WO2009154499A1

WO2009154499A1 - A system for testing well cluster productivity

Info

Publication number: WO2009154499A1
Application number: PCT/RU2008/000386
Authority: WO
Inventors: Aleksander Yurevich Lomukhin; Vladimir Nikolaevich Ulyanov
Original assignee: Schlumberger Canada Ltd; Services Petroliers Schlumberger SA; Prad Research and Development NV; Schlumberger Technology BV; Schlumberger Holdings Ltd
Current assignee: Schlumberger Canada Ltd; Services Petroliers Schlumberger SA; Prad Research and Development NV; Schlumberger Technology BV; Schlumberger Holdings Ltd
Priority date: 2008-06-19
Filing date: 2008-06-19
Publication date: 2009-12-23
Anticipated expiration: 2010-12-19

Abstract

The invention deals with a scope of the geophysical testing methods, namely, the well productivity testing, and can be applied when testing potential flow rate in a well cluster. The system of well cluster productivity monitoring contains the output pipe-mounted low-accuracy flow meters corresponding in number to wells monitored, the flow direction selection unit with inputs connected with wellstream pipe, the first selection unit output is connected with the main pipeline, the second output is connected with a high-accuracy flow meter input, and its output connected with the main pipeline, with this, the system also contains a controlling device connected with the flow meter outputs and the direction selection unit.

Description

A system for testing well cluster productivity

The invention deals with a scope of the geophysical testing methods, namely, the well productivity testing and can be applied when testing potential flow rate in a well cluster.

In Russia, systems for monitoring wells combined in a cluster (as production complexes) are separation-type flow switch-equipped systems. The systems above are various versions of the Sputnik flow meters. They use flow switches for hydrocarbon flows in various wells to measure the productivity of each of them in a cluster for a certain period of time, namely, in cycle (a so called "inelastic" time-table). Various methods for calculating measurement periods in the function of well productivity parameters have earlier been suggested. (F. C. AβpaMOB, A. B. BaptraeB. lTpaκτHHecκaa pacxoβOMeTpirø B HeφτflHθii πpoMtiiπjieHHOcra. OAO "BHHHO3HF", MocKBa, 2002.) (G. S. Abramov, A. V. Barychev. Practical Flowmeter Survey in Petroleum Industry. VNIIOENG OJSC, Moscow, 2002.).

Meter stations are known (HeφτeπpoMticjiθBoe o6opya,OBaHHe. CπpaBOHHHK, M.,«He,zjpa», 1990, CTp. 402-411) (Oil-field Equipment. Reference book, M.,"Nedra", 1990, pp. 402-411) for primary wellstream account that cover some oil field area and for a number of technological and other conditions are grouped in oil gathering, transport, and treatment. In design, they consist of a multi-position liquid switch, a separation displacement tank equipped with control and measuring instruments, automation and controls; they contain an industrial microcontroller (or a computing unit) linked to control and measuring instruments, automation and controls, and a piping system, locking and safety devices (knobs, valves, latches and so on). These stations operate under the cyclic conditions to fill and empty the separation displacement tank using energy of the medium monitored (wellstream), summing production rate of all the wells in the group ( one at a time, as to a program) for some preset measuring time (or a number of cycles).

Common disadvantages of the existing devices of similar purpose are labor-, materials-, metal-consumption and quite a wide range of requirements to mounting, adjustment, operation and repair, as well as the availability of many mechanical, hydraulic and electric units and components. However, this is the cyclic operation of well production meters that is the most material disadvantage, which causes measurement difficulties and errors due to a mechanical lever system to control cycles of filling and emptying a displacement tank by means of a float-level indicator, and the need for regular cleaning fluid-end chambers from all kinds of deposits (contamination), in doing so, the station should be completely shut off.

A close analog to the claimed system can be (RU, patent 2265122) a production rate measuring unit, containing a vertical tank having a lateral tube tangential to the tank shell for wellstream feeding, an upper tube for associated gas draining, a lower tube for liquid draining, and condition- and-position detectors for monitoring a product in the tank chamber; a controller having multiple input (as to a number of detectors) to input electric data signals from these detectors and controlling outputs; and a multi-position liquid (wellstream) switch having inputs for applied wells and two outputs: one of them is hydraulic-piped to the tank via the lateral tube of the latter, and another output of the liquid switch is hydraulic-piped, respectively, to an upper and a lower tank tubes and to a gathering oil-field main, a flow meter-gas counter and a flow meter-liquid counter: each of them being mounted on corresponding pipe; a lower part of the tank being tapered to the liquid draining tube. The lateral tube for wellstream feeding tangential to the tank is mounted on the tank shell at the exit point to a lower cone-shaped part of the tank, a liquid-in-gas detector and a speed control valve driven by this detector via a controller are mounted on the piping between the tank and a flow meter-gas counter.

The disadvantage of the certain engineering solution should be low- accuracy flow meters, which makes difficult to monitor the variation in each cluster well yield immediately, and low measurement reliability and information content.

An engineering problem solved by the invention proposed consists in developing a new cluster well monitoring system.

The technical effect of the system applied consists in increasing the response speed when well working conditions vary and in simultaneous enhancing the measurement reliability and information content.

To achieve the technical effect a cluster well monitoring system is proposed containing output pipe-mounted low-accuracy flow meters corresponding in number to wells monitored, a flow direction selection unit with inputs connected to wellstream pipe, the first selection unit output is connected with the main pipeline, the second output is connected with a high-accuracy flow meter input, and its output connected with the main pipeline; with this, the system also contains a controlling device connected with the flow meter outputs and the direction selection unit. The option of choice for implementation is a personal computer-base controlling device, its software allows for logging data measured with low-accuracy flow meters, synchronizing tic mark and global time, storing measurement data, logging and storing high-accuracy flow meter-measured data, controlling flow direction selection, powering low-accuracy flow meters, interfacing and delivering information to a intercommunicating systems available, storing data in permanent memory. However, an optional version can be a controlling device as a control console containing means of logging wellstream conditions, and means of mechanical direction control. The flow direction selection unit can, particularly, be as a rotary hydraulic switch or as a set of the two-position flow switches, each of them being mounted on the output manifold of one of the wells directing wellstream to the high-accuracy flow meter input or the main pipeline. Design of the direction selection unit depends on working conditions. To apply the system and achieve the engineering effect, flow meters of primary array of up to 18%-accuracy and flow meters of secondary array of up to 3%- accuracy should be proposed.

The engineering solution proposed is related to means and methods of measuring well group productivity. Such a group consists of several surface wells spaced at close range, namely, about one meter forming a so- called well cluster. Finally, all the wellstream arrives into one pipe, therefore, all the cluster wells are hydrodunamically coupled. Wellstream is an oil-, water-, and gas flowing mixture. Productivity should be measured with flow meters mounted both for an individual well and for a well cluster.

The monitoring system developed contains first-array flow meters corresponding in number to wells in a cluster, second- array flow meters (second-array flow meters are of higher accuracy as mentioned above and/or of higher resolution compared to first-array ones), the second- array containing, at least, one similar flow meter, a controlling device, and a direction selection unit. Measuring lines are pipelines connecting wells and flow meters. On passing through the monitoring system, all the wellstream arrives in one pipeline transporting to a processing train. The productivity monitoring system proposed provides the high- quality measurement in each several well. This achieves for the reason that the first-array flow meters manage to monitor wellsteam, though the data are rather coarse (coarse resolution and/or low accuracy). When any first- array flow meter logs significant variation in wellstream, the second-array flow meter (characterized by higher resolution and/or higher accuracy) should be switched to measure wellstream of this well. The well group monitoring system becomes markedly cheaper, since the first-array flow meters are, as a rule, much cheaper than those of second array are. Moreover, the second array can contains one flow meter only. The monitoring system proposed logs significant variation in the productivity of each cluster well, which keeps measurement accuracy from deterioration.

The monitoring system proposed can be implemented as follows (Fig. 1). A well cluster will be equipped with low-accuracy and/or coarse- resolution flow meters, namely, one per each well (first-array flow meters), one high-resolution and/or high-accuracy flow meter (second-array flow meter), direction selection switches who allow for transporting wellstream of any cluster well to a second-array flow meter input.

The first-array flow meters are flow meters who allows for logging total flow (oil+water+gas), determining water content of the flow and measuring gas flow. By obtained data, oil content of the flow can be calculated.

The second-array flow meters are flow meters, preferentially, consisting of a Venturi tube and a gamma densitometer. The second-array flow meters can be of separation type. These flow meters allow for measuring passed oil-, water-, and gas flows individually. The controlling device is a device designed to:

• log first-array flow meter-measured data,

• synchronize tic mark and global time, • store measurement data,

• log and store data measured with the second-array flow meters,

• control flow direction selection system according to a procedure developed,

• power first-array flow meters,

• interface and deliver information to an intercommunicating systems available,

• store data in permanent memory.

The base-case monitoring system proposed will operate as follows. The first-array flow meters keep wellstream measurements. When any first- array flow meter logs significant variation in wellstream (variation in water content or total flow, and in gas content being more than variation limit), wellstream of this well will be directed to a second-array flow meter for high-accuracy measurements. With no significant variation in cluster well productivity for long enough, a second-array flow meter operates under "inelastic time-table" conditions. In this case, it measures the productivity of each cluster well for the prescribed period in cycle, until the first-array flow meters log essential variation in cluster well productivity (more than preset limit).

Further, nature and scope of the invention will be detailed using drawings.

The productivity monitoring system under consideration can be implemented, particularly, as shown in Fig. 1 , for example, using a cluster with N oil-and-gas wells and first-array flow meters, one per each well, pipelines to transport wellstream (water-, oil-, gas mixture), measuring lines to transport wellstream from the first-array flow meters to a direction selection unit, a direction selection unit, bypass pipe designed to transport flow the wells whose productivity a second-array flow meter does not measure at the present moment, a controlling device, and communication and power line between the controlling device and the first-array flow meters.

A personal computer-base controlling unit with controlling outputs coupled to inputs of the direction selection unit, is designed to log first- array flow meter-measured data, synchronize tic mark and global time, store measurement data, log and store second-array flow meter-measured data, control the direction selection system according to a procedure developed, power first-array flow meters, interface and deliver information to an intercommunicating systems available, store data in permanent memory. The controlling unit capabilities are specified by installed software.

Limits for a direction selection unit are specified with regard to well working conditions and /or productivity. These data can be calculated when simulating the system operation or a user can set them by him. The controlling device monitors moments and events when productivity data are above limits.

The first-array flow meters keep measuring wells wherein being mounted (therefore, always or almost always there is a certain time step).

A first-array flow meter, particularly, can contain a flow limiter used to help determine a functional connection between the wellstream and density. For this purpose, the flow meter is equipped with a wellstream- pressure indicator upstream the flow limiter and a pressure drop gage while wellstream passing through the flow limiter. To determine the wellstream density, it is necessary to obtain a wellstream oil-, water-, and gas ratio (since laboratory tests are certain to determine pure oil-, water-, and gas flows individually and are known). To do this, a device should be equipped with a producing watercut detector. A gas detector should be mounted on a well gas-discharge line (but may not be). Partially separated gas (since this is in-deep separation and it conducted in a separator of the electric centrifugal pump) arrives to the surface using the gas-discharge line. Since gas from the gas-discharge line is mix with the rest of the wellstream, the gas detector measurements are to be a low gas content limit. By using a second-array flow meter in wellstream measuring, the wellstream component ratio will be specified. All the detectors in the used unit are equipped with means to transmit data to a controlling device connecting link.

While a second-array flow meter measures any well productivity, a corresponding first-array flow meter is under calibration, i.e. a watering detector and a gas rate detector will be adjusted to zero.

A controlling device logs data on the production complex condition and controls a direction selection unit according to the following algorithm: Should neither of the first-array flow meters logs the situation when the parameters monitored are above the limits, a second-array flow meter controls the cluster wellstream in cycle prescribed. Should either of the first-array flow meters logs the parameters monitored are above the limits, a second-array flow meter will be switched to measure the wellstream conditions wherein the limits logged.

The solution of the problem of monitoring cluster well productivity (production complexes) is based on the fundamental Shannon's theorem. The substance of the theorem is as follows: should any function values are everywhere unknown except for a set of the random points, by enhancing this set the function can be recovered over the whole measurement area, at least, in terms of the convergence of partial integral sums. Generally, the theorem mentioned is rather evident. A method of implementing the monitoring system proposed represents, essentially, a procedure of point (instant of time) selection and obtaining, wherein the unknown function values directly vary to a high accuracy and/or high resolution. Moreover, measurements are continuously kept of low or equal resolution or accuracy, (here, "continuously" is meant "almost always or always ").

