WO2008093049A2 - Systèmes et procédés de gestion de soupapes de régulation de débit dans des systèmes de traitement - Google Patents

Systèmes et procédés de gestion de soupapes de régulation de débit dans des systèmes de traitement Download PDF

Info

Publication number
WO2008093049A2
WO2008093049A2 PCT/GB2008/000229 GB2008000229W WO2008093049A2 WO 2008093049 A2 WO2008093049 A2 WO 2008093049A2 GB 2008000229 W GB2008000229 W GB 2008000229W WO 2008093049 A2 WO2008093049 A2 WO 2008093049A2
Authority
WO
WIPO (PCT)
Prior art keywords
valve
flow
flow rate
control valve
dead band
Prior art date
Application number
PCT/GB2008/000229
Other languages
English (en)
Other versions
WO2008093049A3 (fr
Inventor
Jason D. Dykstra
Original Assignee
Halliburton Energy Services, Inc
Curtis, Philip, Anthony
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from US11/700,397 external-priority patent/US7636614B2/en
Priority claimed from US11/700,533 external-priority patent/US7606636B2/en
Application filed by Halliburton Energy Services, Inc, Curtis, Philip, Anthony filed Critical Halliburton Energy Services, Inc
Publication of WO2008093049A2 publication Critical patent/WO2008093049A2/fr
Publication of WO2008093049A3 publication Critical patent/WO2008093049A3/fr

Links

Classifications

    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
    • G05D7/00—Control of flow
    • G05D7/06—Control of flow characterised by the use of electric means
    • G05D7/0617—Control of flow characterised by the use of electric means specially adapted for fluid materials
    • G05D7/0623—Control of flow characterised by the use of electric means specially adapted for fluid materials characterised by the set value given to the control element

