Enhanced PID (PIDE)

The Enhanced PID (PIDE) instruction provides enhanced capabilities over the standard PID instruction. The instruction uses the velocity form of the PID algorithm. The gain terms are applied to the change in the value of error or PV, not the value of error or PV.

Available Languages

Ladder Diagram

This instruction is not available in ladder diagram.

Function Block

Structured Text

PIDE(PIDE_tag);

Operands

Function Block

Structured Text

See

Structured Text Syntax

for more information on the syntax of expressions within structured text.

PIDE Structure

Description

The PID algorithm regulates the CV output in order to maintain the PV at the SP when the instruction executes in Cascade/Ratio or Auto modes.

When ControlAction is set, the calculated value of EPercent and PVPIDPercent is negated before being used by the control algorithm.

The following table describes how the instruction calculates the PID terms.

Computing CV

The PID control algorithm computes the value for CV by summing Delta PTerm, Delta ITerm, Delta

DTerm, and CV from the previous execution of the instruction, for example CV

n-1

. When CVsetPrevious is set,

CVn-1 is set equal to CVPrevious. This lets you preset CV

n-1

to a specified value before computing the value of CV.

CalculatedCV = CV

n-1

+ DPTerm + DITerm + DDTerm

PIDE Algorithms

The PIDE instruction uses a velocity form PID algorithm similar to that used in most DCS systems. Some advantages to a velocity form algorithm include:

Independent Gains Form

In this form of the algorithm, each term of the algorithm (proportional, integral, and derivative) has a separate gain. Changing one gain only affects that term and not any of the others, where:

Dependent Gains Form

This form of the algorithm changes the proportional gain into a controller gain. By changing the controller gain, you change the action of all three terms (proportional, integral, and derivative) at the same time, where:

Determining Which Algorithm to Use

The PIDE equations above are representative of the algorithms used by the PIDE instruction. You can substitute the change in error values for the change in PV (in percent of span) for the proportional and derivative terms by manipulating the parameters PVEProportional and PVEDerivative. By default, the PIDE instruction uses the change in error for the proportional term and the change in PV for the derivative term. This eliminates large derivative spikes on changes in setpoint.

You can convert the gains used between the different PIDE algorithm forms using these equations:

Each algorithm provides identical control with the appropriate gains. Some people prefer the independent gains style because they can manipulate individual gains without affecting the other terms. Others prefer the dependent gains style because they can, at least to a certain extent, change just the controller gain and cause an overall change in the aggressiveness of the PID loop without changing each gain separately.

Monitoring the PIDE Instruction

There is an operator faceplate available for the PIDE instruction.

Autotuning the PIDE Instruction

The

Logix Designer

Logix Designer

application PIDE autotuner provides an open-loop autotuner built into the PIDE instruction. You can autotune from

PanelView

terminal or any other operator interface devices, as well as the

Logix Designer

application. The PIDE block has an Autotune Tag (type PIDE_AUTOTUNE) that you specify for those PIDE blocks that you want to autotune.

The PIDE autotuner is installed with the application, but you need an activation key to enable it. The autotuner is only supported in function block programming; it is not available in ladder diagram or structured text programming.

Use the Autotune tab to specify and configure the autotune tag for a PIDE block.

Affects Math Status Flags

No

Major/Minor Faults

None specific to this instruction. See Common Attributes for operand-related faults.

Execution

Function Block

Structured Text

When CVInitReq is set, or during instruction first scan, or on a set to cleared transition of CVFault (bad to good), the instruction initializes the CVEU and CV outputs to the value of CVInitValue. If the timing mode is not oversample and EnableIn transitions from cleared to set, the instruction initializes the CVEU and CV values. CVInitialization is cleared after the initialization and when CVInitReq is cleared.

The CVInitValue normally comes from the analog output’s readback value. The CVInitReq value normally comes from the "In Hold" status bit on the analog output controlled by CVEU. The initialization procedure is performed to avoid a bump at startup in the output signal being sent to the field device.

When using cascaded PID loops, the primary PID loop can be initialized when the secondary loop is initialized or when the secondary loop leaves the Cascade/Ratio mode. In this case, move the state of the InitPrimary output and SP output from the secondary loop to the CVInitReq input and CVInitValue input on the primary loop.

The instruction does not initialize and the CVEU and CV values are not updated if CVFault or CVEUSpanInv is set.

