Standards
Certification
Education & Training
Publishing
Conferences & Exhibits
Biological-Chemical Reactor ControlPrinciples and Methods forImproving Product Quality and Optimizing Production Rate
22
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 numerous 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. Greg’s expertise is available on the web site: http://www.modelingandcontrol.com/
http://www.modelingandcontrol.com/
3
Top Ten Things You Don’t Want to Hear in a Project Definition Meeting
• (10) I don’t want any smart instrumentation talking back to me• (9) Let’s study each loop to see if the valve really needs a positioner• (8) Lets slap an actuator on our piping valves and use them for control
valves• (7) We just need to make sure the control valve spec requires the
tightest shutoff• (6) What is the big deal about process control, we just have to set the
flow per the PFD• (5) Cascade control seems awfully complex• (4) The operators can tune the loops• (3) Let’s do the project for half the money in half the time• (2) Let’s go with packaged equipment and let the equipment supplier
select and design the automation system• (1) Let’s go out for bids and have purchasing pick the best deal
4
Introduction
• Biological and chemical process performance is largely determined by reactor performance. The stage for product quality and process efficiency and capacity is set by reaction yield and selectivity.
• An increase in yield can be used to increase production rate for same feed rate or used to decrease raw material costs for the same production rate.
• For batch reactors, an increase in yield can be taken as shorter cycle time for the same charges or as smaller charges for the same cycle time.
• A higher yield reduces downgraded products, recycle, and waste. • Reactor type, reaction rate, and time available for reaction affect yield.• Temperature, concentration, and sometimes pressure affect reaction rate.• Inventory and feed rate determines the amount of time reactants are in
reactor (residence time), which determines time available for reaction.• Process control of temperature, concentration, pressure, inventory, and feed
rate is essential to achieve reaction rate and time for maximum yield.• Endpoint control inherently prevents the accumulation of excess reactants.• Valve position control increases reactant feed rate to limits of utility systems.• Valve position control increases rangeability of utility systems.
5
Reactor TypeDynamics and Control
ReactorType
DynamicResponse
Residence Time Distribution
ReactionRate
AdditionalControl *
CSTRNear-Regulating
Runaway Wide ModerateLevel
Composition
BatchIntegratingRunaway Very Tight Slow None
Fed-BatchIntegratingRunaway Very Tight Slow Time Profile
Plug FlowSelf-Regulating
Runaway Tight Fast Length Profile
GasSelf-Regulating
Runaway Tight Very Fast Composition
* - Additional control besides temperature and pressure control
6
Liquid Reactants (Jacket CTW) Liquid Product Basic Control
TT1-4
TC1-3
TC1-4
AT 1-6
LY 1-8
FY 1-6
FT1-2
FC1-2
reactant A
reactant B
CAS
residencetime calc
CAS
ratiocalc
AC1-6
makeup
return
LY 1-8
FY 1-6
reactant A
reactant B
residencetime calc LT
1-8TT1-3
LC1-8
CAS
ratiocalc
product
ventFT1-1
FC1-1
FC1-7
FT 1-7
CAS
PT1-5
PC1-5
FT 1-5
CTW
7
Liquid Reactants (Jacket CTW) Liquid Product Optimization
TT1-4
TC1-3
TC1-4
AT 1-6
LY 1-8
FY 1-6
FT1-2
FC1-2
reactant A
reactant B
CAS
residencetime calc
CAS
ratiocalc
AC1-6
makeup
return
LY 1-8
FY 1-6
reactant A
reactant B
residencetime calc LT
1-8TT1-3
LC1-8
CAS
ratiocalc
product
ventFT1-1
FC1-1
FC 1-1CAS
ZC1-4OUT
ZC1-4
FC1-7
FT 1-7
PT1-5
PC1-5
FT 1-5
CTW
ZC1-4 is an enhanced PID VPC
Valve position controller (VPC) setpointis the maximum throttle position. TheVPC should turn off integral action toprevent interaction and limit cycles. Thecorrection for a valve position less thansetpoint should be slow to provide a slow approach to optimum. The correction fora valve position greater than setpoint mustbe fast to provide a fast getaway from thepoint of loss of control. Directional velocitylimits in AO with dynamic reset limit in anenhanced PID that tempers integral actioncan achieve these optimization objectives.