Definitions

  • Process plants such as those located at a modern hydrocarbon production site, consist of many different pieces of equipment, each of which can be linked together to form a total process, or sub-units, such as various unit operations and sub-processes. These can be controlled using a network of control loops to achieve a particular end result for the process or a particular unit operation.
  • Each control loop usually controls a particular process variable such as the flow rate of a material.
  • the objective of the loop is to keep the process variable within a required operating range, usually at a desired set-point, to ensure the targeted process result.
  • Each loop experiences, via external sources, or sometimes from internal creation by the loop itself, disturbances that cause the process variable to move away from the set-point. Additionally, interaction with other loops in the process network can cause disturbances that can affect the process variable.
  • sensors gather process variable data.
  • the drilling mud fluid must be provided within specific compositional and flow stream parameters. Sensors can monitor the flow rate of the mud, pressure, density, and other process variables, and this information is typically fed back to a control unit which operates individual control loops for each of the controlled variables.
  • a particular process variable is not easily measured.
  • process variable "observers" are used to provide the indirect measurement, or inference, of a particular process variable. See U.S. Patent Application 11/121,144 as Publication Number US 2006/0235627, filed May 3, 2005, entitled “Methods and Systems for Estimating Density of a Material in a Mixing Process", to Jason D. Dykstra and Justin A.
  • each control loop normally has its own controller which has a specifically-tailored algorithm to achieve a particular process variable control function.
  • the controller usually processes the sensor and set-point information and then decides what must be done to get the process variable back to set-point when a disturbance occurs. This decision takes the form of a control signal.
  • a "final control element" must implement the signal from the controller and apply it to the physical process. Examples of common final control elements for fluid applications are control valves, speed controlled-pumps, and speed- controlled compressors.
  • Common final control elements for solids-handling applications are variable-speed -drives for material flow control devices such as screw conveyors, belt feeders, slide gate valves, and rotary feeder valves.
  • flow control valves are very common in process control applications. They act to increase or decrease the flow of a fluid or solid, in response to a control signal, where a fluid can be a liquid, gas, or vapor, or various combinations thereof.
  • a flow control valve generally consists of at least a valve body, a moveable member within the body for adjusting the open area for flow, a valve-to- actuator linkage, and an actuator.
  • Actuators are powered devices used by an automatic controller to convert a control signal into movement. Actuators are typically powered pneumatically, hydraulically, electrically, or mechanically. For flow control valves, actuators convert the control signal to a physical action via the control valve linkage. Such linkages, being mechanical devices, are subject to wear over time, resulting in many cases, as looseness, or mechanical slack. Such slack in actuator-to-valve connections creates a phenomena called "valve slop". Thus, when the actuator moves the linkage in one direction based on a control signal, there is no opening or closing of the valve until the linkage moves far enough to "take-up" the slack caused by wear.
  • Figure 3 shows an end view of an example rotary-actuated valve shaft 350 as a component of a rotary control valve.
  • the valve was initially in fully closed 351 position 1.
  • the valve actuator was then sent a signal to make a 45 degree rotation 354 to achieve desired position 2 shown as 353.
  • desired position 2 shown as 353.
  • the actuator-valve assembly is worn and has mechanical linkage slack, the actuator will have to first move to take up the "slack" before the shaft actually begins to turn. If the slack in the actuator linkage is equivalent to 45 degrees of rotation, shown as 356, then the actuator will need to signal 90 degrees of rotation shown as 352 to achieve 45 degrees of rotation for desired position 2.
  • dead band is defined as the range through which an input signal can be varied, upon reversal of direction, without initiating an observable change in the output signal. Said differently, dead band results from a temporary discontinuity between the input and output of a device when the input to the device changes direction.
  • FIG. 4 graphically illustrates the dead band concept as applied to flow control valves.
  • the controller signal (“CS") is the input to the valve assembly (to move the actuator), shown as the horizontal axis 410.
  • % OPEN For vertically-actuated valves, it can also be distance traveled.
  • rotary valves it can also be expressed as % ROTATION, or in units of degrees or radians.
  • the flow through a control valve is depicted with relation to the displacement, D, of the control valve vertical stem or rotary shaft. It can be typified as either linear 500 or non-linear 510. It can be expressed as volumetric flow, shown as right-hand vertical axis 530, or as percent of total available flow through the valve, shown as left-hand axis 520.
  • the valve coefficient, C v is the ratio of the flow rate through the valve over the displacement of the valve.
  • C v is dependent on a number of factors, including the condition of the regulating elements of the valve and associated components, such as the valve seat. For example, in a gate valve, the edge of the gate can be worn-away by the abrasive action of fluids or solids. Wear over time can cause C v to change. Gontrol algorithms often utilize the new, unused, as- installed C v values as parameters, since they interact with control algorithm factors such as gain. For liquid applications, C v is often expressed as the number of gallons per minute of 60 0 F water that will flow through a particular valve with a one pound per square inch pressure drop.
  • One equation, as a non-limiting example of a flow model for a control valve, which relates flow through a valve to the valve coefficient is:
  • Figure 6 shows changes in control signal where a valve has a dead band as shown in 620.
