WO2008066414A2 - System for measuring the productivity of a group of wells with a single flow meter - Google Patents

Abstract

This invention relates to the field of measuring technique, more specifically, to the area of measuring the parameters of liquid and gaseous fluids, and can be used in information-measuring systems on objects of oil/water production, transportation, treatment. A system of monitoring comprising a flow meter, a flow multiswitch, and a regulating device, different in the following features. The first outlet of the multiswitch is connected to the main pipeline to the group oil gathering station. The second outlet is connected to the flow meter inlet. The flow meter's production outlet is linked to main pipeline. The regulating device is connected to the outlet by the flow meter signal and to the flow multiswitch.

Description

System for measuring the productivity of a group of wells with a single flow meter

This invention relates to the field of measuring technique, more specifically, to the area of measuring the parameters of liquid and gaseous fluids, and can be used in control of liquid flow parameters, in particularly, in monitoring of current state of oil and gas fields via control of well operation and metering the well production rate.

This invention can be applied in information-measuring systems on objects of oil/water production, transportation, treatment, as well as in metering of well flow rate in a well cluster, and pumping volumes through injection wells.

Now the production rate monitoring systems for well clusters (as a group of production objects) in Russian oil industry are based on separation-type systems equipped with a flow switch. These systems are different modifications of a flow meter setup "Sputnik". All of them are based on hydrocarbon stream switch for metering production rate from a single well in a cluster during a certain time period according to a fixed schedule. There are known methods for estimating the measuring time periods as a function of well production parameters (G. S. Abramov and A. V. Barychev, Practical Flow Metering in Oil Industry, OAO VNIIOENG, Moscow, 2002).

There exits a method based on commercial flow meters (Oil Production Equipment. Handbook, Nedra, Moscow, 1990, pp. 402-411) used for primary accounting of production from wells belonging to certain area of oil field; these wells are collected into groups by technical and other conditions within the intrafield system of oil collection, transportation and treatment. These systems comprise: a multiposition flow switch, separation-type measuring tank with control and automation devices, as well as a commercial microcontroller (or processor unit) linked to communication lines with control instrumentation, and also a system of pipelines, shutoff and safety valves (valves, gates).

These setups operate in a cyclic mode of loading-and-emptying the metering separation tank. The setup is driven by energy of the medium controlled (well fluid) and it sums up the production volume for a given time (or several cycles); this operation scans every well in the cluster (according to a program).

The general shortcoming of existing devices is a high cost and high working-hours consumption. Besides, there exists a set of strict requirements on assembly, commissioning, exploitation and maintenance of this setup, since it consists of many units and elements, both mechanical, hydraulic and electric. But the most essential drawback of this type of metering systems is a cyclic process of production metering and associated challenges and measurement errors: the cycle of loading and emptying of metering tank is handled by a set of mechanical levers and a float level indicator. Besides, the channels of hydraulic part should be cleaned periodically from deposits and impurities; so complete stop of measuring system is required.

The patent RU 2 247 239 teaches about the method of oilfield control through control of a group of wells and accounting the total production by measuring the daily output. According to this method, the output of oil well group is computed by measuring the average flow rate (averaged for a reasonable time) for every single well in the group: every well is connected sequentially (by a program) to the main metering device; this allows recalculation of production flow rate into daily output. Firstly, the most dynamic well of the group is chosen (taking one of most unstable flow parameters). Then the value of relative r.m.s. error of the average flow rate m(q) is calculated and stored in the memory of evaluator (a commercial controller). This gives us the scanning time of this well, and other wells are scanned for the same period. Then the scanning period of every z-th well is corrected: the current r.m.s. error for average flow rate q; is compared with a given value, so the scanning time for the z-th well is prolonged or shorten. If the moments for scanning of flow rate for two wells of a group coincide, the scanning priority is established by the production level of these wells.

The shortcoming of this method and corresponding equipment is technical complexity and impossibility of prompt response to a drastic change in flow rate from one of wells.

