Professional Documents
Culture Documents
Presenter
Greg is a retired Senior Fellow from Solutia/Monsanto and an ISA Fellow. Greg was an adjunct professor in the Washington University Saint Louis Chemical Engineering Department 2001-2004. Presently, Greg contracts as a consultant in DeltaV R&D via CDI Process & Industrial and is a part time employee of Experitec and MYNAH. Greg received the ISA Kermit Fischer Environmental Award for pH control in 1991, the Control Magazine Engineer of the Year Award for the Process Industry in 1994, was inducted into the Control Process Automation Hall of Fame in 2001, was honored by InTech Magazine in 2003 as one of the most influential innovators in automation, and received the ISA Life Achievement Award in 2010. Greg is the author of 20 books on process control, his most recent being Advanced Temperature Measurement and Control. Greg has been the monthly Control Talk columnist for Control magazine since 2002 and has started a Control Talk Blog. Gregs expertise is available on the Control Global and Emerson modeling and control web sites: http://community.controlglobal.com/controltalkblog http://modelingandcontrol.com/author/Greg-McMillan/
Resources
2005
2010
2011
(1) - Delay
Without deadtime I would be out of a job Fundamentals
Goals
A more descriptive name would be total loop deadtime. The loop deadtime is the amount of time for the start of a change to completely circle the control loop and end up at the point of origin. For example, an unmeasured disturbance cannot be corrected until the change is seen and the correction arrives in the process at the same point as the disturbance. Process deadtime offers a continuous train of values whereas digital devices and analyzers offer non continuous data values at discrete intervals, these delays add a phase shift and increase the ultimate period (decrease natural frequency) like process deadtime.
Sources
Minimize delay (the loop cannot do anything until it sees and enacts change)
Pure delay from process deadtimes and discontinuous updates
Piping, duct, plug flow reactor, conveyor, extruder, spin-line, and sheet transportation delays (process deadtimes set by mechanical design - remaining delays set by automation system design) Digital device scan, update, reporting, and execution times (0.5T) Analyzer sample processing and analysis cycle time (1.5T) Sensitivity-resolution limits Backlash-deadband Mixing, column trays, dip tube size and location, heat transfer surfaces, and volumes in series (process lags set by mechanical design - remaining lags set by automation system design) Thermowells Electrodes Transmitter damping Signal filters
(2)- Speed
Speed kills - (high speed processes and disturbances and low speed control systems can kill performance) Fundamentals
The rate of change in 4 deadtime intervals is most important. By the end of 4 deadtimes, the control loop should have completed most of its correction. Thus, the short cut tuning method (near-integrator) is consistent with performance objectives. Make control systems faster and make processes and disturbances slower Control system
PID tuning settings (gain, reset, and rate) Slewing rate of control valves and velocity limits of variable speed drives Steps - Batch operations, on-off control, manual actions, SIS, startups, and shutdowns Oscillations - limit cycles, interactions, and excessively fast PID tuning Ramps - reset action in PID Degree of mixing in volumes due to agitation, boiling, mass transfer, diffusion, and migration
Goals
Sources
Disturbances
Process
(3) - Gain
All is lost if nothing is gained Fundamentals
Gain is the change in output for a change in input to any part of the control system. Thus there is a gain for the PID, valve, disturbance, process, and measurement. Knowing the disturbance gain (e.g. change in manipulated flow per change in disturbance) is important for sizing valves and feedforward control.
Goals
Sources
Maximize control system gains (maximize control system reaction to change) and minimize process and disturbance gains (minimize process reaction to change).
PID controller gain Inferential measurements (e.g. temperature change for composition change in distillation column) Slope of control valve or variable speed drive installed characteristic (inherent characteristic & system loss curve) Measurement calibration (100% / span). Important where accuracy is % of span Process design Attenuation by upstream volumes (can be estimated) Attenuation by upstream PID loops (transfer of PV variability to controller output)
For a discussion of unifying concepts check out Deminar #9 Process Control Improvement Primer Sept 8, 2010 Recording
Go to Deminars under Archives heading on top of web page
http://modelingandcontrol.com/author/Greg-McMillan/
(4) - Resonance
Dont make things worse than they already are Fundamentals
Oscillation period close to ultimate period can be amplified by feedback control
Goals
Make oscillation period slower or control loop faster
Sources
Control loops in series with similar loop deadtimes (e.g. multiple stage pH control) Control loops in series with similar tuning and valve stiction and backlash Day to night ambient changes to slow loops (e.g. column temperature control)
(4) - Resonance
For all of you frequency response and Bode Plot Fans
resonance (amplification) of disturbances amplitude ratio is proportional to ratio of break frequency lag to disturbance period
1 1
no attenuation of disturbances
Faster Tuning
(5) Attenuation
If you had a blend tank big enough you would not need control Fundamentals
Attenuation increases as the volume of the blend tank increases and the ultimate period of the control loop decreases.