  • the process controller is calling for changes in the process variable of flow, but because the dead band is so great in relation to the finer calls of control, the calls are never acted on. The net result is precision and responsiveness in overall process control is lost. Further, as wear or adjustments occur over time, the dead band increases, and process control gets increasingly less effective.
  • Figure 7 shows the error effect of quantization as analog signals are converted to digital signals ("A/D" conversion).
  • the signaled flow rate from a flow meter can be a 5 to 20 milliamp analog electrical signal 700, which is a continuous range of values.
  • the signal is now represented by a relatively small set of discrete symbols or integer values shown as digitized signal 710.
  • Quantization can introduce error into signal processing because signal definition is lost between the different digital intervals, as shown in 720.
  • the true analog signal increase from 720A to just under 720B, the converted digital signal value remains the same. This amount of signal definition is thus lost. Because most modern control systems utilize A/D converters, further error is thus often introduced into control algorithms.
  • the invention relates to systems and methods for managing flow control valves in process systems.
  • the operational condition of a valve can be managed based at least on changes in the valve dead band.
  • a non-linear dynamic model can determine in real-time the present valve dead band by modeling the actual flow through the valve and the commanded drive signal to the valve.
  • the present dead band can be used to update the valve control algorithm and to make a maintenance decision regarding the valve.
  • the present flow coefficient can also be determined in real-time by removing the dead band from the drive signal and using that modified control signal in a flow rate model of the valve to adaptively find the present value of the coefficient with reference to the actual flow through the valve.
  • the flow coefficient can also be used to update the valve control algorithm and to make a maintenance decision regarding the valve.
  • the disclosed ideas are used in real-time to monitor the mechanical condition of a flow control valve acting to control the flow rate of a flow stream in a process system.
  • the disclosed ideas are used in real-time to monitor the slack in the linkage in a flow control valve and/or the degree of wear in the flow-restricting elements of a flow control valve acting to control the flow rate of a flow stream in a process system.
  • the disclosed ideas are used in real-time to update the control algorithm of a flow control valve acting to control the flow rate of a flow stream in a process system.
  • the disclosed ideas are used to perform predictive maintenance on a flow control valve acting to control the flow rate of a flow stream in a process system.
  • a system for managing the operational condition of a flow control valve in a process system comprising: a flow control valve acting on a flow stream; a sensor configured to measure the flow rate of said flow stream; a non-linear dynamic model for determining the present value of the dead band of said control valve based at least on the control signal to said control valve and the flow measurement of said sensor; and an output of said present dead band value as an indication of the operational condition of said valve.
  • a system for managing the operational condition of a flow control valve in a process system comprising: a flow control valve acting on a flow stream; a sensor configured to measure the flow rate of said flow stream; a computer system connected to said valve and said sensor; and configured to provide: non-linear dynamic modeling of the present value of the dead band of said valve based at least on the control signal to said valve and the i flow rate measured by said sensor; and ' an output of said present dead band value as an indication of the operational condition of said flow control valve.
  • a system for controlling the flow rate of a flow stream comprising: a flow control valve acting on a flow stream; a computer system connected to said valve and said sensor; and configured to provide:
  • a flow control valve in a process system comprising:
  • a method for performing predictive maintenance on a flow control valve in a process flow stream comprising:
  • a flow control valve being used to regulate the flow rate of a flow stream without taking said valve off-line from acting on said flow process stream; and (b) performing maintenance on said valve based at least on (a).
  • Figure 1 shows one embodiment of the present innovations for a control method and system for managing flow control valves in process systems.
  • Figure IA shows a preferred embodiment of the present innovations for a control method and system for managing flow control valves in process systems.
  • Figure 2 shows another embodiment of the present innovations for a control method and system for managing flow control valves in process systems using an observer rather than a sensor for determination of the actual flow rate of a process flow stream.
  • Figure 3 shows a rotary-actuated flow control valve as an example of a flow control valve that can be managed using at least one of the present innovations.
  • Figure 4 shows a graphical depiction of dead band in a generalized process control action which can be managed using at least one of the present innovations.
  • Figure 5 shows a graphical depiction of flow control valve action which can be managed using at least one of the present innovations.
  • Figure 6 shows a graphical depiction of an effect of dead band in the process control of fluid flow using a flow control valve which can be managed using at least one of the present innovations.
  • Figure 7 illustrates the effect of analog-to-digital (AfO) signal conversion and quantization that can be managed using at least one of the present innovations.
  • AfO analog-to-digital
  • Figure 8 shows results for flow rate measurements and estimations from Example 1 using a preferred embodiment of the present innovations.
  • Figure 8A shows results for flow rate measurements and estimations from
  • Figure 9 shows results for dead band determinations as "valve slop" from Example 1 using a preferred embodiment of the present innovations.
  • Figure 10 shows one embodiment of a general purpose computer system suitable for implementing the control systems and methods of the present innovations.
  • these innovative concepts include, in one embodiment, non-linear modeling of the dead band dynamics of a control valve as expected in normal operation in real time.