There is known a device (RU 2 265 122) for measuring the production rate from a group of oil wells. This metering device has a vertical tank with a side tangential connection pipe for feeding of production fluid, and having a top nipple for discharging the well head gas and a bottom nipple for liquid draining. The device is equipped with sensors for taking the fluid parameters and position in the tank, with a controller with multichannel inlet (according to number of sensors) for input of information signals from sensors and with control terminals. It is also equipped with a multiposition switch of the fluid (with inlets according to the number of connected wells) and two outlets. One of switcher outlets is connected to the side tangential connection pipe of the tank and another one is hydraulically connected to the top and bottom nipples of the tank and connected to the common oil field manifold. The system is equipped with a gas flow meter and liquid flow meters installed in corresponding pipelines, wherein the lower part of the tank is taped towards the liquid drainage outlet. The side tangential connection pipe for feeding of production fluid is installed at the place of junction of cylindrical and tapered part of the tank. The pipe between the tank and gas flow meter is equipped with a sensor of liquid droplets in the gas flow; this sensor controls the throttle valve through the controller. According to the described method, the territorial group of oil wells are connected to the multiposition flow switch. The controller has a program for switch drive; this program set consequently every well for metering the flow rate. The process of measuring the production rate is similar to existing technology: the fluid from one group of wells is fed through the side tangential connection pipe to the vertically installed tank. Since this pipe is tangential to the tank body, the input well fluid gets an intensive rotary motion along the wall. The motion intensity is proportional to the feed velocity. After a certain time, dynamic equilibrium of components at the tank's inlet and outlets establishes. Both gas and liquid flow meters give almost steady readings. The explanation is that the well fluid even under most unfavorable flow mode in the exhaust line (slug mode) and at the moment feeding to the production rate metering device (to the vertical tank) is mixed with rotating mass of liquid and being separated into gaseous and liquid phases due to a high difference in densities. The liquid droplet detector in the described device operates in routine mode: it sends a signal when the gas optical transparency becomes below a certain level. The signal informs the controller that low-dispersion droplets are entrained by gas flow. The controller uses this signal to vary the throttle valve cross- section.

The shortcoming of this device is the use of flow meters with a low accuracy resulting in irregular and low time resolution; this device fails to react promptly to a change in production rate for every well in a cluster. This also reduces the informative value and reliability of output data.

The technical task solved by disclosed invention is a new method for monitoring of production from wells allocated on a restricted area.

The technical result of this invention is a higher reactivity to variations in well operating parameters with simultaneous improvement in reliability and information value of measured data. This technical result is achieved by using a system of production monitoring for closely-spaced wells (well cluster) connected via a flow multiswitch to the main pipeline. The system comprises pipelines, from at least two wells, connected to inlets of the flow multiswitch and the first outlet of multiswitch is connected to the main pipeline delivering the production fluid from the well cluster to the group oil gathering station and the second outlet of multiswitch is connected to the input of flow meter on the path to the main pipeline. This system has an additional regulating device, reacting to the flow meter readings and controlling the flow multiswitch. The flow multiswitch has an opportunity to connect all inlets to the first type of outlet and an opportunity to connect one of input flows to the second outlet.

For the preferred embodiment of invention, the regulating device is designed on the basic of a personal computer having a software for data acquisition about flow meter readings, for synchronizing of time counting and global time, for measuring data storage, for flow directional control, for data communication via communication system, as well as for data storage in nonvolatile memory. Another embodiment of same system is possible; this is a control board with instrumentation for data registration (data about single well flow rate) and tools for mechanical control of flows. A flow multiswitch can be designed as a hydraulic rotary switch or as an assembly of double-trigger switches; every switch is installed on the exhaust line of every well with opportunity to forward the product flow either to the inlet of flow meter or to the main pipeline from the cluster. The design of the flow multiswitch depends on the operation conditions. To achieve the formulated technical objectives, the high-accuracy (few percent) flow meter should be used in all embodiments.

The disclosed invention relates to the tools and methods of metering the production rate of a group of oil or gas wells. This group comprises several wells, preferably with heads within a small distance and making up a so- called well cluster. Ultimately, the whole production from all wells is collected into single pipe, so all wells in a cluster are hydraulically connected. The well production is oil-water-gas fluid. Production rate can be metered using the approach of the total or single-component (oil, water, gas) flow. Usually the well production is measured with flow meters installed either to a single well or to the whole cluster. The disclosed invention can be applied for control of fluid flow rate in the water-injection wells being a member of the same cluster. For the last case the problem becomes simpler because the injection fluid is water with known density (or includes gas in known proportions). The fluid injection is a well known tool in oil industry for sustaining the reservoir pressure.