Goals
Maximize attenuation by increasing volume and mixing and making loops faster
Sources
Mixed volume size and degree of mixing Control loop speed
(5) Attenuation
The attenuation of oscillations can be estimated from the expression of the Bode plot equation for the attenuation of oscillations slower than the break frequency where (f) is the filter time constant, electrode or thermowell lag, or a mixed volume residence time
Af = Ao *
to 2 f
Equation is also useful for estimating original process oscillation amplitude from filtered oscillation amplitude to better know actual process variability (measurement lags and filters provide a attenuated view of real world)
Goals
Improve sensitivity and resolution
Sources
In measurements, minimum change detected and communicated (e.g. sensor threshold and wireless update trigger level) and quantized change (A/D & D/A) Minimum change that can be manipulated (e.g. valve stick-slip sensitivity and speed resolution)
x o o x o x
o o x x
o x o o x
Actual Transmitter Response
0.02% 0.01%
o x o x
1 2
o 0.00% x 0
4 6 5 Digital Updates
10
x o o x o x o x o x
x o o x o o x
1
o x
x
2 3 4 6 5 Digital Updates 7
o 0.00% x 0
10
(7) Hysteresis-Backlash
No problem if you dont ever change direction Fundamentals
Hysteresis is the bow in a response curve between full scale traverses in both directions. Normally much smaller and less disruptive than backlash Backlash (deadband) is minimum change measured or manipulated once the direction is changed - once past backlash-deadband limit you get full change Both Hysteresis and backlash will cause a limit cycle if there are 2 or more integrators in the process or control system.
Goals
Minimize backlash and deadband
Sources
Pneumatic instrument flappers, links, and levers (hopefully these are long gone) Rotary valve and damper links, connections, and shaft windup Variable speed drive setup parameter to eliminate hunting and chasing noise
(7) Hysteresis-Backlash
Hysteresis
100%
x 0 x x x x x x
True Process Variable x
x
Actual Transmitter Response
x
0
4 6 5 Digital Updates
10
(7) Hysteresis-Backlash
Backlash (Deadband)
Deadband is 5% - 50% without a positioner ! Deadband
Stroke (%)
Signal (%)
(8) Repeatability-Noise
The best thing you can do is not react to noise Fundamentals
Noise is extraneous fluctuations in measured or manipulated variables Repeatability is difference in readings for same true value in same direction Often repeatability is confused with noise
Goals
Minimize size and frequency of noise and do not transfer noise to process
Sources
Noise
Bubbles Concentration and temperature non-uniformity from imperfect mixing Electromagnetic interference (EMI) Ground loops Interferences (e.g. sodium ion on pH electrode) Velocity profile non-uniformity Velocity impact on pressure sensors
Repeatability
Sensitivity and resolution
(8) Repeatability-Noise
Official definition of repeatability obtained from calibration tests
100% 90% 80% Process Variable and Measurements 70% 60% 50% 40% 30% 20% 10% 0% 0 0 0 0
True Process Variable
0 0 0 0 0 0 0
Actual Transmitter Response
Repeatability
4 6 5 Digital Updates
10
(8) Repeatability-Noise
Practical definition of repeatability as seen on trend charts
100% 90% 80% 70% Process Variable and Measurements 60% 50% 0 40% x 30% 20% 10% 0% 0
True Process Variable
0 x
Actual Transmitter Response
0 x
x x x Repeatability
4 6 5 Digital Updates
10
(8) Repeatability-Noise
Noise as seen on trend charts
100% 90% 80% 70% Process Variable and Measurements 60% 50% 0 x 40% 30% 20% 10% 0%
True Process x Variable
x 0
Actual Transmitter Response
x 0
x 0 Noise
0 x
x 0
0 x
x 0
0 x
4 6 5 Digital Updates
10
(9) Offset-Drift
There is always an offset and drift, it is matter of size and consequence Fundamentals
The deviation of the peak in the distribution of actual values from true value Drift shows up as a slowly changing offset Minimize offset and nonlinearity by smart transmitters and sensor matching and smart tuned digital positioners with accurate internal closure member feedback Manufacturing tolerance, degradation, de-calibration, and installation effects (process and ambient conditions and installation methods and location)
Goals
Sources
(9) Offset-Drift
Offset (Bias)
100% 90% 80% 70% Process Variable and Measurements 60% 50% 40% 30% 20% 10% 0% 0 0 0 0
True Process Variable
0 0 0 0 0 0 0 0
x
Bias
x
Actual Transmitter Response
x
2 3 4 6 5 Digital Updates 7 8 9 10
x
1
(9) Offset-Drift
Drift (Shifting Bias)
100% 90% 80% Process Variable and Measurements 70% 60% x 50% 0 40% 30% 20% 10% 0% 0 x
True Process Variable
0 x
5 Months
10
(10) Nonlinearity
Not a problem if the process is constant, but then again if the process is constant, you do not need a control system Fundamentals
While normally associated with a process gain that is not constant, in a broader concept, a nonlinear system occurs if a gain, time constant, or delay changes anywhere in the loop. All process control systems are nonlinear to some degree. Goals Minimize nonlinearity by process and equipment design (e.g. reagents and heat transfer coefficients), smart transmitters and sensor matching, valve selection, signal characterization, and adaptive control
Sources
Control valve and variable speed drive installed characteristics (flat at high flows) Process transportation delays (inversely proportional to flow) Digital and analyzer delays (loop delay depends upon when change arrives in discontinuous data value update interval) Inferred measurement (conductivity or temperature vs. composition plot is a curve) Logarithmic relationship (glass pH electrode and zirconium oxide oxygen probe) Process time constants (proportional to volume and density)
(10) Nonlinearity
100% 90% 80% 70% Process Variable and Measurements 60% 50% 40% 30% 20% 10% 0% x 0 0 0
True Process Variable
x 0
0 0
x x
0 0 0 0
x
Nonlinearity
x
Actual Transmitter Response
x
0 x
x x
4 6 5 Digital Updates
10
Bias
Frequency of Measuremen ts
Frequency of Measurements
Bias
Measured Values
True Value
Frequency of Measurements
Measured Values
True Value
Frequency of Measurements
Bias
Measured Values
True Value
Measurements
Measurements
Measurements
Measurements
Measurements
Is the permanent pressure loss for the meter size acceptable? Do O-rings and gaskets meet worst case corrosive and temperature operating conditions? Are gaskets not projecting into flow stream? Is meter centerline concentric with piping centerline? Do the upstream and downstream straight run lengths for vortex meters meet the ASME guideline for 0.8 beta ratio flow tubes (e.g. 20 pipe diameters upstream for long bends)? Do the upstream and downstream straight run lengths for turbine meters meet the ASME guideline for 0.6 beta ratio flow tubes (e.g. 10 pipe diameters upstream for long bends)? Do the upstream and downstream straight run lengths for magnetic flow meters meet the ASME guideline for 0.4 beta ratio flow tubes (e.g. 5 pipe diameters upstream for long bends)?