  • the present innovations include embodiments that uses an adaptive parametric controller to determine the flow valve coefficient of a flow control valve during operation in real time.
  • FIG. 1 shows a generalized embodiment of the methods and systems of the present innovations.
  • a flow control valve 3 can be controlled by control system 1 with control signal 5.
  • a flow stream 2 can enter valve 3 and also pass through flow rate sensor 4, which can be a flow meter of a type as known to one skilled in the art of flow meter design and selection. Note that flow rate sensor can be located prior to valve 3, with little or no effect on the benefits of the present innovations.
  • Stage 4 can provide flow rate signal 6 to control system 1, to non-linear dead band modeling stage IB-I, and to a summing stage for use in determining the present value of the flow valve coefficient.
  • Stage IB-I can model and determine the present value of the dead band of valve 3, shown as dead band signal 7, based on the current control signal
  • Dead band removal stage 1B-2 can remove the dead band determined in Stage IB-I from control signal 5, and then output the modified control signal to valve coefficient stage 1B-3.
  • Stage 1B-3 can determine the valve coefficient based on the modified control signal and the output of adaptive parametric controller stage 1B-5.
  • Valve dynamics stage 1B-4 can calculate the predicted flow rate from a physical flow model of control valve 3, and the predicted rate, based on a given flow coefficient, can be compared in a summation stage against the actual flow rate from sensor 4. The difference can be. caused by a change in the given valve coefficient.
  • Adaptive stage 1B-5 can process the difference to determine the present valve coefficient by driving the difference from said summing stage to zero.
  • the systems and methods of the present innovations can thus regulate flow stream 2 as action IA.
  • the innovations can determine the present value of the dead band and valve dead band as action IB.
  • the innovations can then update the control algorithms in control system 1 as action 1C.
  • the innovations can also make a' prediction and/or decision on valve mechanical failure as action ID by comparing the present values of the dead band and valve coefficient against various decision criteria as generally known to one skilled in the art of flow control valve maintenance.
  • a maintenance decision criteria can be to perform maintenance when ,the dead band exceeds 15% of total control signal span.
  • Figure IA shows a preferred embodiment of a control system and method for the present innovations as applied to flow control of a process stream in a continuous process.
  • Control system 10 can receive a set point 5A for the desired flow rate of flow stream 7A.
  • Control system 10 can operate on and command flow control valve 20 to physically control the flow rate of flow stream 7A to that set point using control drive signal 15.
  • a flow rate sensor 30 can sense the actual rate and feed-back the value to control system 10 using signal 35. Note that flow sensor 30 can alternatively be placed before flow control valve 20 without any significant effect on the benefits of the present innovations.
  • Control system 10 can utilize a variety of control algorithms to effect control of the flow rate as known to one skilled in the art of process control.
  • the actual flow rate value as signal 37 can also be sent to summing stage 90 to be compared against the output of flow control valve dynamics calculation stage 70, as described later.
  • the actual flow rate value signal can also be sent to a non-linear modeling system 40 for determination and modeling of the dead band, e.g. mechanical slack, associated with flow control valve 20.
  • Modeling system 40 can receive the commanded control drive signal 15, and can then determine the dead band based on that signal and signal 37. ..- , .
  • Stage 40 can operate as follows.
  • a non-linear modeling system can be used to determine the dead band. This' model assumes a dead band behavior and uses the drive inputs and flow outputs to characterize this dead band. Due to quantization errors, this is accomplished in two steps.
  • the dead band region is approximated by looking for directional changes in the valve and directly estimating the size of the dead band region.
  • an adaptive system is used to further refine the dead band by comparing the error of the valve model with the actual system and changing the estimate of the dead band such that the error goes to zero.
  • the error of the valve model and actual system is pre-filtered to only adapt when the system travels through the dead band zone so it does not adapt to changes in the flow characteristics of the valve due to supply or pressure changes. Due to the quantization error on the embodied system, without the adaptive component it can have an error up to 40% depending on present valve wear.
  • a neural network can also be used to determine the dead band.
  • the present value of the dead band can be represented as signal 45.
  • Stage 50 can then receive and modify control signal 15 for the effect of the dead band by removing the dead band value from the control signal.
  • the effects of the dead band can removed before the valve coefficient is determined in stage 60 so the system does not erroneously fit the nonlinear dead band into the valve coefficient calculation.
  • the dead band is virtually removed from the system before the valve coefficient is determined, and as the control element dead band increases or decreases due to wear or adjustment, stage 50 can thus proactively correct for the change.
  • the corrected or modified control signal can then be inputted into stage 60 to determine the flow coefficient for flow control valve 20 based on the modified signal and the output of adaptive controller stage 80.
  • This system can allow for the monitoring of the valve without regard to the dead band, therefore enabling the detection of failure in the supply system.
  • Stage 60 can use adaptive parametric controller 80 to find the best fitting valve coefficient. This can also be done using a least squares method or other ways to curve- fit the data to a "line of best fit” or statistical "best fit” as is well known to a skilled -person in statistics. For, example, “best fit” methods are routinely applied to calculating values of constants in equations to allow equations, or "lines of best fit” to represent data sets with minimum error. Three illustrative definitions of the "line of best fit" are: (1) The University of Cambridge (England) website
  • the line is characterized by two features, slope and intercept.
  • the formula for the best-fit straight line can be used to predict one variable from another. Available on the Internet at: http://www.st- andrews.ac.uk/academic/psychology/teaching/ glossary.shtml#L, last visited on October 17, 2006.
  • the adapted flow coefficient can be inputted in flow control valve dynamic model stage 70.
  • Stage 70 can calculate the flow rate of flow stream 5 using a flow model of valve 20.
  • adaptive controller 80 can adjust the present value of the coefficients or constants within the dynamic model to achieve substantially a zero difference from stage 90.
  • stage 40 and stage 60 can feedback present values of the dead band 45A and the flow control valve coefficients 85, respectively, to control system 10.
  • Control system 10 can use those present values to update its control algorithm or algorithms to improve its control over flow control valve 20.
  • Figure 8 shows the recorded (e.g. "measured") data for the actual flow rate as measured by the actual flow meter, represented in Figure IA as flow rate sensor 30, and as regulated by the actual rotary control valve (flow control valve 20), which was being signaled by an automatic controller (within control system 10).
  • Figure 8 also shows a simulated flow rate as measured by the actual flow meter, represented in Figure IA as flow rate sensor 30, and as regulated by the actual rotary control valve (flow control valve 20), which was being signaled by an automatic controller (within control system 10).
  • Figure 8 also shows a simulated
  • stage 40 Because the non linear model of stage 40 is a learning model, approximately 3000 seconds (about 50 minutes) were required to converge on the valve dead band at 5.1%, from an initial estimate of zero at zero seconds. In actual on-going use of the present innovations, this learning period would thus take an initial 50 minutes and then not be repeated again, for example, until the valve was removed from service, maintained, and then re-installed and started-up again. Additionally, measure-able wear in mechanical linkages generally occurs over days or weeks of control valve use, not in a matter of less than an hour.
  • Figure 8A shows three enlarged areas from three time periods in Figure 8 to illustrate the increased accuracy of control valve characterization as non-linear model 40 converged on the value of the valve dead band using the systems and methods of the present innovations.
  • the periods were Period 1, from about 500 to 1000 seconds, Period 2, from about 2000 to 2500 seconds, and Period 3 (near the end of the total run), from about 3500 to 4200 seconds.
  • the measured flow rate (signal 37) is individually apparent and distinct from the calculated flow rate (e.g. "estimated” in Figure 8) and has periods of relatively constant flow rates as flow rate set points were frequently changed to various desired flow rates and the controller responded by driving the control valve to achieve the new desired rate.
  • the calculated flow rate represented in Figure IA as the output of stage 70 going into summing stage 90, is doing a relatively poor job of matching the measure flow rate.
  • the calculated flow rate is much closer to the actual flow rate from Period 1, and in Period 3, they are virtually indistinguishable.
  • the methods and systems of the present innovations increased in accurate monitoring of the operational condition of the valve as evidenced by increased accuracy in the calculated flow rate relative to the actual flow rate.
  • the present innovations would be part of the control system operating a control valve.
  • the present innovations so utilized can map the wear of the control valve over time and in real time, such that a decision can be made to perform maintenance.
  • Figure 2 is another embodiment of the present innovations applied to a solids flow stream 25 in which flow control valve 220 is a solid-handling valve or metering device instead of a fluid as.in Figure IA.
  • stream 25 is being continuously charged into a continuous mixing stage 230, which is also receiving a separate fluid flow stream, such as water (not shown), which can be easily measured for flow rate.
  • the flow rate is not easily measured.
  • the solids flow rate can be inferred from the use of a density observer 245 using density readings from a density sensor 240, because in this case, the specific density of the solid is different than the specific density of the fluid. See U.S. Patent Application 11/121,144, referred to previously, for details of how the density observer can provide signal 37 as an observed value of the solids flow rate of stream 25.
  • Figure 2 operates essentially the same as Figure IA.
  • a system for managing the operational condition of a flow control valve in a process system comprising: a flow control valve acting on a flow stream; a sensor configured to measure the flow rate of said flow stream; a non-linear dynamic model for determining the present value of the dead band of said control valve based at least on the control signal to said control valve and the flow measurement of said sensor; and an output of said present dead band value as an indication of the operational condition of said valve.
  • a system for managing the operational condition of a flow control valve in a process system comprising: a flow control valve acting on a flow stream; a sensor configured to measure the flow rate of said flow stream; a computer system connected to said valve and said sensor; and configured to provide: non-linear dynamic modeling of the present value of the dead band of said valve based at least on the control signal to said valve and the flow rate measured by said sensor; and an output of said present dead band value as an indication of the operational condition of said flow control valve.
  • a system for controlling the flow rate of a flow stream comprising: a flow control valve acting on a flow stream; a computer system connected to said valve and said sensor; and configured to provide: (i) a control signal to a flow control valve acting on a flow stream; (ii) a determination of the present flow rate of said flow stream; (Hi) a determination of the; present dead band of said valve based at least on (i) and (ii); (iv) a determination of the present flow coefficient of said valve based at least on (i), (ii), and (iii); and (v) an update to a controller providing said control signal to said control valve based at least on the results of actions (iii) and (iv).
