Copyright © 2014 Rockwell Automation, Inc. All Rights Reserved. PUBLIC INFORMATION PID Controller Tuning Advancing the State-of-the-Art with Patent-Pending Modeling Control Station
Jun 21, 2015
Copyright © 2014 Rockwell Automation, Inc. All Rights Reserved.
PUBLIC INFORMATION
PID Controller TuningAdvancing the State-of-the-Art with Patent-Pending Modeling Control Station
Robert Rice, PhDVice President, Engineering
Control Station, Inc.
PID Controller Tuning Advancing the State-of-the-Art with Patent-Pending Modeling
Copyright © 2014 Control Station, Inc. All Rights Reserved
Outline of Discussion
Introduction to Process ControlBrief history of Process Control
Introduction to Process Behavior and the Control Objective
Why understanding the process is fundamental to controlling itThe importance of stating the correct control objective
PID Controller Tuning MethodThe PID Controller
What is a PID Controller Examples of the PID controllers (e.g. PI vs PID)
Theory Vs the Real-WorldQuestions and Answers
Copyright © 2014 Control Station, Inc. All Rights Reserved
History of Feedback / PID Control
300BC – 1200 AD Float Regulators used in Water Clocks (P-Only Control)
Used a float to control the inflow of water through a valve; as the level of water fell the valve opened and replenished the reservoir. This float regulator performed the same function as the ball and cock in a modern flush toilet.
1700 – 1900 : Industrial RevolutionCentrifugal (Flyball) Governors (P-Only Controller)
This device employed two pivoted rotating flyballs which were flung outward by centrifugal force. As the speed of rotation increased, the flyweights swung further out and up, operating a steam flow throttling valve which slowed the engine down. Thus, a constant speed was achieved automatically.
1900 – Current : Mass ManufacturingPneumatic, Electronic, Model Predictive ControllersPID Control
Copyright © 2014 Control Station, Inc. All Rights Reserved
PID Tuning and Optimization
A well controlled process has less variability in the measured process variable (PV), so the process can be operated close to the maximum profit constraint.
Copyright © 2014 Control Station, Inc. All Rights Reserved
Process OptimizationUnderperforming Controllers Can Cripple Plant Profitability
0% 50% 100%
Controllers That are Operated in Manual
Mode
Controllers That are Poorly Tuned or De-Tuned
to Mask Other Issues
Control Systems That are Not Properly Configured to Meet Their Objective
20%
30%
65%
Copyright © 2014 Control Station, Inc. All Rights Reserved
Significant Opportunity:Uncovering the Value of Improved Control
ProductionThroughput
Production Yield
EnergyConsumption
ProductionDefects
2 – 5%
5 – 15% 25 – 50%
5 – 10%
Benefits of regular PID tuning can be found across a production facility:
Copyright © 2014 Control Station, Inc. All Rights Reserved
Steps to Controller Design and Tuning
Identify the Controller and
Specify the Design Level of Operation (DLO)
and Control Objective
Find
Perform a “Bump Test” and Collect Dynamic
Process Data
Step
Fit a Model to the Process Data
Model
Use Tuning Correlations to
Calculate Tunings Based
on Model
Tune
Implement and Test results
Test
Document the Tuning Process
Document
Copyright © 2014 Control Station, Inc. All Rights Reserved
Step 1: Find Controller, Specify Objective
How do you identify PID loops that need to be retuned?
