through topside measurements, including a model-based tracking control algorithm for bottom pressures or injection gas flow rate combined with a nonlinear state estimation method.

The invention provides further still for the bottom-hole pressures and injection gas flow rate can be estimated by a nonlinear state estimation method from the topside measurements obtained by sensors.

The invention provides further for the nonlinear state estimation method is based on a third- order dynamic model of artificial gas lift processes.

The invention provides further wherein said third-order dynamic model of artificial gas lift processes is based on mass balance of gas in the annulus, mass balance of gas in the tubing and mass balance of oil in the tubing. The invention provides further still for said model-based tracking control algorithm is regarded as multi-input and multi-output control system, with the openings of both gas injection choke and oil production choke being control commands, and any combination of two variables from the injection gas flow rate and bottom-hole pressures in an annulus and a tubing being process variables.

In accordance with the invention there is provided an apparatus for achieving transition of an initial operating mode of an artificial gas lift process to an optimal operating mode and maintaining said optimal operating mode under variable parameters of the process, said artificial gas lift well comprising an annulus with at least one gas injection choke and oil production tubing including at least one oil production choke, characterized in that gas injection chokes and oil production chokes are continuously controlled by a model-based tracking control algorithm, and bottom pressures are estimated by nonlinear state estimation method or measured by particular sensors.

The invention provides further for achieving transition of an initial operating mode of an artificial gas lift process to an optimal operating mode and maintaining said optimal operating mode under variable parameters of the process where that said model-based tracking control algorithm of artificial gas lift process comprises an optimal trajectory generation algorithm (module) and a tracking controller.

The invention provides further still in that said stability of the artificial gas lift process is analyzed through open-loop simulations by manipulating both gas injection choke and oil production choke continuously between 0% to 100%, to develop a stability map.

The invention provides further by different operating modes of artificial gas lift processes by manipulating gas injection choke and oil production choke, between; a stable operating mode (region), an unstable operating mode (region) and a none operation mode (region).

The invention provides further where the stability analysis drives the closed-loop system from initial operating mode in the unstable region into optimal operating mode in the stable region with least process transition time.

The invention provides further still where the tracking controller includes; measurement or estimation of process variables; bottom pressures or gas injection flow rate.

The invention provides further where measurements of bottom pressures or gas injection flow rate are inputs.

The invention provides further where manipulation of the gas injection choke controls the gas flow rate into the annulus and manipulation of the oil production choke controls the oil flow out tubing.

The invention provides further still where the bottom pressures and gas injection flow rate are estimated by using topside measurements, or are measured by sensors.

In accordance with the invention there is provided the apparatus for achieving transition of an initial operating mode of an artificial gas lift process to an optimal operating mode and maintaining said optimal operating mode under variable parameters of the process where the nonlinear state estimation method uses topside measurements of artificial gas lift process as inputs.

The invention provides further in that the topside measurements of artificial gas lift process are measured by particular sensors or devices. The stability analysis of artificial gas lift process is studied through open-loop study by continuously changing the openings of both gas injection choke and oil production choke from 0 to 100%, where a stability map is developed. The concept of optimal trajectory is established by introducing the cost function of minimal transition time. The set-point values of controlled variables is a sequence of equilibrium points, which is calculated offline and selected by optimal trajectory generation algorithm.

The tracking controller is designed based on state estimation of process variables, where a nonlinear state estimation method is utilized. This tracking controller is used to realize online closed-loop control with a closed-form optimal control law.

BRIEF DESCRIPTION OF THE DRAWINGS

These and other features of the invention will become more apparent by the following description of the embodiment, which is made by way of example, with reference to the accompanying drawings in which:

Figure 1 shows a schematic of an artificial gas lift process, along with invented control strategy;

Figure 2 shows the invented control structure, which is corresponding to Figure

1

Figure 3 shows the phenomenon of casing-heading oscillations of flow;

Figure 4 shows the scheme of open-loop simulations for artificial gas lift process, where opening of both gas injection choke and oil production choke are inputs;

Figure 5 shows a stability map, which is the stability analysis of artificial gas lift process with different choke openings;

Figure 6 shows the relation between gas injection flow rate and different choke openings;

Figure 7 shows the relation between oil production flow rate and different choke openings;

Figure 8 shows an illustration of selected optimal point on stability map;

Figure 9 shows an example of arbitrary straight trajectory on stability map; Figure 10 shows the set-point values for controlled variables and corresponding ramp limits;

Figure 11 shows the comparison between straight trajectory and optimal trajectory plotted on stability map; and

Figure 12 shows the comparison of transition time between straight trajectory and optimal trajectory. DETAILED DESCRIPTION OF THE INVENTION

Referring now to Figure 1 , which shows a schematic of an artificial gas lift process. The gas lift process has two main circulation channels, one is the annulus 1 which is the channel for compressed gas and the other is tubing 2, which is the channel for the mixture (gas and liquid). The gas 9 is injected through the gas lift choke 4, which is also called surface injection valve, into the annulus. At the bottom of the annulus, there is a one-way valve called injection valve 3 and normally is an orifice, which is the connection between annulus and tubing. Gas flows through this orifice, into the tubing producing a mixture of gas and liquid 1 0. The injected gas from the annulus decreases the average density of oil in the tubing, resulting in lower hydrostatic pressure gradient. Flowever, the reduction of pressure leads to higher flow rate of oil from the reservoir 1 1 , which in turn increases the production. At the top of the tubing, a production choke 5 is placed to control the production flow from tubing into the next operation unit, for example, a separator.