  • a method for managing the operational condition of a flow control valve in a process system comprising: (a) regulating the flow rate of a flow stream with a flow control valve acting on said flow stream; (b) determining in real-time the present value of the dead band of said valve; and (c)updating the controller providing the control signal to said control valve based at least on the results of action (b).
  • a method for performing predictive maintenance on a flow control valve in a process flow stream comprising: (a) determining in real-time the present value of the dead band and the present value of the flow coefficient of a flow control valve being used to regulate the flow rate of a flow stream without taking said valve off-line from acting on said flow process stream; and (b) performing maintenance on said valve based at least on (a).
  • a method for controlling the flow rate of a flow stream comprising: (a) providing a control signal to a flow control valve acting on a flow stream; (b) determining the present flow rate of said flow stream; (c) determining the present dead band of said valve based at least on actions (a) and (b); (d) determining the present flow coefficient of said valve based at least on actions (a), (b), and (c); and (e) updating a controller providing said control signal to said control valve based at least on the results of actions (c) and (d).
  • the methods and systems of the present application can be configured or combined in various schemes.
  • the combination or configuration depends partially on the required flow rate control precision and accuracy and the operational envelope of the process flow stream.
  • One of ordinary skill in the art of process control, with- the benefit of this disclosure will- recognize the appropriate combination or configuration for a chosen application.
  • flags such as a particular process variable out of range which may define the reliability of the data or provide variables to use for process control.
  • One of ordinary skill in the art, with the benefit of this disclosure will recognize the appropriate additional measurements that would be beneficial for a chosen
  • such measurements taken by the methods and systems of the present application may also be sent to an external computing and analysis system for further processing or use.
  • control system 10 application, or across different embodiments, and therefore different reference data sets or curves or models fitted to such data sets may be employed, maintained, or stored in control system 10 or an external computing and analysis system which can communicate with control system 10.
  • control system 10 or an external computing and analysis system which can communicate with control system 10.
  • the methods' and systems of the present innovations can be implemented on general-purpose computers or laptop computer or microprocessor system, or an external computing and analysis system, in addition to being embodied in manufacturing control hardware, as long as such embodiments possess adequate 20 computing resources, memory, and communication capacity to perform the necessary operations requested of them.
  • Figure 10 shows one embodiment of such a computer system 1000 for implementing one or more embodiments of the methods and systems of the present innovations.
  • system 1000 includes central processor unit (CPU) 25 1100 which can communicate with various system devices via communications BUS
  • CPU 1100 can execute codes, instructions, programs, and scripts which it accesses from various disk based systems which can be secondary storage 1200, ROM 1300, RAM 1400, or the network communication components 1600.
  • the set of instructions to CPU 1100 can comprise input instructions that receives data or models from an external system.
  • system 1000 can have more than one CPU chip to increase computing power and resources.
  • various system devices can include memory devices such as secondary storage 1200, read only memory (ROM) 1300, random access memory (RAM) 1400.
  • System 1000 can connect to other systems such as the systems of the present innovations via input/output (I/O) components 1500 and network or communication components 1600.
  • I/O input/output
  • network or communication components 1600 the signal outputs from system 1000 to the flow control valve 20 of
  • Figure IA (or valve 220 in Figure 2) can be converted from a digital to an analog signal by a digital to analog converter (DAC) 1700.
  • DAC digital to analog converter
  • additional signal conditioning can be conducted on system 1000 output signals to appropriately communicate with various control elements and actuators such as flow control valves 20 or 220.
  • secondary storage 1200 can comprise one or more disk drives or tape drives for permanent storage of data and as extra memory if RAM 1400 is not of sufficient capacity for a given operation. Secondary storage 1200 can store programs that are loaded into RAM 1200 if such programs are selected for running.
  • ROM 1300 can store instructions and data that can be read during the running of programs. ROM 1300 is a non-volatile memory device.
  • RAM 1400 can be used to store data and to store computing instructions. Speed of access to ROM 1300 and RAM 1400 can be faster than to secondary storage 1200.
  • input/output components 1500 can include video monitors, printers, touch screen displays, liquid crystal displays, keyboards, keypads, on-off buttons, dials, mouse pointing devices, track balls, voice recognizers, card readers, tape readers, and various combinations thereof.
  • network communications components 1600 can be ethernet cards, universal serial bus interface cards, serial interfaces, token ring cards, fiber distributed data interface cards, modems, modem banks, wireless local area network cards, radio transceiver cards such as "Global System for Mobile Communications" radio transceiver cards, and various combinations thereof.
  • components 1600 can enable CPU 1100 to communicate with an Internet or with intranets.
  • CPU 1100 can receive information from the nets, or can output information to the nets.
  • Such information can be a computer data signal embodied in a carrier wave or a baseband signal.
  • the baseband signal or signal embedded in a carrier wave, or other types of signals currently used or hereafter developed, can be generated according to several methods well known to one skilled in the art.
  • RS-422 or RS-485 can be used to allow links to control system 10 or to an external computing and analysis system, or to multiple external units.
  • a 4-20 milliamp analog output signal can be used to allow external processing of the system measurements.
  • the methods of the present invention can be embodied in a computer readable medium, including a compact disk.

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)
  • Flow Control (AREA)

Abstract

L'invention concerne des systèmes et des procédés qui permettent de gérer l'état opérationnel de soupapes de régulation de débit dans des systèmes de traitement. Au fur et à mesure qu'une soupape de régulation de débit subit une usure mécanique au cours de son fonctionnement, les modifications physiques qui touchent la soupape peuvent altérer sa zone morte et son coefficient d'écoulement. Un modèle dynamique non linéaire permet de déterminer la zone morte actuelle, par modélisation de la relation entre l'écoulement réel à travers la soupape et le signal de commande appliqué à la soupape. On peut mesurer le coefficient d'écoulement de soupape actuel en soustrayant la zone morte du signal de commande et en utilisant le signal modifié dans un modèle de débit de soupape afin de trouver de manière adaptative la valeur de coefficient d'écoulement qui permet de faire correspondre l'écoulement prédit à partir du modèle de débit avec l'écoulement actuellement mesuré par un capteur de débit. On peut utiliser la zone morte et le coefficient de débit actuels pour actualiser des algorithmes de régulation de soupape et prendre des décisions concernant l'entretien de la soupape.
PCT/GB2008/000229 2007-01-31 2008-01-23 Systèmes et procédés de gestion de soupapes de régulation de débit dans des systèmes de traitement WO2008093049A2 (fr)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
US11/700,397 US7636614B2 (en) 2007-01-31 2007-01-31 Systems for managing flow control valves in process systems
US11/700,533 US7606636B2 (en) 2007-01-31 2007-01-31 Methods for managing flow control valves in process systems
US11/700,533 2007-01-31
US11/700,397 2007-01-31

Publications (2)

Publication Number Publication Date
WO2008093049A2 true WO2008093049A2 (fr) 2008-08-07
WO2008093049A3 WO2008093049A3 (fr) 2009-11-05

Family

ID=39674548

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/GB2008/000229 WO2008093049A2 (fr) 2007-01-31 2008-01-23 Systèmes et procédés de gestion de soupapes de régulation de débit dans des systèmes de traitement

Country Status (1)

Country Link
WO (1) WO2008093049A2 (fr)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0227297A2 (fr) * 1985-11-18 1987-07-01 Otis Elevator Company Ascenseur hydraulique comportant une valve motorisée programmée dynamiquement
US6286532B1 (en) * 2000-05-13 2001-09-11 Ford Global Technologies, Inc. Control system and method for controlling valve

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0227297A2 (fr) * 1985-11-18 1987-07-01 Otis Elevator Company Ascenseur hydraulique comportant une valve motorisée programmée dynamiquement
US6286532B1 (en) * 2000-05-13 2001-09-11 Ford Global Technologies, Inc. Control system and method for controlling valve

Also Published As

Publication number Publication date
WO2008093049A3 (fr) 2009-11-05

Similar Documents

Publication Publication Date Title
US7606636B2 (en) Methods for managing flow control valves in process systems
US7636614B2 (en) Systems for managing flow control valves in process systems
US10494906B2 (en) Method and a control system for optimizing production of a hydrocarbon well
Chai et al. Optimal operational control for complex industrial processes
US11782401B2 (en) Apparatus and methods to build deep learning controller using non-invasive closed loop exploration
CN101484858B (zh) 流量控制阀的控制方法和控制系统
JP4680930B2 (ja) ウェーブレット分解を用いてバルブの可能性のある欠陥標識を識別する装置及び方法
US20200033895A1 (en) Pressure-type flow control device and flow control method
Kordestani et al. A modular fault diagnosis and prognosis method for hydro-control valve system based on redundancy in multisensor data information
CN107574785B (zh) 流体管道网络的需求管理和控制的方法
Al-Qutami et al. Development of soft sensor to estimate multiphase flow rates using neural networks and early stopping
US20160314409A1 (en) Method and system for real time production optimization based on equipment life
Nguyen et al. Remaining useful lifetime prognosis of controlled systems: a case of stochastically deteriorating actuator
RU2695468C2 (ru) Переход к закрытию для позиционеров клапанов управления
Jordanou et al. Online learning control with echo state networks of an oil production platform
Nguyen et al. Feedback control system with stochastically deteriorating actuator: remaining useful life assessment
US10338543B2 (en) Integrated model predictive control and advanced process control
WO2008093049A2 (fr) Systèmes et procédés de gestion de soupapes de régulation de débit dans des systèmes de traitement
WO2004025384A1 (fr) Procede et appareil de determination d'une hysteresis
JP2007514999A (ja) クッション・フィルタを用いてバルブを動作させる装置及び方法
Sanzo' et al. Virtual metering and allocation using machine learning algorithms
Sivagamasundari et al. A new methodology to compensate stiction in pneumatic control valves
Malaterre Control of irrigation canals: why and how?
Nguyen et al. A deterioration model for feedback control systems with random environment
Mohammadzaheri et al. Efficient neuro-predictive control of a chemical plant

Legal Events

Date Code Title Description
NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 08701903

Country of ref document: EP

Kind code of ref document: A2