Examples

Example 1

The easiest way to implement a PIDE instruction is to create a function block routine in a program in a periodic task. The default timing mode for the PIDE instruction is periodic. When the PIDE instruction is used in a periodic task and in periodic timing mode, it automatically uses the periodic task's update rate as its delta t update time. All you need to do is wire the process variable analog input into the PV parameter on the PIDE instruction and wire the CVEU out of the PIDE instruction into the controlled variable analog output.

Optionally, you can wire the analog input’s fault indicator (if one is available) into the PVFault parameter on the PIDE instruction. This forces the PIDE into Manual mode when the analog input is faulted and stops the PIDE CVEU output from winding up or down when the PV signal is not available.

Function Block

Structured Text

PIDE_01.PV := Local:1:I.Ch0Data;

PIDE_01.PVFault := Local:1:I.Ch0Fault;

PIDE(PIDE_01);

Local:2:)O.Ch0Data :=PIDE_01.CVEU;

Example 2

Cascade control is useful when externally-caused upsets to the controlled variable occur often, which then cause upsets to the process variable you are trying to control. For example, try to control the temperature of liquid in a tank by varying the amount of steam fed into a heating jacket around the tank. If the steam flow suddenly drops because of an upstream process, the temperature of the liquid in the tank eventually drops and the PIDE instruction then opens the steam valve to compensate for the drop in temperature.

In this example, a cascaded loop provides better control by opening the steam valve when the steam flow drops before the liquid temperature in the tank drops. To implement a cascaded loop, use a PIDE instruction to control the steam valve opening based on a process variable signal from a steam flow transmitter. This is the secondary loop of the cascaded pair. A second PIDE instruction (called the primary loop) uses the liquid temperature as a process variable and sends its CV output into the setpoint of the secondary loop. In this manner, the primary temperature loop asks for a certain amount of steam flow from the secondary steam flow loop. The steam flow loop is then responsible for providing the amount of steam requested by the temperature loop in order to maintain a constant liquid temperature.

Function Block

Structured Text

PrimaryLoop.PV := Local:1:I.CH0Data;

PIDE(PrimaryLoop);

SecondaryLoop.PV := Local:1:I.Ch1Data;

PIDE(SecondaryLoop);

Local:2:O.Ch0Data:= SecondaryLoop.CVEU;

For a cascaded pair of loops to work correctly, the secondary loop must have a faster process response than the primary loop. This is because the secondary loop’s process must be able to compensate for any upsets before these upsets affect the primary loop’s process. In this example, if steam flow drops, the steam flow must be able to increase as a result of the secondary controller’s action before the liquid temperature is affected.

To set up a pair of cascaded PIDE instructions, set the

AllowCasRat

input parameter in the secondary loop. This allows the secondary loop to be placed into Cascade/Ratio mode. Next, wire the

CVEU

from the primary loop into the

SPCascade

parameter on the secondary loop. The

SPCascade

value is used as the SP on the secondary loop when the secondary loop is placed into Cascade/Ratio mode. The engineering unit range of the CVEU on the primary loop should match the engineering unit range of the PV on the secondary loop. This lets the primary loop scale its 0-100% value of CV into the matching engineering units used for the setpoint on the secondary loop.

The PIDE instruction supports several other features to more effectively support cascade control. Wire the

InitPrimary

output on the secondary loop into the

CVInitReq

input on the primary loop and wire the SP output of the secondary into the

CVInitValue

input on the primary. This sets the CVEU value of the primary loop equal to the SP of the secondary loop when the secondary loop leaves Cascade/Ratio mode. This allows a bumpless transfer when you place the secondary loop back into Cascade/Ratio mode. Also, wire the

WindupHOut

and WindupLOut outputs on the secondary loop into the

WindupHIn

and

WindupLIn

inputs on the primary loop. This causes the primary loop to stop increasing or decreasing, as appropriate, its value of CVEU if the secondary loop hits a SP limit or CV limit and eliminates any windup on the primary loop if these conditions occur.

Example 3

Ratio control is typically used to add a fluid in a set proportion to another fluid. For example, if you want to add two reactants (say A and B) to a tank in a constant ratio, and the flow rate of reactant A may change over time because of some upstream process upsets, you can use a ratio controller to automatically adjust the rate of reactant B addition. In this example, reactant A is often called the "uncontrolled" flow since it is not controlled by the PIDE instruction. Reactant B is then called the "controlled" flow.