31
Measurements
Measurements
m m
Measurements
34
Measurements
35
Measurements
36
Measurements
Measurements
38
Measurements
(1) RTD sensor sheath insulation errors can be significant for temperatures above 400oC Operations may change a setpoint to account for the offset from drift. However the drift is unpredictable and the replacement of the sensor causes a error in operating point.
39
Measurements
Bare Sensing Element Type Thermocouple 1/8 inch sheathed and grounded Thermocouple 1/4 inch sheathed and grounded Thermocouple 1/4 inch sheathed and insulated Single Element RTD 1/8 inch Single Element RTD 1/4 inch Dual Element RTD 1/4 inch
40
Measurements
41
Measurements
42
Measurements
43
Measurements
44
Measurements
45
Measurements
Thermowell tip should be in center of pipe to reduce cross sectional temperature profile error from wall and mixing with enough length to reduce conduction error
46
Measurements
flush
An optimum location provides good velocity, sufficient mixing, and minimal time delay Transportation delay is volume between outlet of equipment and sensor divided by flow
47
Measurements
Desuperheater Installation
For Desuperheaters, the distance from the outlet to the thermowell depends upon the performance of the Desuperheater, process conditions, and the steam velocity. To give a feel for the situation there are some simple rules of thumb for the straight piping length (SPL) to the first elbow and the total sensor length (TSL). Actual SPL and TSL values depend on quantity of water required with respect to the steam flow rate, temperature differential between water and steam, water temperature, pipe diameter, steam velocity, model, type, etc. and are computed by software programs. SPL (feet) = Inlet steam velocity (ft/s) * 0.1 (seconds residence time) TSL (feet) = Inlet steam velocity (ft/s) * 0.2 (seconds residence time) Typical values for the inlet steam velocity, upstream of the Desuperheater range from 25350 ft/s (7.6 to 107 m/sec). Below 25 ft/s there is not enough motive force to keep water suspended in steam. Velocities higher than 350 ft/s cause the Desuperheater to vibrate and damage the unit to the point where it breaks apart
48
Measurements
Measurements
50
Measurements
51
Measurements
Food is burning in the operators kitchen The only loop mode configured is manual An operator puts his fist through the screen You trip over a pile of used pH electrodes The technicians ask: what is a positioner? The technicians stick electrodes up your nose The environmental engineer is wearing a mask The plant manager leaves the country Lawyers pull the plugs on the consoles The president is on the phone holding for you
52
Measurements
aH = hydrogen ion activity (gm-moles per liter) cH = hydrogen ion concentration (gm-moles per liter) cOH = hydroxyl ion concentration (gm-moles per liter) = activity coefficient (1 for dilute solutions) pH = negative base 10 power of hydrogen ion activity pKw = negative base 10 power of the water dissociation constant (14.0 at 25oC)
53
Measurements
pH 8
pH 9
pH 10
Measured pH
pH 7.5 pH 7 pH 6.5 pH 6
pH 5
54
Measurements
Influent pH
pH Set Point
Sensor Drift
4
Reagent to Feed Flow Ratio
The error in a pH feedforward calculation increases for a given sensor error as the slope of the curve decreases. This result Combined with an increased likelihood of Errors at low and high pH means feedforward Could do more harm than good when going from the curves extremes to the neutral region.
Flow feedforward (ratio control of reagent to influent flow) works well for vessel pH control if there are reliable flow measurements with sufficient rangeability
Feedforward Reagent Error
Feedforward control always requires pH feedback correction unless the set point is on the flat part of the curve, use Coriolis mass flow meters and have constant influent and reagent concentrations
55
Measurements
Measurement becomes slow from a loss of active sites or a thin coating of outer gel
High acid or base concentrations can affect glass gel layer and reference junction potential Increase in noise or decrease in span or efficiency is indicative of glass electrode problem Shift or drift in pH measurement is normally associated with reference electrode problem
56
Measurements
Most calibration adjustments chase the short term errors shown below that arise from concentration gradients from imperfect mixing, ion migration into reference junction, temperature shifts, different glass surface conditions, and fluid streaming potentials. With just two electrodes, there are more questions than answers.
A B pH
B A
A B
time
57
Measurements
For more Information on Middle Signal Selection see Feb 5, 2010 post http://www.modelingandcontrol.com/2010/02/exceptional_opportunities_in_p_11.html
58
Measurements
>100% increase in life from new glass designs for high temperatures
25 C
50 C 75 C Process Temperature
100 C
High acid or base concentrations (operation at the extremes of the titration curve) decrease life for a given temperature. A deterioration in measurement accuracy (e.g. electrode efficiency) and response time often accompanies a reduction in life.