Reactive: Response to the Operators Needs Proactive: Analyze Process Data to determine PID Loops that contribute to increased process variability
Proactive Monitoring Should:Identify Mechanical, Process and Controller Tuning Related IssuesProvide Root-Cause DetectionRecommendation for Corrective ActionDisplay Customizable Reports
Copyright © 2014 Control Station, Inc. All Rights Reserved
Step 1: Find Controller, Specify ObjectiveGood Control is “SIMPLE”
Copyright © 2014 Control Station, Inc. All Rights Reserved
Step 1: Find Controller, Specify Objective
What is/are the primary Control Objective(s)?Maintain Liquid Level In the Reflux DrumMaintain Column StabilityPrevent Environmental Release by avoid Drum Hi Limit
Reflux Drum – Level Control Example
Copyright © 2014 Control Station, Inc. All Rights Reserved
A bump test must generate a response that clearly dominates the random (noisy) PV behavior
Here the PV moves about 4 times the noise band, a good value
Step 2: Step or Bump the ProcessData Should Show “Cause and Effect”
Copyright © 2014 Control Station, Inc. All Rights Reserved
Open loop tests require the controller output to be stepped
Closed loop tests require a sharp controller output change
Step 2: Step or Bump the ProcessGood Bumps Tests
sharp COmovement
sharp COmovement
Copyright © 2014 Control Station, Inc. All Rights Reserved
Step 2: Step or Bump the ProcessBad Bump Tests
AVIODDisturbance Driven Data &
Slow Ramp CO Changes
Copyright © 2014 Control Station, Inc. All Rights Reserved
Step 3: Fit a Process Model
Self-RegulatingIf all inputs & outputs are held constant, the process will seek a steady-stateEx: Heat Exchanger
Non Self-RegulatingProcess will only reach a steady-state at its ‘balancing’ pointEx: Surge Tank
Types of Process Behavior
Copyright © 2014 Control Station, Inc. All Rights Reserved
Self-Regulating
∙
→ Process Gain
→ Time Constant [time]→ Deadtime [time]
Non Self-Regulating
∗ ∙
∗ → Integrator Gain ∙
→ Deadtime [time]
Step 3: Fit a Process ModelSimple First Order Models for Modeling
All models are wrong, some are useful -George Box
Copyright © 2014 Control Station, Inc. All Rights Reserved
63%∆
∆∆
Process Gain How Far
How Far does the PV Move for Change in the Output
Process Time Constant How Fast
How Fast does it take the PV to reach 63% of its total change
Process Deadtime How Much Delay
How much delay is there from when the CO is changed until the PV first moves
Step 3: Fit a Process ModelFirst Order Plus Deadtime (Self-Regulating Model)
Copyright © 2014 Control Station, Inc. All Rights Reserved
Integrating Process Gain How Far and How Fast
How Far and How Fast does the PV Move when the CO is moved from its balancing point
Process Deadtime How Much Delay
How much delay is there from when the CO is changed until the PV first moves
Step 3: Fit a Process ModelFirst Order Plus Deadtime (Non Self-Regulating Model)
Copyright © 2014 Control Station, Inc. All Rights Reserved
Step 3: Fit a Process Model
By Hand or Autotune Approach Sufficient for Simplest of ControllersSoftware Modeling Much More Robust
Handle Open/Close LoopNoisy / Non-Steady State Conditions
Tunings Only As Good as the Model
SIMPLE
Copyright © 2014 Control Station, Inc. All Rights Reserved
Step 4: Tune the PID Loop
1
First compute, , the closed loop time constant (a small provides an aggressive or quick response)Choose your performance using these rules:
aggressive: is the larger of 0.1 or 0.8moderate: is the larger of 1 or 8conservative: is the larger of 10 or 80
PI tuning correlations use this and the FOPDT model values:
and
IMC Tuning Correlation: Dependent PI, Self-Regulating Process
Copyright © 2014 Control Station, Inc. All Rights Reserved
Step 4: Tune the PID Loop
1
First compute, , the closed loop time constant (a small provides an aggressive or quick response)Choose your performance using these rules:
aggressive: is the larger of 0.1 or 0.8moderate: is the larger of 1 or 8conservative: is the larger of 10 or 80
PID tuning correlations use this and the FOPDT model values:
1 0.50.5 0.5 2
IMC Tuning Correlation: Dependent PID, Self-Regulating Process
Copyright © 2014 Control Station, Inc. All Rights Reserved
Step 4: Tune the PID Loop
1
The closed loop time constant, , should be as large as possible, but still fast enough to arrest or recover from a major disturbance.PI tuning correlations use this and the FOPDT Integrating model values:
1∗
22
IMC Tuning Correlation: Dependent PID, Non Self-Regulating Process
Copyright © 2014 Control Station, Inc. All Rights Reserved
Step 4: Tune the PID Loop
Flow Loops 3 to 5 times the Open Loop Time Constant,
Pressure Loops 2 to 4 times the Open Loop Time Constant,
Temperature Loops 1 to 3 times the Open Loop Time Constant,
Closed Loop Time Constant Rules of Thumb
Copyright © 2014 Control Station, Inc. All Rights Reserved
Step 4: Tune the PID Loop
Set point tracking (servo) response as changes
Expected PI Controller Response
Copyright © 2007 by Control Station, Inc. All Rights Reserved.
Conservative Moderate Aggressive
Copyright © 2014 Control Station, Inc. All Rights Reserved
Kc*2
Kc/2
Kc
Step 4: Tune the PID LoopChallenges of PI Control: Self-Regulating Processes
Base Case Performance
2
Copyright © 2007 by Control Station, Inc. All Rights Reserved.
Ti/2 Ti 2Ti
Copyright © 2014 Control Station, Inc. All Rights Reserved
2*Kc
Kc / 2
Kc
Step 4: Tune the PID LoopChallenges of PI Control: Non Self-Regulating Processes
Ti/2 Ti 2Ti
Copyright © 2014 Control Station, Inc. All Rights Reserved
Step 4: Tune the PID Loop
PID shows decreased oscillations compared to PI performancePID has somewhat:
Shorter Rise TimeFaster Settling TimeSmaller Overshoot
PI vs PID Set Point Tracking Response
Copyright © 2014 Control Station, Inc. All Rights Reserved
Step 5: Implement and Test Results
Testing PID Controllers Typically Involve
Adjust Set-Point to ensure adequate tracking
Did the Process Variable Overshoot?Did the Controller Output Move too much?