To illustrate the proposed system efficiency, performances of the two systems for monitoring cluster well productivity were numerically simulated. Fig. l gives the first system that operated according to the procedure proposed. The second system consisted only of the first-array flow meters that kept measuring continuously. The systems were simulated for identical data sets obtained when monitoring performing well productivity. These are petroleum wells in West Siberian fields. In experiments, resolutions of the first-array flow meters were varied (symbol δ_FM in diagram). Fig.2 gives the results for the first system. A diagram of Fig.3 shows the results for the second system. A vertical axis in the diagrams is the ratio measurement error relative to initial productivity data. Obviously, to obtain such a low error in the first system the first-array flow meters can be used of resolution in order of magnitude lower than that of the second system. With regard to the peculiar features and the cost of the flow meters monitoring well productivity, price difference of the flow meters can be thousand percents.

Horizontal axes in the diagrams of Fig.2 and Fig.3 are data amount logged by the monitoring systems when operating. Obviously, the more data is obtained; the minor is the measurement error.

Thus, a problem of monitoring production complexes can be broader than simply a problem of increasing measurement accuracy. This can represent as a problem of increasing measurement information content. In fact, should there are wells among many production complexes monitored whose yield almost unchanged with time, it is little sense in often high- accuracy measurements or, in other words, there is no information content. Therefore, the second-array flow meters should be used for wells of more unstable yield. A problem of measurement information content can be defined more strictly. Let a production complex whose yield varies be a data source (hereinafter, a most adequate approach to determine data amount being a so-called probabilistic approach according to Kolmogorov- [A. H. KojiMoropOB. TpH πoflxo^a _κ oπpe^eneHHio ΠOHΛTH» "κojiHHecτBθ HHφopMaijHH". HoBoe B )KH3HH, Hayice H TexHHKe. CepHH "MaTeMaTHKa H κκ6epHeτHκa", >IHB. 1991, σrp. 24-29].) [A. N. Kolmogorov. Three Approaches to Definition of Data Amount Concept. New Developments in Life, Science, and Technology. "Mathematics and Cybernetics" Series, January. 1991, pp.24-29].). In this case the well yield monitoring system is a "data receiver". Let us consider a formula: dl = ∑I,(t, δ,Δt) -Δt - ∑η(t, δ_FM)- I™(t, δ_FM,At_FM ) - At_FM ≥ O (l) all time steps all time steps where

At_FM is the measurement intervals, δ_FM is the resolution of the flow meters of first or second groups depending on which of the high-resolution device logs in this instant of time t at the given production complex,

I™ (t, δ_FM ,Δt_FM ) is the data flow the monitoring system logs, I_r (t, δ ,Δt) is the information content transmitted from a production complex. Data flows possess three very important properties. These are always measurable; weakly tend to zero for production complexes whose yield tends to constant; the higher resolution and/or accuracy, the more data flow. η(t,δ_FM) is the function of the monitoring system sensitivity to data. 0 < η(t, δ_FM ) < 1 , due to evident engineering limits, the second-array device should not be always switched to log the yield of the certain production complex being busy, and the switching should not be done instantly. The value of this function should be appreciated using Erlangian formulas from the waiting theory - [A. Α. XHHHHH, ΠOΛ pe^. B. B. FHe^eHKO. Pa6oτti no MaτeMaτHHecκθH τeopHH MaccoBoro o6cjry5KHBamui. Focyzi,apcτBeHHoe H3ΛaτejitcτBθ φH3Hκo-MaτeMaτHHecκoH JiHτepaτypbi, MocKβa, 1963, CTp. 199-208]. [A. Yu. Khinchin, Edited by B. V. Gnedenko. Works on Mathematical Waiting Theory. State publishing House of physical-and- mathematical literature, Moscow, 1963, pp. 199-208]. When simulating the procedure patented, the values of η(t,δ_FM) were within the range of 0.98...1, δ is the preset resolution for production complex monitoring. dl is data losses to be foregone.

As the results have shown of simulating the monitoring system proposed and the existing Sputnik-type monitoring systems, this is dl that is the main characteristic of the monitoring system operation for the reason that it is directly related to measurement accuracy of each individual well yield. (With this, values such as δ_m are of marginal importance - Fig.2, 3.). By the simulation results the correlation coefficient between dl and the measurement error is about 80% - refer to, e.g. Fig. 4.

The approach to define a problem of monitoring complex productivity (as a problem of information content calculating) allows for explaining the result obtained. Besides, the problem definition proposed and the calculation method (represented by formula 1) can be applied to estimate the efficiency of any system of the group monitoring despite of the measuring unit design and a structure of the system itself. To do this, it suffices (e.g., in the course of a numerical experiment) to determine a type of relation between data losses and errors of the productivity characterization.

The system proposed is intended to maximizing, if possible, δ_FM at each instant of time and for each well, enhancing the use of the second- array flow meters and the least value dl , i.e. the minimum error in well production measuring.

Therefore, the monitoring system proposed provides the optimum use of the flow meters of various resolution and/or accuracy for monitoring the well cluster productivity and each individual well in this cluster. With this, the whole productivity monitoring system becomes much cheaper without noticeable deterioration of the productivity measurement accuracy. Examples of the system implementation:

Let the monitoring system is mounted as shown in Fig.1. A number of well is (1-3) N = 3. The initial well productivity (1 - 3) measured using first-array flow meters 4 - 6 on pipelines 8 (water content = water flow / (water flow + oil flow) * 100%):

Table 1

The switching limits for all the wells are 10% of the initial productivity. The second-array flow meter 9 measures the productivity of all the cluster wells in cyclic, i.e. 2 days per each well.

1. At a certain instant of time the well 1 watering will become equal to 21%. Since the data are within the switching limits (+5%) the second-array flow meter 9 keeps cyclic measurements.

2. At a certain instant of time the oil flow from well 2 will become equal to 80 mVday (-20%). Since the switching limit is not achieve (if not in this instant of time), the direction selection unit 7 switches the measuring line 5 of well 2 to the second-array flow meter input 9. Using bypass 11 oil from wells 1 and 3 arrives to the main pipeline 13. All the well 2 productivity data should be corrected for time required for measurements.

3. Should at the time required to correct the productivity data of the well 2 of example 2 switching limit higher is achieved in wells 1 and 3, the second-array flow meter 9 is switched to the well 3 for correction and then to the well 1 for correction as the well 3 productivity (as to oil) is higher (extra operating term of the switching unit 7).

4. At a certain instant of time the gas flow from well 3 will become equal to 1200 mVdays (+20%). Since the switching limit is achieved, (if not in this instant of time) the direction selection unit 7 switches the measuring line 10 of the well 3 to the second-array flow meter input 8. All the well 2 productivity data should be corrected for time required for measurements.

5. In measurements, all the data from the first-array flow meters 4 - 6, the unit 7 and the second-array flow meters 9 transmit to the controlling device 12, and from there via the line 14 to the intercommunicating systems.

On correcting the productivity of any well the switching limits are calculated in percent relative to corrected productivity data.

Claims

1. A system of well cluster productivity monitoring containing the pipe- mounted flow meters and the flow direction selection unit, differing in the fact that it contains the output pipe-mounted low-accuracy flow meters corresponding in number to wells monitored, the flow direction selection unit with inputs connected to wellstream pipe, the first selection unit output is connected with the main pipeline, the second output is connected with a high-accuracy flow meter input and its output connected with the main pipeline, with this, the system also contains a controlling device connected with the flow meter outputs and the direction selection unit.

2. The system according to item 1 differing in that the controlling device is designed for logging low-accuracy flow meter-measured data, synchronizing tic mark and global time, storing measurement data, logging and storing high-accuracy flow meter-measured data, controlling direction selection, powering low-accuracy flow meters, interfacing and delivering information to a intercommunicating systems available, storing data in permanent memory.

3. The system according to item 1 differing in that the direction selection unit is designed as a rotary hydraulic switch.

4. The system according to item 1 differing in the direction selection unit is designed as a set of the two-position flow switch, each of them being mounted on the output well manifold directing wellstream to the high- accuracy flow meter input or the main pipeline.