The disclosed system of monitoring (Fig. 1) comprises: wells 1 - 3, flow multiswitch 4, flow meters 5, measurement lines 6, bypass line 7, regulation tool 8, main pipeline 9, and data transfer line 10. Measurement lines are pipelines connecting wells to the multiswitch. After production fluid passes through the monitoring system, all production from all wells is collected into a single pipeline for transporting to downstream facilities.

The disclosed system of production rate monitoring guarantees a high quality of measuring the well productivity, water cut, and gas-oil ratio for every well in a cluster. This is achieved by the following procedure: the flow meter controls the total productivity of all production sites and does not distinguish information about productivity of a specific production site. When a significant variation in total productivity is recorded, the flow meter starts measuring the individual productivity of every well in a cluster according to assigned cyclic mode. This reduces the cost of a monitoring system, since the flow multiswitch will wear out slower than for a constant cyclic switching mode. The disclosed monitoring system ensures that all significant changes in well productivity would be recorded without loss in accuracy of productivity measurement for every well.

The disclosed monitoring system has the following embodiment (see Fig. 1). The well cluster is equipped with a flow meter and a flow multiswitch that is capable to forward the production flow from any well in the cluster to the inlet of the flow meter.

The flow meter usually includes a Venturi tube and a gamma densitometer (or it can be separation-type device). This type of flow meter can be used for simultaneous measurement of flow rates for oil, water and gas.

A regulating device is required for synchronizing the time counts and global time, for storing measurement data, acquisition and storing of flow meter data, for control of multiswitch according the chosen procedure, data transfer in communication system, data storage in nonvolatile memory.

All measuring information is forwarded through available channels to the centralized storage (the data base), so it can be retrieved any time for visualization, control of measurement quality and optimization of the monitoring system. Thus, the system in the preferred variant of embodiment includes: executable modules, which, in particularly, can operate on a platform of control device and the data base for centralized storage of measuring data, place, and time of measurements. The said executable modules are required for several purposes: data acquisition through available channels and storing of this information in the centralized data base, processing of measuring results. The later results are by- component composition of well production, interpolation of this composition. The said modules are required for storing of measurement results in the centralized data base or transfer of information via available communication lines, for statistic processing of measurement results aimed to adaptation of the monitoring system to current situation, in particularly, for estimating the said parameter "significance of variation in total productivity", for modeling of monitoring system in time, and for results visualization.

In the preferred embodiment, the monitoring system operates in the following way.

A flow meter runs steadily and measures the total productivity of all wells. When an event of significant variation in total productivity is registered (this may be a change in water cut, total flow rate, or gas-to-oil ratio), the flow meter is rendered to cyclic measurement (scanning) of productivity for every well in a cluster.

Below we present the essence of this invention using graphic material.

The productivity control system can be configured, in particularly, as shown in Fig. 1, at the example of a cluster with N oil wells; the system comprises: pipelines for transportation of well fluid (water-oil-gas mixture), measurement lines for transportation of well fluid to the flow multiswitch, a multiswitch, a bypass (for transportation of production from wells beyond the measuring scope at the current moment), a regulating device, and also the lines for communication and power between the regulation device and the flow meter.

The regulating device is designed on the basic of a personal computer having software for data acquisition about flow meter readings, for synchronizing of time counting and global time, for storing measurement data, for flow directional control according to the developed technique, power feeding of the flow meter, for data communication via communication system, as well as for data storage in nonvolatile memory. The options of regulating device are installed by the software or possible additional executable modules. The switching thresholds of the flow multiswitch are assigned from current parameters of operating wells or/and from their productivity. These parameters can be calculated from the system models or assigned directly (e.g., by results of direct measuring of flow rate for every well). Events of exceeding of controllable parameter above the switching thresholds are controlled by the regulating device.

The flow meter carries out continuous measuring of productivity for one well either for all wells. (This means that measurements are carried out always or almost always with a certain time step.)

Additional gas flow meters can be installed on the gas line of well heads. Partially separated gas is exhausted through this line to the day surface. (Separation is partial only because it occurs in the separator attached to the electric centrifugal pump at a high depth, which means a high pressure and temperature). Since the gas from the gas line is mixed with the rest production of the well downstream to well head, the readings of the gas flow meter can be considered as a low limit for the gas content in the production flow. Then the flow meter is connected for measuring the productivity of a single well, this means a more accurate measurement of fluid composition from this well. All sensors of gas flow meters are equipped with tools for data transfer to the regulating device via the communication line. Since the well-head gas is dissolved in oil, any change in the gas flow rate in the gas line evidences about changes in oil production from this well. This fact can be used for arranging of priorities in cyclic scanning of well with a single flow meter.