To perform ratio control with a PIDE instruction, set the

AllowCasRat

and

UseRatio

input parameters. Wire the uncontrolled flow into the

SPCascade

input parameter. When in Cascade/Ratio mode, the uncontrolled flow is multiplied by either the

RatioOper

(when in Operator control) or the RatioProg (when in Program control) and the resulting value is used by the PIDE instruction as the setpoint.

Function Block

Structured Text

PIDE_01.PV := ControlledFlow;

PIDE(PIDE_01);

Local:2:O.Ch0Data := PIDE_01.CVEU;

Switching Between Program Control and Operator Control

The PIDE instruction can be controlled by either a user program or an operator interface. You can change the control mode at any time. Program and Operator control use the same ProgOper output. When ProgOper is set, control is Program; when ProgOper is cleared, control is Operator.

The following diagram shows how the PIDE instruction changes between Program control and Operator control.

(1) The instruction remains in Operator control mode when ProgOperReq is set.

Operating Modes

The PIDE instruction supports the following PID modes.

The Cascade/Ratio, Auto, and Manual modes can be controlled by a user program when in Program control or by an operator interface when in Operator control. The Override and Hand modes have a mode request input that can only be controlled by a user program; these inputs operate in both Program and Operator control.

Selecting the Setpoint

Once the instruction determines program or operator control and the PID mode, the instruction can obtain the proper SP value. You can select the cascade/ratio SP or the current SP.

Cascade/Ratio SP

The cascade/ratio SP is based on the UseRatio and ProgOper values.

Current SP

The current SP is based on the Cascade/Ratio mode, the PVTracking value, auto mode, and the ProgOper value.

SP High/Low Limiting

The high-to-low alarming algorithm compares SP to the SPHLimit and SPLLimit alarm limits. SPHLimit cannot be greater than PVEUMax and SPLLimit cannot be less than PVEUMin.

Updating the SPOper and SPProg Values

The PIDE instruction makes SPOper = SP or SPProg = SP to obtain bumpless control switching between Program and Operator control or when switching from Cascade/Ratio mode.

PV High/Low Alarming

The high-high to low-low alarming algorithm compares PV to the PV alarm limits and the PV alarm limits plus or minus the PV alarm deadband

(1) During instruction first scan, the instruction clears all the PV alarm outputs. The instruction also clears the PV alarm outputs and disables the alarming algorithm when PVFaulted is set.

PV Rate-of-Change Alarming

PV rate-of-change (ROC) alarming compares the change in the value of PV over the PVROCPeriod against the PV positive and negative rate-of-change limits. The PVROCPeriod provides a type of deadband for the rate-of-change alarm. For example, if you use a ROC alarm limit of 2°F/second with a period of execution of 100 ms, and an analog input module with a resolution of 1°F, then every time the input value changes, a ROC alarm is generated because the instruction sees a rate of 10°F/second. However, by entering a PVROCPeriod of at least 1 sec, the ROC alarm is only generated if the rate truly exceeds the 2°F/second limit.

The ROC calculation is only performed when the PVROCPeriod has expired. The rate-of-change is calculated as:

ElapsedROCPeriod = ElapsedROCPeriod + ElapsedTimeSinceLastExecution

If ElapsedROCPeriod

³

PVROCPeriod then:

Once PVROC has been calculated, the PV ROC alarms are determined as follows:

(1) During instruction first scan, the instruction clears the PV ROC alarm outputs. The instruction also clears the PVROC alarm outputs and disables the PV ROC alarming algorithm when PVFaulted is set.

Converting the PV and SP Values to Percent

The instruction converts PV and SP to a percent and calculates the error before performing the PID control algorithm. The error is the difference between the PV and SP values. When ControlAction is set, the values of EPercent, E, and PVPIDPercent are negated before being used by the PID algorithm.

(1) PVPIDPercent and Deviation are internal parameters used by the PID control algorithm.

Deviation High/Low Alarming

Deviation is the difference in value between the process variable (PV) and setpoint (SP). Deviation alarming alerts the operator to a discrepancy between the process variable and the setpoint value.

The high-high to low-low alarming algorithm compares the deviation to deviation alarm limits and the deviation alarm limits plus or minus the deadband.

(1) During instruction first scan, the instruction clears the deviation alarm outputs. The instruction also clears the deviation alarm outputs and disables the alarming algorithm when PVFaulted or PVSpanInv is set.

Zero Crossing Deadband Control

You can limit CV such that its value does not change when error remains within the range specified by ZCDeadband (| E |

£

ZCDeadband).