59
Measurements
Glass electrodes get slow as they age. High temperatures cause premature aging
60
Measurements
New
Old #1
Old #2
61
Measurements
AE
pressure drop for each branch must be equal to keep the velocities equal
drain
AE
AE
20 to 80 degrees
20 pipe diameters static mixer or pump
The bubble inside the glass bulb can be lodged in tip of a probe that is horizontal or pointed up or caught at the internal electrode of a probe that is vertically down
AE
10 OD
AE
10 OD
AE
flush
Series arrangement preferred to minimize differences in solids, velocity, concentration, and temperature at each electrode!
62
Measurements
AE
AE
AE
AE
Series arrangement preferred to minimize differences in solids, velocity, concentration, and temperature at each electrode!
63
AE
10 OD
AE
10 OD
hole or slot
Measurements
Select best glass and reference electrolyte for process Use smart digital transmitters with built-in diagnostics Use middle signal selection of three pH measurements
Inherent auto protection against a failure, drift, coating, loss in efficiency, and noise (February 5, 2010 entry on http://modelingandcontrol.com/author/Greg-McMillan//)
Allocate time for equilibration of the reference electrode Use in place standardization based on a sample with the same temperature and composition as the process. If this is not practical, the middle value of three measurements can be used as a reference. The fraction and frequency of the correction should be chosen to avoid chasing previous calibrations Insure a constant process fluid velocity at the highest practical value to help keep the electrodes clean and responsive
64
Measurements
Noisy measurement pH decreases with temperature pH increases with temperature pH increases with salt or solvent concentration Constant 7 pH measurement Decrease in measurement range (decrease in electrode efficiency) Daily drift in measurement (change in electrode offset) Hourly drift in measurement Off-scale pH measurement
Measurements
Incredibly tight pH control via 0.001 pH wireless resolution setting still reduced the number of communications by 60% Temperature compensated wireless pH controlling at 6.9 pH set point
66
Measurements
67
Measurements
68
Measurements
pH Measurement Checklist
The following checklist is not intended to cover all the specification requirements but some of the major application details to be addressed for automation component. The following list assumes the materials of construction have been properly specified and that the sensor will work safely and reliably with acceptable accuracy for the maximum possible temperature and pressure. For a detailed understanding see Chapter 6 in the ISA book Essentials of Modern Measurements and Final Elements in the Process Industries http://www.isa.org/Template.cfm?Section=Books3&template=/Ecommerce/ProductDisplay.cfm &ProductID=10764 Do O-rings and gaskets meet worst case corrosive and temperature operating conditions? Is the best glass used for the worst case temperature, pH, and chemicals that can attack glass (e.g. general purpose, high pH, high temperature, sterilizable, and HF resistant)? For pH < 1 or > 12 would conductivity or density give a better concentration measurement? Is the best reference design and fill used for the accuracy and speed requirement and worst case temperature, low water or pure water solutions that have low conductivity, salts and chemicals concentrations that change junction potential, plug junction, and poison reference Flowing junction offers the most constant reference potential, has the fastest junction equilibration, and eliminates plugging-poisoning, but requires pressurized reservoir Aperture junctions tiny hole has lowest junction potential but is susceptible to plugging Double and triple junction references can slow down internal contamination rate Replaceable junction can fix electrode before plugging -poisoning is problem Large surface solid reference can essentially eliminate plugging, contamination , and poisoning but the junction potential may be large and slow to equilibrate
69
Measurements
pH Measurement Checklist
Is the chemical attack, premature aging from high temperature, or dehydration (non-aqueous solvents or low water concentrations) so severe that automated retractable insertion needed? Is the solution conductivity so low (e.g. condensate, boiler feedwater, deionized water) a special assembly is needed to provide low sample flow, diffuser, and electrolyte reservoir ? Can a VP connector be used to quickly locally disconnect electrode cable eliminating the need to disconnect transmitter and retract cable through conduit or flex to prevent twisting of cable? Is a smart electrode with stored calibration record available for selected electrode? Is a smart transmitter to detect glass and reference problems available? Is solution pH temperature compensation needed besides Nernst temperature compensation? Is a solution ground needed for impedance diagnostics and ground potential elimination? Can a wireless transmitter be used to get latest features and enable portability of measurement to test the best electrode and location (least deadtime and least noise-bubbles)? Is the electrode installed with tip pointing down at a 30-60o angle to prevent bubble inside tip? Are electrodes always wetted even for batching and during shutdown of continuous operations? Is middle signal selection needed to eliminate response to single failure and noise? Is stream velocity and electrode protective shroud design the best for process conditions? Velocity 5-10 fps and exposure of glass to flow helps prevent coatings Velocity 0.1-1 fps and shroud reducing flow impingement helps decrease abrasion Does the electrode tip extend into the center line of pipe and past baffles in vessel?
70
Measurements
pH Measurement Checklist
Is electrode location free from flashing (e.g. not on pump suction or valve discharge)? Are electrodes sufficiently downstream from pump or static mixer to reduce concentration and pressure fluctuations but not so far as to increase deadtime by more than 3 sec? Are insertion electrodes in series used to ensure same velocity and composition? Are electrode and transmitter connections always dry? Do electrical connections and enclosure meet electrical area classification and codes in plant?