Introduce a Load Change or Disturbance
Did the Process Variable Recover quick enough?
Updated Tuning Parameters MUST be tested
NOTE: PID Controllers work off of controller error (SP-PV), if there is no error, there is nothing for the PID Controller to do. You MUST introduce controller error, and force the controller to respond before you know if your tuning changes improved the system.
Copyright © 2014 Control Station, Inc. All Rights Reserved
Step 6: Document, Document, Document.
WhoWho is accountable for the changes?
WhatWhich loop has been tuned, what were the As Found and Recommended Tuning Values?
WhenWhen was the Loop Adjusted?
WhyWhy was this particular loop tuned?
Copyright © 2014 Control Station, Inc. All Rights Reserved
Processes Have Time Varying Behavior
The CO to PV behavior described by an ideal FOPDT model is constant, but real processes change every day because:
surfaces foul or corrode mechanical elements like seals or bearings wearfeedstock quality varies and catalyst activity decays environmental conditions like heat and humidity change
So the values of , and that best describe the dynamic behavior of a process today may not be best tomorrow
As a result, controller performance can degrade with time and periodic retuning may be required
Copyright © 2014 Control Station, Inc. All Rights Reserved
Example Process: Heat Exchanger
Process Variable (PV)Set Point (SP)Controller Output (CO)Disturbances (D)
Copyright © 2014 Control Station, Inc. All Rights Reserved
Heat Exchanger Shows Nonlinear Behavior
Processes often exhibit changing (or nonlinear) behavior as operating level changesAs a result, “best” tuning can change if the set point moves the PV across a range of operation
Copyright © 2014 Control Station, Inc. All Rights Reserved
Controller’s Robust Stability
What does it mean for a controller to be Robustly Stable?Controller Robustness measures the Ability to Tolerate Variations in Process Behavior (e.g., Nonlinearity)
Visual Robust Stability PlotPlots Plant-Model Mismatch in Gain vs. Plant-Model Mismatch in Dead TimeStable and Unstable Regions shown on Plot
Copyright © 2014 Control Station, Inc. All Rights Reserved
Options For Tuning
Manual Tuning• Time Consuming, and may not yield consistent results.• Results vary on experience
Push Button Auto-Tune• For Simple / Fast Loops (e.g. Flow)• Requires “Steady-State” Starting Condition• Generally not recommended for Level Loops or Slow Temperature
(Batch Temperature) Controllers
Dedicated PID Tuning / Modeling Package• Handle All Types of Processes
• From Fast Flows, to Slow Batch Temperature or Furnace Temperature Control
• Customize Controller Response to Match Objective
Copyright © 2014 Control Station, Inc. All Rights Reserved
Full Non-Steady State ModelingOpen and Closed Loop ModelingSupports All Rockwell Automation PLCs (SLC to Logix)
Monitor 100s to 1000s of PIDsIdentify Interactions, Valve and Tuning IssuesCustomizable Alerts and Reports
Rockwell Automation Encompass ProductsFor Controller Tuning and Control Loop Performance Monitoring
Copyright © 2014 Control Station, Inc. All Rights Reserved
Summary
First Order Models provide Important InformationHow Far?; How Fast?; With How Much Delay?Fit by Hand or Use Software
Systematic Approach to Tune PID Controllers Internal Model Control (IMC) Tuning
Uses the FOPDT Model in the Tuning CorrelationSpecifying the Single Adjustable Tuning Parameter,
Decrease for a Faster, More Aggressive ResponseIncrease to Increase Robustness
Understanding Robust StabilityProcesses Change over time and with Operating LevelController Performance can degrade over timeSelect Tunings which balance performance with robust stability
Copyright © 2014 Control Station, Inc. All Rights Reserved
Questions?
Thank you for attending!
Contact Information:
Bob Rice, PhD
Vice President, Engineering+1-860-872-2920, ext. 1601+1-860-420-7158 (m)[email protected]
Copyright © 2014 Rockwell Automation, Inc. All Rights Reserved.
We care what you think!
On the mobile app:1. Locate session using
Schedule or Agenda Builder2. Click on the thumbs up icon on
the lower right corner of the session detail
3. Complete survey4. Click the Submit Form button
38
Please take a couple minutes to complete a quick session survey to tell us how we’re doing.
2
3
4
1
Thank you!!
Copyright © 2014 Rockwell Automation, Inc. All Rights Reserved.
PUBLIC INFORMATION
Questions?