(1)

When ZCOff is cleared, ZCDeadband > 0, error has crossed zero for the first time, (i.e. En >= 0 and En-1 < 0 or when En <= 0 and En-1 > 0), and | En | <= ZCDeadband, the instruction sets ZCDeadbandOn.

(2)

On the transition to Auto or Cascade/Ratio mode, the instruction sets En-1 = En.

The instruction disables the zero crossing algorithm and clears ZCDeadband under these conditions:

Feedforward Control

Compute CV by summing CV from the zero crossing algorithm with ∆FF. The value of ∆FF = FF – FFn-1. When FFSetPrevious is set, FFn-1=FFPrevious. This lets you preset FFn-1 to a specified value before the instruction calculates the value of ∆FF.

Selecting the Control Variable

Once the PID algorithm has been executed, select the CV based on program or operator control and the current PID mode.

CV Windup Limiting

Limit the CV such that its value cannot increase when WindupHIn is set or decrease when WindupLIn is set. These inputs are typically the WindupHOut or WindupLOut outputs from a secondary loop. The WindupHIn and WindupLIn inputs are ignored if CVInitializing, CVFault, or CVEUSpanInv is set.

CV Percent Limiting

The following diagram illustrates how the instruction determines CV percent limiting.

(1) During instruction first scan, the instruction clears the alarm outputs.

CV High/Low Limiting

The instruction always performs alarming based on CVHLimit and CVLLimit. Limit CV by CVHLimit and CVLLimit when in auto or cascade/ratio mode. When in manual mode, limit CV by CVHLimit and CVLLimit when CVManLimiting is set. Otherwise limit CV by 0 and 100 percent.

(1) During instruction first scan, the instruction clears the alarm outputs.

CV Rate-of-Change Limiting

The PIDE instruction limits the rate-of-change of CV when in Auto or Cascade/Ratio mode or when in Manual mode and CVManLimiting is set. A value of zero disables CV rate-of-change limiting.

The CV rate-of-change is calculated as:

where DeltaT is in seconds.

Once CV rate-of-change has been calculated, the CV rate-of-change alarms are determined as follows:

(1) During instruction first scan, the instruction clears the alarm output. The instruction also clears the alarm output and disables the CV rate-of-change algorithm when CVInitializing is set.

(2) When in Auto or Cascade/Ratio mode or when in Manual mode and CVManLimiting is set, the instruction limits the change of CV.

Updating the CVOper and CVProg Values

If not in the Operator Manual mode, the PIDE instruction sets CVOper = CV. This obtains bumpless mode switching from any control to the Operator Manual mode.

Primary Loop Control

Primary loop control is typically used by a primary PID loop to obtain bumpless switching and anti-reset windup when using Cascade/Ratio mode. The primary loop control includes the initialize primary loop output and the anti-reset windup outputs. The InitPrimary output is typically used by the CVInitReq input of a primary PID loop. The windup outputs are typically used by the windup inputs of a primary loop to limit the windup of its CV output.

(1) During instruction first scan, the instruction sets InitPrimary.

(2) When CVInitializing is set or when not in Cascade/Ratio mode the instruction sets InitPrimary.

(3) When CVInitializing is cleared and in Cascade/Ratio mode, the instruction clears InitPrimary.

(4) During instruction first scan, the instruction clears the windup outputs. The instruction also clears the windup outputs and disables the CV windup algorithm when CVInitializing is set or if either CVFaulted or CVEUSpanInv is set.

(5) The instruction sets WindupHOut when SPHAlarm is set, or when ControlAction is cleared and CVHAlarm is set, or when ControlAction is set and CVLAlarm is set.

The SP and CV limits operate independently. A SP high limit does not prevent CV from increasing in value. Likewise, a CV high or low limit does not prevent SP from increasing in value.

(6) The instruction clears WindupHOut when SPHAlarm is cleared, and not (ControlAction is cleared and CVHAlarm is set), and not (ControlAction is set and CVLAlarm is set).

(7) The instruction sets WindupLOut when SPLAlarm is set, or when ControlAction is cleared and CVLAlarm is set, or when ControlAction is set and CVHAlarm is set.

The SP and CV limits operate independently. A SP low limit does not prevent CV from increasing in value. likewise a CV low or high limit does not prevent SP from increasing in value.

(8) The instruction clears WindupLOut when SPLAlarm is cleared and not (ControlAction is cleared and CVLAlarm is set) and not (ControlAction is set and CVHAlarm is set).

Processing Faults

The following table describes how the instruction handles execution faults: