Patent Application
20190101141
The present invention generally relates to control valves and, more particularly, to methods and apparatus for controlling a double-acting pneumatic actuator for a process control valve.
Many industrial processes use control valves to control the flow rates of fluids through pipes. These control valves are opened and closed by actuators, the positions of which are set by positioners using feedback from actuator position sensors and process settings. The goal of the positioner and actuator combination is to quickly and accurately control actuator position and to minimize deviations in the actuator position in response to the forces generated by the process fluids flowing through the valve.
A double-acting pneumatic actuator changes position by adjusting pressures in two pneumatic chambers, where the pressures push on a piston connected to a stem. The stem, in turn, translates the motion of the piston to adjust an opening of a control valve to change the flow of a process fluid. When the forces from the flow of process fluid move the actuator, the process may be adversely affected, and additional wear on actuator components may reduce the life of the actuator. Stiffness of the actuator presents an engineering trade-off between being able to rapidly control the actuator position and limiting the effect of the buffeting forces in the valve on fluctuations in the position of the actuator. Reducing undesired variations in stiffness of valve actuators would improve the quality and the durability of the process control system.
Implementations of methods and corresponding systems of this disclosure can control the position of a double-acting pneumatic actuator in view of constraints set on the stiffness of the actuator. The system can increase the stiffness by increasing the opposing pressures in the two pneumatic chambers and decrease the stiffness, conversely, by decreasing the two pressures. Thus, by controlling the two pneumatic signals to the actuator chambers, the disclosed methods can change the position of the actuator while simultaneously increasing or decreasing the stiffness.
The system can be configured to measure the pressures in the two chambers and use the measurements to compute a numerical indicator of stiffness. A control loop can be configured to use the numerical indicator of stiffness as one of the outputs to control, along with controlling the position of the actuator.
The method for controlling the actuator may include: obtaining a constraint on a numerical indicator of the stiffness of the actuator and obtaining a set point for the position of the actuator, measuring the position of the actuator, computing the numerical indicator of the stiffness of the actuator, computing control signals to adjust the pressures in the two pneumatic chambers so as to minimize an error in the position of the actuator in view of the constraint on the numerical indicator of the stiffness of the actuator, and/or activating pneumatic devices to adjust the pressures in the pneumatic chambers of the actuator in response to the control signals.
Alternatively, the method can use the measured pressures (without necessarily calculating the indicator of the stiffness) by: obtaining a set point for the position of the actuator, measuring the position of the actuator, measuring sensor outputs indicative of the pressures in the two chambers of the actuator, using measured position and the two measured pressures in computing control signals to adjust the pressures in the two pneumatic chambers so as to minimize the error in the position of the actuator, and/or activating pneumatic devices to adjust the pressures in the pneumatic chambers of the actuator in response to the control signals.
Different implementations may define the constraint on the stiffness of the actuator or constraints on the pressures in the pneumatic chambers of the actuator in different ways. Some implementation may define the constraints on the pressures in the chambers by setting minimum and/or maximum pressure values for one or each of the chambers, setting target pressures (i.e. set points) for the pressures in one or each of the chambers, or defining an indicator as a mathematical combination of the pressure values (such as an average) in the two chambers, and giving the indicator a corresponding set point or an acceptable range. Some implementations may define the numerical indicator of the stiffness of the actuator as a mathematical combination (such as a weighted sum) of the pressure values in the two chambers, or include position measurements in the definition and the computation of the indicator. These implementations may define corresponding constraints as acceptable ranges in the indicators of the stiffness, or as a set point in the indicators of the stiffness.
Different implementations may compute control signals in different ways. In some implementations, the constraints may set whether the pressure in a given chamber will increase or decrease, and the position error will set the corresponding magnitudes of the increase or the decrease. Alternatively, multi-input multi-output (MIMO) control loops can use an error in the indicator of the stiffness or an error in pressure, along with an error in position, to compute the control signals for each of the two chambers. These computations may comprise calculating a weighted sum of the error in the position of the actuator, a velocity of the actuator, and the error in the numerical indicator of the stiffness of the actuator (or the error in a chamber pressure).
Additionally, different implementations can translate the computed control signals into pneumatic signals generated by pneumatic devices in different ways. In some implementations, the magnitudes of the control signals may set the magnitudes of flow rates or pressures generated by the pneumatic devices. In other implementations, the control signals may activate each of the pneumatic devices to provide a constant flow rate for a duration consistent with the magnitude of the corresponding control signal.
An implementation of the double-acting actuator 1 can also contain a spring (not shown) in one of the chambers for fail-open or fail-closed action. The spring would place the actuator at one limit of the actuator's range when, for example, the chambers depressurize due to a leak. The spring can offset the balance between the two chamber pressures.
In some implementations, a mechanism in an alternative actuator can translate the linear motion of the pistons into rotary motion of the stems by means of rack and pinion, scotch yoke, or another mechanism. These implementations can contain more than one piston and more than two chambers.
Position sensor 11 is configured to detect the position of the actuator 1. While in the illustrated implementation this position indicates a linear displacement of the stem 5, in implementations with an alternative rotary actuator, an alternative sensor can be configured to measure angular displacement of some portion of the alternative rotary actuator. Pressure sensors 14a and 14b are configured to indicate the pressures in chambers 2 and 3, respectively. The pressure sensors can be located at the outlet ports 6a-b for the chambers. The sensors can also be integrated into the body of the positioner 10 at locations which are fluidly connected to the chambers 2 and 3 by pneumatic lines. Sensors 11, 14a and 14b communicatively couple to the processing unit 16.
The processing unit 16 is configured to detect outputs of position and pressure sensors 11, 14a and 14b. The processing unit 16 is communicatively coupled to the interface 18. In one implementation, the processing unit 16 can comprise a microprocessor. In other implementations, the processing unit 16 can comprise field programmable gate arrays (FPGAs) or analog circuits. The processing unit 16 is configured to output electrical control signals to the pneumatic devices. The processing unit 16 can also be configured to compute other signals, such as diagnostic information about the positioner and the actuator.
The interface 18 can include wired and wireless connections, circuitry for communications and signal processing, non-transient memory and a human-machine interface. Thus, the processing unit 16 can obtain set points and constraints for controlling the actuator in a variety of ways. The set point of position may be a dynamically changing value which is communicated by a process controller to the interface 18 using a predetermined communication protocol. The constraint on the stiffness of the actuator may or may not change dynamically. In implementations with preset constraints on the stiffness, the processing unit 16 may use the interface 18 to retrieve the stiffness constraint by accessing nonvolatile memory where the constraint can be stored.
The implementation shown in
In operation, the processing unit 16 communicates to the interface 18 to obtain the set point of position as well as a constraint on the stiffness of the actuator 1. The processing unit 16 also reads the sensors 11, 14a and/or 14b to obtain the displacement of the actuator and/or the pressures in the chambers 2 and 3 of the actuator 1. For any given implementation, the processing unit 16 can compute a numerical indicator of the stiffness of an actuator from the collected sensor data in accordance to the way that the constraint on the numerical indicator of the stiffness of the actuator is defined. In one implementation, the constraint on the stiffness of the actuator can be defined as an acceptable range for the average of the pressures in the two chambers 2 and 3. In such implementation, the processing unit 16 computes the average of the two pressures measured by the pressure sensors 14a-b and compares it to the constraint. The processing unit 16 then computes two control signals and communicates the signals to activate two out of the four I/P transducers 20a-d.
Only one transducer needs to be activated for each pneumatic chamber at a given time in order to change the pressure in the chamber. Therefore, two control signals computed by the processing unit 16 are sufficient, as they can activate two of the four transducers 20a-d for each new control action. I/P transducers 20a and 20b activate, respectively, the supply and exhaust of the upper pneumatic chamber 2 of the actuator 1. I/P transducers 20c and 20d activate, respectively, the supply and exhaust of the lower pneumatic chamber 3 of the actuator 1. In one implementation, positive control signals can activate 20a and 20c, while negative control signals can activate 20b and 20d. The pneumatic amplifiers 24a-d amplify the flow rates of the transducers. The pneumatic summation components 27 and 28 connect the flow to the outlets of the pneumatic chambers 2 and 3. Active I/P transducer 20a results in control fluid flowing into chamber 2. Active I/P transducer 20b results in control fluid flowing out of chamber 2. Active I/P transducer 20c results in control fluid flowing into chamber 3. Active I/P transducer 20d results in control fluid flowing out of chamber 3. Another implementation may use only two I/P transducers if each is configured to control bi-directional flow.
At block 101, the stiffness constraint is obtained. In one implementation, the stiffness constraint can be preset by the manufacturer of the positioner 10, while in another it can be preset by any operator with appropriate access privileges. Additionally, an operator or engineer can change the stiffness constraint in response to process requirements or changes detected in the process control equipment. The changes in the stiffness constraint can be communicated to the interface 18 by means of a predetermined communication protocol. The stiffness constraint may be specified in terms of a constraint on a numerical indicator of stiffness, the indicator calculated in a predetermined way depending on implementation. In one implementation, this indicator can be specified to reflect variations of actuator position, as measured by the position sensor 11. In another implementation, the indicator can be specified in terms of a combination, such as a weighted sum, of the pressures in the two chambers 2 and 3, measured by the pressure sensors 14a-b. Also, the position of the actuator can be combined with the pressures to arrive at an indicator of stiffness. In some implementations, the constraint on the indicator of stiffness may be dependent on available supply pressure.
At block 104, the processing unit 16 obtains the set point of the position. The set point can be the most recent value communicated to the interface 18 by a process controller or it can be a value stored by the interface in response to operator input. At block 108, the processing unit 16 reads the current value of the measured position from sensor 11. At block 112, the processing unit 16 computes the numerical indicator of stiffness. This computation corresponds to the way that the constraint on stiffness is specified. In one implementation, for example, the numerical indicator of stiffness can be calculated as an average of the pressures in the two chambers 2 and 3, and the constraint can be specified as an acceptable range of this average. In such implementation, block 112 comprises reading the pressures measured by sensors 14a and 14b, from which the average is computed. Another implementation may compute the numerical indicator of stiffness by adding the weighted pressures in the two chambers 2 and 3 in order to account for some asymmetry in the actuator. Such asymmetry could be due to the presence of springs in the actuator, or because of the difference in the area of the piston exposed to pressure in each chamber. For an implementation in which the numerical indicator of stiffness is computed by other means, only the sensors which measure the quantities involved in the computation need to be read, though other sensors may also be read. For example, the processing unit 16 can collect statistics on the variation of the position of the actuator in the absence of control signals in order to compute the numerical indicator of stiffness.
At block 116, the processing unit 16 computes the control signals. The processing unit 16 uses the computed control signals to activate the I/P transducers at block 120. In the illustrated implementation of
Control signals at block 116 can be calculated in a variety of different ways in different implementations. In a simple implementation, the pressures in both chambers 2 and 3 are reduced by a control action if the computed indicator of stiffness is too high with respect to the constraint, and, conversely, the pressures in both chambers are increased by a control action if the computed indicator of stiffness is too low with respect to the constraint. In such implementation, the change in position is set by the difference in the rates at which the two pressures are raised or lowered. In another implementation, the stiffness constraint obtained in 101 is a set point for the numerical indicator of stiffness, and the control signals are computed by a control algorithm which treats this set point as another input to the control loop, along with actuator position. Because there are two control signals as well as two input signals, this implementation employs Multiple Input Multiple Output (MIMO) control.
The processing unit 16 obtains the measurements of the position of the actuator and the pressures in the chambers at blocks 208 and 212, respectively. At block 216, the processing unit 16 computes the control signals based on the position and pressure measurements, as well as the associated set points or constraints. In one example, if a pressure measurement in chamber 2 of
The control loop for implementing the MIMO implementation of the process in
The control signals can be computed as weighted sums of the errors in position and the stiffness indicator. In another implementation of this computation, one can add a factor for the velocity of the actuator, which can also be interpreted as the rate of change in the position error and can be computed using the difference in consecutive measurements of position. The resulting control signal can then be written as:
C
a
=K
p,a
e
x
−K
v,a
e
{dot over (x)}
−K
s,a
e
s
C
b
=−K
p,b
e
x
+K
v,b
e
{dot over (x)}
−K
s,b
e
s
where Ca is the control signal for the upper chamber, Cb is the control signal for the lower chamber, Kp,a and −Kp,b are the position feedback gains for chambers, −Kv,a, Kv,b are the velocity feedback gains, −Ks,a, −Ks,b are the feedback grains for the indicator of stiffness, ex is the error in the position, e{dot over (x)} can be interpreted as rate of change of the error in position, or simply, as the velocity of the actuator and es is the error in a numerical indicator of stiffness. In some implementations, the numerical indicator of stiffness may be replaced by a pressure indicator, and es is then replaced by the error in the pressure indicator, ep.
Control signals Ca and Cb can be computed in a variety of ways different from the weighted sum of the errors. For example, the feedback gains can change values depending on the errors or other process parameters. Also, terms proportional to errors raised to integer or non-integer powers can be included in the computation of errors.
The effect of control signals Ca and Cb on the I/P transducers depends on a given implementation. In one implementation, analog I/P transducers can be used, and the control signals can set the magnitude of flow rates into or out of the corresponding chambers, with the flow rate magnitudes, for example, proportional to the control signals. In another implementation, especially suitable for digital I/P transducers, the duration of a fixed flow rate is controlled. For example, if Ca is positive, while Cb is negative, but its absolute value is twice that of Ca, then I/P transducers 20a and 20d in
The control of flow rate durations, as opposed to the flow rate magnitudes, is particularly applicable to the implementation of a control method in which the control actions are updated at pre-defined intervals. The fraction of the interval during which a given flow path is active can be proportional to the magnitude of the corresponding control signal, the magnitude determining the duration of an electrical pulse to the associated I/P transducer.
The foregoing detailed description has been given for clearness of understanding only, and no unnecessary limitations should be understood therefrom, as modifications will be obvious to those skilled in the art. Additionally, throughout this specification, plural instances may implement components, operations, or structures described as a single instance. Although individual operations of one or more methods are illustrated and described as separate operations, one or more of the individual operations may be performed concurrently or may be performed in an alternate order to the order illustrated. Structures and functionality presented as separate components in example configurations may be implemented as a combined structure or component. Similarly, structures and functionality presented as a single component may be implemented as separate components. These and other variations, modifications, additions, and improvements fall within the scope of the subject matter herein.
Throughout this specification, actions described as performed by the processing unit 16 or other similar devices (or routines or instructions executing thereon) generally refer to actions or processes of a processor manipulating or transforming data according to machine-readable instructions. The machine-readable instructions may be stored on and retrieved from a memory device communicatively coupled to the processor. That is, methods described herein may be embodied by a set of machine-executable instructions stored on a non-transitory computer readable medium (i.e., on a memory device). The instructions, when executed by one or more processors of a corresponding device (e.g., a server, a mobile device, etc.), cause the processors to execute the method. Where instructions, routines, modules, processes, services, programs, and/or applications are referred to herein as stored or saved on a computer readable memory or on a computer readable medium, the words “stored” and “saved” are intended to exclude transitory signals.
Unless specifically stated otherwise, discussions herein using words such as “processing,” “computing,” “calculating,” “determining,” “identifying,” “presenting,” “displaying,” or the like may refer to actions or processes of a machine (e.g., a computer) that manipulates or transforms data represented as physical (e.g., electronic, magnetic, or optical) quantities within one or more memories (e.g., volatile memory, non-volatile memory, or a combination thereof), registers, or other machine components that receive, store, transmit, or display information.
When implemented in software, any of the applications, services, and engines described herein may be stored in any tangible, non-transitory computer readable memory such as on a magnetic disk, a laser disk, solid state memory device, molecular memory storage device, or other storage medium, in a RAM or ROM of a computer or processor, etc. Although the example systems disclosed herein are disclosed as including, among other components, software and/or firmware executed on hardware, it should be noted that such systems are merely illustrative and should not be considered as limiting. For example, it is contemplated that any or all of these hardware, software, and firmware components could be embodied exclusively in hardware, exclusively in software, or in any combination of hardware and software. Accordingly, persons of ordinary skill in the art will readily appreciate that the examples provided are not the only way to implement such systems.
Thus, while the present invention has been described with reference to specific examples, which are intended to be illustrative only and not to be limiting of the invention, it will be apparent to those of ordinary skill in the art that changes, additions or deletions may be made to the disclosed embodiments without departing from the spirit and scope of the invention.
| Number | Date | Country | |
|---|---|---|---|
| 62565929 | Sep 2017 | US |