71
Stroke (%)
Digital positioner will force valve shut at 0% signal Stick-Slip is worse near closed position 0 Signal (%)
dead band
The dead band and stick-slip is greatest near the closed position Plugging and laminar flow can occur for low Cv requirements and throttling near the seat Consider going to reagent dilution. If this is not possible checkout out a laminar flow valve for an extremely low Cv and pulse width modulation for low lifts
8 Set point 6
Control Band
Influent pH
B A
Where: A = distance to center of reagent error band on abscissa from influent pH B = width of allowable reagent error band on abscissa for control band Er = allowable reagent error (%) Frmax = maximum reagent valve capacity (kg per minute) Fimax = maximum influent flow (kg per minute) Ss = allowable stick-slip (resolution limit) (%)
73
Less backlash but wear of piston O-ring seal from piston pitch is concern
Stick-slip from rack and gear teeth - particularly bad for worn teeth
Terminal Box
Control Signal
SV
ZZZZZZZ
Piston
SV SV
W Check Valve
Air Supply
Volume Tank
F is h e r V 1 5 0 H D /1 0 5 2 ( 3 3 )/3 6 1 0 J
(% )
55 50 45 40 0 .5 % S te p s 0 .5 % S te p s 1 % S te p s 1 % S te p s 2 % S te p s 2 % S te p s 5 % S te p s 5 % S te p s 1 0 % S te p s 1 0 % S te p s
65 60
N e le s R 2 1 /Q P 3 C /N P 7 2 3
(% )
55 50 45 40 35
In p u t S ig n a l A c tu a to r P o s itio n F lo w R a te ( F ilte r e d ) 0 50 100 150 200 250 300 350 400 450
T im e ( s e c o n d s )
Maintenance test of 25% or 50% steps will not detect dead band - all valves look good for 10% or larger steps
10
30 25 20 15 10 5 0 0 1 2 3 4 5
Time (sec)
10
100 90 80 70 60 50 40 30 20 10 0 0 10 20 30 40 50 60 70 80 90 100 Inherent Characteristic Installed Characteristic 1 Installed Characteristic 2 Installed Characteristic 3 Installed Characteristic 4
Valve pressure drop ratio (PR) for installed characteristic: Characteristic 1: PR = 0.5 Characteristic 2: PR = 0.25 Characteristic 3: PR = 0.125 Characteristic 4: PR = 0.0625
Valve pressure drop ratio (PR) for installed characteristic: Characteristic 1: PR = 0.5 Characteristic 2: PR = 0.25 Characteristic 3: PR = 0.125 Characteristic 4: PR = 0.0625
Level (%)
Real Rangeability
Minimum fractional flow coefficient for a linear trim and stick-slip:
C x min =
Sv X v max
[ Sv X v max 1]
Minimum fractional flow coefficient for an equal percentage trim and stick-slip:
C x min = R
Q x min =
C x min PR + (1 PR ) C x min
2
Rv =
1 Q x min
Cxmin = minimum flow coefficient expressed as a fraction of maximum (dimensionless) Pr = valve pressure drop ratio (dimensionless) Qxmin = minimum flow expressed as a fraction of the maximum (dimensionless) Rv = rangeability of control valve (dimensionless) R = range of the equal percentage characteristic (e.g. 50) Xvmin = maximum valve stroke (%) Sv = stick-slip near closed position (%)
Low stem packing friction Low sealing and seating friction of the closure components Booster(s) on positioner output(s) for large valves on fast loops (e.g., compressor anti-surge control) Valve sizing for a throttle range that provides good linearity [4]: o 5% to 75% (sliding stem globe), o 10o to 60o (Vee-ball) o 25o to 45o (conventional butterfly) o 5o to 65o (contoured and toothed butterfly) Online diagnostics and step response tests for small changes in signal Dynamic reset limiting using digital positioner feedback [2]
Volume Booster with Integral Bypass (Furnace Pressure and Surge Control)
Signal from Positioner Adjustable Bypass Needle Valve
Bypass
Increase air line size Port A Supply Port B Control Signal
1:1
Increase connection size
Volume Booster
ZZZZZZZ
Air Supply
93
PID Controllers
PID Controllers
seconds/repeat
SP
Time (seconds)
Contribution of Each PID Mode for a Step Change in the Set Point Structure of PID on error (=1 and =1)
PID Controllers
set point
48
52
time
PID Controllers
%CO
Controller is in Manual
%PV
SP
Time (seconds)
PID Controllers
%CO
SP (%) 0.63SP
%PV
Time (seconds)
PID Controllers
SP
SCLR
SUB
% %
PID
% %CO
SCLR
OUT (e.u.)
AO
Control Valve
MV (e.u.)
Process Equipment
PID
AI
PV - Primary Variable SV - Second Variable* TV - Third Variable* FV - Fourth Variable* * - additional HART variables
DCS
The scaler block (SCLR) that convert between engineering units of application and % of scale used in PID algorithm is embedded hidden part of the Proportional-Integral-Derivative block (PID)
To compute controller tuning settings, the process variable and controller output must be converted to % of scale and time units of deadtimes and time constants must be same as time units of reset time and rate time settings!
PID Controllers
SP Feedforward
PID Controllers
If the process changes during the setpoint response (e.g. reaction or evaporation), the resting value can be captured from the last batch or startup If the process changes are negligible during the setpoint response, the resting value can be estimated as:
the PID output just before the setpoint change for an integrating (e.g. batch) process the PID output just before the setpoint change plus the setpoint change divided by the process gain for a self-regulating (e.g. continuous) process
For self-regulating processes such as flow with the loop deadtime (o) approaching or less than the largest process time constant (p ), the logic is revised to step the PID output immediately to the resting value. The PID output is held at the resting value for the T98 process response time (T98 = o + 4 p ).
PID Controllers
Structure 3 Rise Time = 8.5 min Settling Time = 8.5 min Overshoot = 0%
Structure 1 Rise Time = 1.6 min Settling Time = 7.5 min Overshoot = 1.7%
Structure 1 + Bang-Bang Structure 1 + SP FF Rise Time = 0.5 min Rise Time = 1.2 min Settling Time = 6.5 min Settling Time = 0.5 min Overshoot = 0.2% Overshoot = 1.3%
PID Controllers
Deadtime Compensation
The simple addition of a delay block with the deadtime set equal to the total loop deadtime to the external reset signal for the positive feedback implementation of integral action described in Deminar #3 for the dynamic reset limit option http://www.screencast.com/users/JimCahill/folders/Public/media/f093eca1-958f4d9c-96b7-9229e4a6b5ba . The controller reset time can be significantly reduced and the controller gain increased if the delay block deadtime is equal or slightly less than the process deadtime as studied in Advanced Application Note 3 http://www.modelingandcontrol.com/repository/AdvancedApplicationNote003.pdf
PID Controllers
PID Controllers
PID Controllers
PID Controllers
Profile Control
PID Controllers
Primary PV
PID Controllers
PID
PV
PID
PV
Out
AO
PID*
I/P
Relay
Control Valve
Flow Meter
Process
AI
AI
Process Sensor
PID Controllers
Feedforward Applications
Feedforward is the most common advanced control technique used - often the feedforward signal is a flow or speed for ratio control that is corrected by a feedback process controller (Flow is the predominant process input that is manipulated
to set production rate and to control process outputs (e.g. temperature and composition))
Blend composition control - additive/feed (flow/flow) ratio Column temperature control - distillate/feed, reflux/feed, stm/feed, and bttms/feed (flow/flow) ratio Combustion temperature control - air/fuel (flow/flow) ratio Drum level control - feedwater/steam (flow/flow) ratio Extruder quality control - extruder/mixer (power/power) ratio Heat exchanger temperature control - coolant/feed (flow/flow) ratio Neutralizer pH control - reagent/feed (flow/flow) ratio Reactor reaction rate control - catalyst/reactant (speed/flow) ratio Reactor composition control - reactant/reactant (flow/flow) ratio Sheet, web, and film line machine direction (MD) gage control - roller/pump (speed/speed) ratio Slaker conductivity control - lime/liquor (speed/flow) ratio Spin line fiber diameter gage control - winder/pump (speed/speed) ratio
Feedforward is most effective if the loop deadtime is large, disturbance speed is fast and size is large, feedforward gain is well known, feedforward measurement and dynamic compensation are accurate Setpoint feedforward is most effective if the loop deadtime exceeds the process time constant and the process gain is well known
For more discussion of Feedforward see May 2008 Control Talk http://www.controlglobal.com/articles/2008/171.html
PID Controllers
Feedforward Implementation - 1
Feedforward gain can be computed from a material or energy balance ODE * & explored for different setpoints and conditions from a plot of the controlled variable (e.g. composition, conductivity, pH, temperature, or gage) vs. ratio of manipulated variable to independent variable (e.g. feed) but is most often simply based on operating experience
* http://www.modelingandcontrol.com/repository/AdvancedApplicationNote004.pdf Plots are based on an assumed composition, pressure, temperature, and/or quality For concentration and pH control, the flow/flow ratio is valid if the changes in the composition of both the manipulated and feed flow are negligible. For column and reactor temperature control, the flow/flow ratio is valid if the changes in the composition and temperature of both the manipulated and feed flow are negligible. For reactor reaction rate control, the speed/flow is valid if changes in catalyst quality and void fraction and reactant composition are negligible. For heat exchanger control, the flow/flow ratio is valid if changes in temperatures of coolant and feed flow are negligible. For reactor temperature control, the flow/flow ratio is valid if changes in temperatures of coolant and feed flow are negligible. For slaker conductivity (effective alkali) control, the speed/flow ratio is valid if changes in lime quality and void fraction and liquor composition are negligible. For spin or sheet line gage control, the speed/speed ratio is valid only if changes in the pump pressure and the polymer melt quality are negligible.
Dynamic compensation is used to insure the feedforward signal arrives at same point at same time in process as upset
Compensation of a delay in the feedforward path > delay in upset path is not possible
PID Controllers
Feedforward Implementation - 2
Feedback correction is essential in industrial processes
While technically, the correction should be a multiplier for a change in slope and a bias for a change in the intercept in a plot of the manipulated variable versus independent variable (independent from this loop but possibly set by another PID or MPC), a multiplier creates scaling problems for the user, consequently the correction of most feedforward signal is done via a bias. The bias correction must have sufficient positive and negative range for worst case. Model predictive control (MPC) and PID loops get into a severe nonlinearity by creating a controlled variable that is the ratio. It is important that the independent variable be multiplied by the ratio and the result be corrected by a feedback loop with the process variable (composition, conductivity, gage, temperature, or pH) as the controlled variable.
Feedforward gain is a ratio for most load upsets. Feedforward gain is the inverse of the process gain for setpoint feedforward.
Process gain is the open loop gain seen by the PID (product of manipulated variable, process variable, and measurement variable gain) that is dimensionless.
Feedforward action must be in the same direction as feedback action for upset. Feedforward action is the opposite of the control action for setpoint feedforward. Feedforward delay and lag adjusted to match any additional delay and lag, respectively in path of upset so feedforward correction does not arrive too soon. Feedforward lead is adjusted to compensate for any additional lag in the path of the manipulated variable so the feedforward correction does not arrive too late. The actual and desired feedforward ratio should be displayed along with the bias correction by the process controller. This is often best done by the use of a ratio block and a bias/gain block instead of the internal PID feedforward calculation.
PID Controllers
PID Controllers
Open loop backup used for prevention of compressor surge and RCRA pH violation
PID Controllers
PID Controllers
Top Ten Reasons Why an Automation Engineer Makes a Great Spouse or at Least a Wedding Gift
(10) Reliable from day one (9) Always on the job (8) Low maintenance (minimal grooming, clothing, and entertainment costs (7) Many programmable features (6) Stable (5) Short settling time (4) No frills or extraneous features (3) Relies on feedback (2) Good response to commands and amenable to real time optimization (1) Readily tuned
121
PID Controllers
+
Kc
PID integral mode is restructured to provide integral action to match the process response in the elapsed time (reset time set equal to process time constant) PID derivative mode is modified to compute a rate of change over the elapsed time from the last new measurement value PID reset and rate action are only computed when there is a new value If transmitter damping is set to make noise amplitude less than communication trigger level, valve packing and battery life is dramatically improved Enhancement compensates for measurement sample time suppressing oscillations and enabling a smooth recovery from a loss in communications further extending packing -battery life
Elapsed Time
TD
+
Kc
Elapsed Time
PID Controllers
KL
Load Upset
Delay Lag Gain Secondary Delay Fv Secondary Lag
Gain
Kv
p2
p2
p1
p1
Kp
PV
Valve
Hopefully p1 is the largest lag in the loop For integrating processes: Ki = Kv (Kp / p1) Km 100% / span
Process
%CO
PID Kc Ti Td
% %PV
c2
Lag
c1
Km
m2
m2
m1
m1
Delay
Controller
Measurement
First Order Approximation: o v + p1 + p2 + m1 + m2 + c + v + p2 + m1 + m2 + c1 + c2 (set by automation system design for flow, pressure, level, speed, surge, and static mixer pH control)
123
PID Controllers
Ko = %PV / %CO
Self-regulating process gain (%/%)
Noise Band
ideally
p1
Time (seconds)
124
PID Controllers
%CO
Wireless Default Update Rate ramp rate is %PV1 / t1 ramp rate is %PV2 / t2
Time (seconds)
Wireless default update rate must be fast enough that excursion for maximum ramp rate is less than wireless trigger level that is set just larger than measurement noise
125
PID Controllers
Ko = %PV / %CO
Runaway process gain (%/%)
Acceleration
Wireless presently not advisable for runaway Tests are terminated before a noticeable acceleration leading to characterization as an integrating process %CO
For safety reasons, tests are terminated after 4 deadtimes Maximum speed in 4 deadtimes is critical speed
1.72%PV
%PV
Noise Band
must be
1 p
Time (seconds)
126
PID Controllers
Ki =
Ko
= Max( % PV / t ) / %CO
The above equation can be solved for the process time constant by taking the process gain to be 1.0 or for more sophistication as the average ratio of the controlled variable to controller output Tuning test can be done for a setpoint change if the PID gain is > 2 and the PID structure is PI on Error D on PV so you see a step change in controller output from the proportional mode
127
PID Controllers
3 o TNI = TSR o + 4 o
If the process time constant is greater than 6 times the deadtime
o 6 o
Then the near integrating tuning test time is reduced by > 90%:
128
PID Controllers
4 Ko
Ko
o
PV PV Sum
PV0 Initial Controlled Variable at time 0
CO
SUB
CO
Ki
PV
Switch
PV
1 ODE (Ki)
PV
First Principle Parameters = f (Ki) For the manipulation of jacket temperature to control vessel temperature, the near integrator gain is
K i = (U A ) / (C p M o )
Since we generally know vessel volume (liquid mass), heat transfer area, and process heat capacity, We can solve for overall heat transfer coefficient (least known parameter) to provide a useful ordinary differential equation (ODE) for a first principle model (1).
129
PID Controllers
Ex =
(o + o )
Eo
Largest lag in loop that is ideally set by large process volume
Integrated error is proportional to the ratio of loop deadtime squared to 63% response time
(Important to minimize quantity of product off-spec and total energy and raw material use)
o2 Ei = Eo (o + o )
Wireless default update rate affects ultimate performance limit because of default update rate is additional loop deadtime
130
PID Controllers
1 Ex = Eo (1 + Ko Kc )
Integrated error decreases as the controller gain increases and reset time decreases but is essentially the open loop error multiplied by the reset time plus signal delays and lags for systems when total deadtime >> process time constant
Ei =
Ti + tx + f Ko K c
Eo
Rise time (time to reach a new setpoint) is inversely proportional to controller gain
Tr =
PID Controllers
Kc = 0.4
Ko o
* - Ziegler Nichols method closed loop modified to be more robust and less oscillatory
Ti = 4 o
Td = p 2 = 0.5 o
1 Kc = 0.4 Ki o 1 Ki o
1 Kc = 0.4 Ko
Ti = 0.5 o
Td = 0
1.0 for Enhanced PID if Wireless Default Update Rate > Process Response Time !
Kc = 0.5
Ti = 4 o
Td = p 2 = 0.5 o Td = 2 p 2 = o
Kc = 0.6
Ko o
'o
Ti = 40 o
1 Kc = 0.6 Ki o
These tuning equations provide maximum disturbance rejection but will cause some overshoot of setpoint response
Wireless default update rate affects fastest controller tuning because of default update rate is additional loop deadtime
132
PID Controllers
o + w + v
T63
Eo
(o + w + v )2 Ei = Eo T63
T63 = o + w + o
w = Min( T , S )
T = 0.5 Tw
0.5 Sm S = ( % PV / t ) max
Eo
( % PV / t ) max = Ki ( Eo / Ko )
Ki =
Ko
( % PV / t ) max =
S =
0.5 Sm o Eo
133
PID Controllers
v =
( % PV / t ) max =
Eo
( %CO / t ) max =
K x Eo Ko o
v =
0.5 Sv Ko o K x Eo
Increase in process gain from elimination of controller reaction to noise by wireless trigger level or PID threshold sensitivity setting decreases deadtime from valve stick-slip, resolution, or deadband
134
PID Controllers
Kc = controller gain (dimensionless) Ki = integrating process gain (%/sec/% or 1/sec) Kp = process gain (dimensionless) also known as open loop gain L = change in load (engineering units) Fv = change valve flow (engineering units) %PV = change in process variable (%) SP = change in setpoint (engineering units) SPff = setpoint feedforward (engineering units) t = change in time (sec) tx = execution or update time (sec) o = total loop dead time (sec) f = filter time constant or well mixed volume residence time (sec) m = measurement time constant (sec) p2 = secondary (small) self-regulating process time constant (sec) p1 = primary (large) runaway process time constant (sec) p1 = primary (large) process time constant (sec) Ti = integral (reset) time setting (sec/repeat) Td = derivative (rate) time setting (sec) Tr = rise time for setpoint change (sec) to = oscillation period (sec) = Lambda (closed loop time constant or arrest time) (sec) f = Lambda factor (ratio of closed to open loop time constant or arrest time)
135
PID Controllers
PID Controllers
ZC1-4 OUT
FC
CAS
LC
1-1
1-6 reactant A
LY FT
ZC
1-4
1-5 1-3
1-8
1-3
1-2
TT TC
reactant B
FT
1-4
1-4 return
1-2
Valve position controller (VPC) setpoint is the maximum throttle position. The VPC should turn off integral action to prevent interaction and limit cycles. The correction for a valve position less than setpoint should be slow to provide a slow approach to optimum. The correction for a valve position greater than setpoint must be fast to provide a fast getaway from the point of loss of control. Directional velocity limits in AO with dynamic reset limit in an enhanced PID that tempers integral action can achieve these optimization objectives.
AT
AC
1-6
1-6
makeup CTW
FC
1-7
FT
product
1-7
137
PID Controllers
ZY1-1 OUT
FC
CAS
LC
1-1
1-8
PT PC
1-6 reactant A
LY FT
1-5
FT
TC
1-8
1-3
1-3
1-2
PT PC
1-4 reactant B
FT
1-4
1-1
ZC
LC
1-9
1-9
AT
AC
1-6
1-6
LT
1-9 BFW
FT
1-9
FC
1-7
ZC
1-4
FT
product
1-7
138
PID Controllers
139
PID Controllers
Valve position control should minimize interactions with temperature loop, oscillations from valve backlash and stick-slip, and prevent saturation of temperature control valve for large disturbances
Feature Direction Velocity Limits Dynamic Reset Limit Adaptive Tuning Feedforward Enhanced PID
Function Limit VPC Action Speed Based on Direction Limit VPC Action Speed to Process Response Automatically Identify and Schedule Tuning Preemptively Set VPC Out for Upset Suspend Integral Action until PV Update
Advantage 1 Prevent Running Out of Valve Direction Velocity Limits Eliminate Manual Tuning Prevent Running Out of Valve Eliminate Limit Cycles from Stiction & Backlash
Advantage 2 Minimize Disruption to Process Prevent Burst of Oscillations Compensation of Nonlinearity Minimize Disruption Minimize Oscillations from Interaction & Delay
140
PID Controllers
1-5
AC AT FT
1-6
1-6
Fast reaction, short residence time, and high heat release prevents inverse response in manipulation of reactant feed rate for temperature control.
BFW steam
1-3a
TT
FC FY
BFW
steam
1-3b
TT
1-3
1-3
1-1 BFW
FT
1-6 1-1 CA S
FC
1-3c
1-2
gas reactant B
FT
Temperature controller inherently maximizes reactant feed rate to amount permitted by the number of BFW coils in service
141
1-2
PID Controllers
V
FC LC FC
FC
TC
Change in temperature controller output does not affect column until level controller output changes so high level PID gain or feedforward is needed The level control of reflux provides some internal reflux control by decreasing reflux for an decrease in vapor flow from cold weather or rain storm
LC FC
FC
142
PID Controllers
V
FC LC FC
FC
TC
FC
LC FC
143
PID Controllers
V
FC LC FC
D
If the distillate flow (D) was on auto with a local set point instead of cascade with a remote set point from level, the material balance would be fixed and we would have separation control rather material balance control. Separation control is not recommended because it requires much larger adjustments to the V/F, energy per unit feed to control composition severely limiting the range of control (size of upsets and set point changes handled before low and high V causes weeping and flooding)
FC
FC
TC
FC
LC
144
PID Controllers
V
FC LC FC
FC
TC
steam level
FC
Inverse Response
LC FC
145
PID Controllers
+1%D/F
-1%D/F
146
PID Controllers
Temperature Sensitivity
100 99.8 99.6 99.4
DISTILLATE PRODUCT
COMPOSITION, wt%
BOTTOM PRODUCT
PID Controllers
Disturbance Analysis
148
PID Controllers
149
PID Controllers
150