An invented control strategy is also shown in Figure 1 , where a nonlinear state estimation method 6 is designed to estimate the bottom pressures in both the annulus and the tubing or gas injection flow rate by measuring the top pressures. The closed-loop system is a two-input and two-output system, where the openings of both a gas injection choke and an oil production choke are manipulated variables (MV), the estimation of bottom pressures in both annulus and tubing are controlled variables (CV). The controller used in a tracking controller, which is designed based on model predictive control principles.

Figure 2 shows the invented control structure, which is a control block diagram corresponding to Figure 1 . In order to suppress casing-heading instability using feedback control, a controller based on a simplified third order model, which can characterize the dynamics of artificial gas lift process, is designed. Further, a nonlinear state estimation method may be designed based on this third order model to estimate bottom pressures, where the inputs of observer come from real plant or other more accurate models. The casing-heading instability can be well suppressed if the bottom pressures can be controlled, because the cyclic oscillations result from the interactions between bottom pressure in annulus and that in tubing.

Figure 3 shows the typical phenomenon of oscillations in gas and oil flow, which is known as casing-heading instability. It is an open-loop simulation results at certain openings of gas injection choke and oil production choke.

Figure 4 shows the structure of open-loop simulation for artificial gas lift process, where both gas injection choke and oil production choke are continuously manipulated from 0 to 1 00%. Under different conditions of choke opening, the equilibrium points of bottom pressures can be calculated based on the simplified third order model, which is described by three first order ordinary differential equations. It is noted that the equilibrium points of bottom pressures are obtained offline, which will be used as set-point values in optimal trajectory control. In addition, the stability of the artificial gas lift process can be determined by observing whether the flow rates of oil or gas experience periodic oscillations.

Figure 5 shows a stability map, which is drawn based on the open-loop simulations described in Figure 4. The artificial gas lift process is divided into three operation modes, that is, stable operation region, unstable operation region and none operation region. As may be seen from Figure 5,‘small’ opening of gas injection choke or‘large’ opening of oil production choke results in unstable production flow, while‘large’ opening of gas injection choke or‘small’ opening of oil production choke brings about stable production flow. This gives an instruction for stabilizing artificial gas lift process by manual manipulation of both gas injection choke and oil production choke. This stability map inspires the idea of research on optimal trajectory control for artificial gas lift process. Figure 6 shows the relation between gas injection flow rate and different openings of both gas injection choke and oil production choke. This is a simulation result corresponding to Figure 5. As shown in Figure 6, there is a trend of decrease in gas injection rate with the decrease of both choke openings. In practical production, there should be certain constraints on gas supply equipment. For example, minimum gas injection flow rate and maximum gas injection flow rate.

Similar to Figure 6, Figure 7 shows the relation between oil production flow rate and different openings of both gas injection choke and oil production choke. An‘optimal point’ on the stability map is selected. For purpose of explanation the minimum and maximum gas injection flow rate in Figure 6, are moved to Figure 7 in dotted lines. This is one constraint for determination of optimal point. The optimal point in petroleum production is not the one with maximum production, but the one with maximum economic profit. Thus, many factors may affect the selection of optimal point, following which an arbitrary point is chosen as the optimal point. It is noted that the simplification with losing generality, is the main contribution of this invention which focuses on the development of trajectory control, rather than determination of optimal point.

Figure 8 shows the stability map together with the selected optimal point (0.9, 0.8), which means that the opening of gas injection choke is 90% and the opening of oil production choke is 80%. This optimal point is also the target operation point. The optimal trajectory generation algorithm is invented to drive any initial point in unstable region to this optimal point in shortest transition time.

Figure 9 shows an example of straight trajectory, where the trajectory is divided into 2 stages. The first stage obeys the principle of ‘escape the unstable region as soon as possible’, which means that the path of this stage is almost perpendicular to the boundary between unstable and stable regions. The second stage lays in the stable region, which is the one to be optimized.

Figure 10 shows set-point values and their ramp limits for controlled variables. The set-point values are equilibrium points of controlled variables, which are calculated based on third order mathematical model with respect to different inputs.

Figure 1 1 shows the comparison between straight trajectory and optimal trajectory plotted on stability map. The straight trajectory (solid curve) is the second stage in Figure 9, which is compared with the optimal trajectory (dotted curve) in stable operation region.

Figure 12 shows the result of transition time, which is corresponding to Figure 1 1 . The transition time of optimal trajectory is less than that of straight trajectory, which means that it is faster to reach the optimal point if we follow the optimal trajectory.