The regulating device acquires information about production sites and governs the flow multiswitch according to the following algorithm: if there is no event of exceeding the threshold for controllable parameters of the total productivity, the flow meter keeps operating in the same mode and meters productivity of the entire cluster. If controllable parameters of total productivity exceed a switch threshold, the flow meter is switched into cyclic scanning mode for metering of individual well of the cluster. The data about more accurate component composition of product fluid of all wells is recognized valid since the moment of registration of the last considerable change in the parameters of the whole production of wells.

The disclosed invention about monitoring of well cluster productivity (production sites) is based on Shannon's theorem. This theorem states that if values of a function are unknown everywhere, except of an ensemble of random points, the growth of this ensemble can restore the function for the whole domain of function, at least, enough for convergence of partial integral sums. In general, the said theorem is quite obvious. The implementation of the disclosed monitoring system is, actually, a method for choice (time moment) and obtaining the points where the values of unknown function are measured directly with a high accuracy and/or high resolution. Besides, the parameters of total productivity are measured regularly. (Here we mean "always or almost always ".)

To illustrate the efficiency of disclosed system, simulation of operation for two different monitoring systems was carried out. The first system is those depicted in Fig. 1. The second system includes a set of flow meters (one flow meter per every well in a cluster). The system was modeled using the same set of data from real wells (the oil wells from Western Siberian oilfields). During our numerical experiments, the resolution of flow meters in the second type of system was varied (symbol δ_m on charts). For the first system, a threshold of "substantial change" for parameters of total productivity was varied (the same symbol δ_FM in charts since this is essentially the same); the flow meter was assumed to be absolutely accurate. The result of operation of the first system is presented in Fig. 2. The result of operation of the second system is presented in Fig. 3. The vertical axis in charts is the relative error of productivity measurement (relative the initial data about productivity). Obviously, a single flow meter in the first system provides quite an accurate data about productivity of the entire well cluster.

The horizontal axis in charts shown in Fig. 2 and Fig. 3 is the plotting for amount of information collected in the monitoring system during operation. Obviously, the more information, the lower the measurement error. Additional studies demonstrated that the reason why information was lost is insignificant. For example, information can be lost by reason of low accuracy of monitoring system or it was lost because the total productivity has been measured instead of measuring productivity for individual well. Here it is important that we have a relation between the accuracy of monitoring and the information loss.

Thus, it is possible to state the problem of monitoring for a group of production sites. The problem is wider that just a better accuracy of measuring: this is a task of higher information value of measurements. Indeed, objects with almost steady productivity can be met among the controlled sites. Therefore, frequent repeating measurements of these objects would not increase the informative value. Therefore one can use one flow meter for measuring the entire productivity of the well cluster.

A more accurate problem statement for informative measurements is the following. Let us consider an object with varying productivity; this object is a source of information. (Hereafter the proper definition of information is so called probability approach by Kolmogorov [A. N. Kolmogorov, Three approaches to definition of concept "amount of information". New in Life, Science, and Technology, Series "Mathematics and Cybernetics", January 1991, pp. 24-29].) In this case the monitoring system is the information "receiver". The situation is described by formula: dl = ∑I_r(t, δ,AO -At- ∑η(t, δ_FM) -I™(t, δ_m,At_m) -At_m ≥ O (1) all time steps all time steps where At_m is the time interval between measurements, δ_FM is the resolution of measuring devices at the given time t for the given production object, I™ (t, δ_FM ,At_m )is the information flow accepted by the monitoring system, I_r (t, δ ,M) is the information flow from the production objects. Information flows have three important properties. They are: always can be measured, tend to zero for production objects if the productivity tends to a constant; the higher the resolution and/or accuracy of measurement, the higher the information flow.

Here η(t,δ_FM)\s the responsive function for a monitoring system. O ≤η(t, δ_FM) ≤l , because (due to obvious technical limitations) the flow meter during cyclic scanning cannot be switched immediately to the well after an event (drastic change in productivity), so extra changes in productivity can occur since the event registration (drastic change in entire productivity) till the measuring of productivity for the specific well. Thus, some information after the event of drastic change in entire productivity can be lost irreversibility. The value of this function can be estimated by Erlang function from the mass service theory [A. Y. Khinchin, ed. By B.V. Gnedenko, Collection of papers on mathematical theory on mass service. State Publishing House of Physical and Mathematical literature, Moscow, 1963, pp. 199-208]. In our modeling of developed approach the values of η(t,δ_m) varied in the range 0.88...1, where δ is the prescribed resolution required for monitoring of a production object, dl is the inevitable loss of information.

As modeling for the system developed and for the existing monitoring system "Sputnik" has revealed, the value of dl is the key feature for monitoring systems, since it is related directly to the accuracy measurement for every individual well. (The value of δ_m has only secondary significance - see Figs. 2 and 3.) Our modeling demonstrated that the correlation coefficient between dl and the measuring error is about 80% - e.g., see Fig. 4.

The claimed approach to the problem formulation on control of productivity for a group of production objects (as a problem of evaluating the informative value) explains the obtained result. Besides, the problem statement and method of calculation (explained by relationship 1) can be applied for estimating the efficiency of any systems of group monitoring regardless the meter design and arrangement of the system. Usually this is enough (e.g., through simulation experiment) to determine the type of relation between the information loss and errors in measuring the productivity parameters.

The disclosed system is aimed to maximizing δ_FM at every time moment and for every well: this pursues optimization of usage of the flow meter and minimization of dl, i.e., the lowest measurement error for well productivity.

So, the disclosed monitoring system allows optimizing of operation with a single flow meter for control of productivity of a well cluster and individual wells in this cluster. This makes the productivity monitoring system less expensive without compromising on accuracy of parameters measurement.

Herein we give an example of system embodiment:

We take a monitoring system configured as in Fig. 1 with N = 3. The initial productivity of well is (watercut = water flow rate / (water flow rate + oil flow rate) * 100%) the following:

The switching threshold is established as 5% for every parameter of initial productivity. The flow meter initially measured the total productivity for entire cluster.

At a certain time the watercut for the well number 1 increases to 21%. This means that the watercut for entire product increases by 0.14%. Since this is below the switch threshold (+5%), the flow meter keep measuring the total productivity of all wells (cluster).

At a certain time the oil flow rate from well number 2 becomes equal 70 m³/day. The total productivity by liquid measured by the flow meter drops by 8.6%. This exceeds the threshold, so the flow meter operates in the mode of scanning of all wells in the cluster.

If there is a discrepancy between the actual total parameter (e.g., oil productivity) and the sum of productivities obtained during the scanning regime for every well, and this discrepancy is higher than 5%, the cyclic scanning of well productivities is repeated.

After more accurate data on well productivities have been obtained, the new switching thresholds are calculated relative to the new data on productivities.

Claims

What is claimed is:

1. A system of monitoring for a group of wells comprising a flow meter, a flow multiswitch, and a regulating device, different in the following features: the first outlet of the multiswitch is connected to the main pipeline to the group oil gathering station, the second outlet is connected to the flow meter inlet, and the flow meter's production outlet is linked to main pipeline; the regulating device is connected to the outlet by the flow meter signal and to the flow multiswitch, wherein the said multiswitch has the opportunity of connecting all inlets to its own first outlet, and the opportunity of connecting of all flows to the second outlet.

2. The system as in claim 1, wherein the regulating device has the facility for data acquisition from the flow meter, synchronization of time counting and global time, measurement data storage, control of flow directions, power feeding for the flow meter, performing communication and data transfer via communication lines, and storing the data in nonvolatile memory.

3. The system as in claim 1, wherein the flow multiswitch is a rotary hydraulic switch.

4. The system as in claim 1, wherein the flow multiswitch is a set of two- position flow switch, and every said switch at the outlet pipeline of every well with facility to direct the flow to the inlet of a high-accuracy flow meter or to the main pipeline.

5. The system as in claim 1, wherein the system includes a set of executable modules operating within the regulating devices and providing access to all measurement results suitable for processing and visualization, as well as for data treatment in the form of graphs and tables.

6. The system as in claim 5, wherein the executable modules can calculate statistic parameters of the group well productivity.

7. The system as in claim 5, wherein the executable modules can model the system operation.

8. The system as in claim 5, wherein the executable modules can modify the operation algorithm for the regulating device.

9. The system as in claim 1, wherein it includes the centralized data base about measurement results.