8
Liquid Reactants (Jacket BFW) Liquid Product Basic Control
AT 1-6
LY 1-8
FY 1-6
FT1-2
FC1-2
reactant A
reactant B
CAS
residencetime calc
CAS
ratiocalc
AC1-6
steam
PT1-4
TC1-3
PC1-4
LY 1-8
FY 1-6
reactant A
reactant B
residencetime calc LT
1-8TT1-3
LC1-8
CAS
ratiocalc
product
ventFT1-1
FC1-1
LT1-9
LC1-9
FC1-7
FT 1-7
PT1-5
PC1-5
FT 1-5
BFW FT 1-9
9
Liquid Reactants (Jacket BFW) Liquid Product Optimization
AT 1-6
LY 1-8
FY 1-6
FT1-2
FC1-2
reactant A
reactant B
CAS
residencetime calc
CAS
ratiocalc
AC1-6
steam
PT1-4
TC1-3
PC1-4
LY 1-8
FY 1-6
reactant A
reactant B
residencetime calc LT
1-8TT1-3
LC1-8
CAS
ratiocalc
product
ventFT1-1
FC1-1
LT1-9
LC1-9
BFW
ZC1-4
ZC1-9
ZY 1-1
FC1-1CAS
ZY1-1OUT
low signalselector
FC1-7
FT 1-7
PT1-5
PC1-5
FT 1-5
FT 1-9
ZC1-4 & ZC-9 are enhanced PID VPC
10
Gas Reactants (Jacket BFW) Gas Product Optimization
AT 1-6
FT1-2
FC1-2
CAS
AC1-6
TC1-3
FY 1-6
gas reactant A
TT1-3b
TT1-3a
CAS
ratiocalc
product
FT1-1
FC1-1
steamBFW
gas reactant B
TT1-3c
steamBFW
steamBFW
TY 1-3
high signalselector
average bedtemperatures
Fluidized BedCatalytic Reactor
PT1-5
PC1-5
FT 1-5
Fast reaction, short residence time, and high heat release prevents inverse response in manipulation of reactantfeed rate for temperature control.
Temperature controller inherentlymaximizes reactant feed rate toamount permitted by the numberof BFW coils in service
11
Gas & Liquid Reactants (Jacket CTW) Liquid Product End Point Control
TT1-4
PC1-5
TC1-3
TC1-4
AT 1-6
LY 1-8
FY 1-6
FT1-2
FC1-2
gas reactant A
liquid reactant B
CAS
residencetime calc
CAS
ratiocalc
AC1-6
makeup
return
LY 1-8
FY 1-6
residencetime calc
CAS
ratiocalc
product
purgeFT1-1
FC1-1
FC1-5
FT 1-5
LT1-8
TT1-3
PT1-5
LC1-8
CAS
FC1-7
FT 1-7
CTW
Pressure control inherentlyprevents an excess of gas reactant providing endpointcontrol for liquid product in continuous andfed-batch reactors.
12
Gas & Liquid Reactants (Jacket CTW) Gas Product End Point Control
TT1-4
TC1-3
TC1-4
AT 1-6
LY 1-8
FY 1-6
FT1-2
FC1-2
liquid reactant A
gas reactant B
CAS
residencetime calc
CAS
ratiocalc
AC1-6
makeup
return
LY 1-8
FY 1-6
residencetime calc LT
1-8TT1-3
LC1-8
CAS
ratiocalc
purge
product
FT1-1
FC1-1
CAS
FC1-7
FT 1-7
PC1-5
FT 1-5
PT1-5
CTW
Level control inherentlyprevents an excess of liquidreactant providing endpointcontrol for gas product incontinuous reactors.
13
Liquid Reactants (Jacket CTW) Gas & Liquid Products Basic Control
TT1-4
TC1-4
AT 1-6
LY 1-8
FY 1-6
FT1-2
FC1-2
reactant A
reactant B
residencetime calc
CAS
ratiocalc
AC1-6
makeup
return
LY 1-8
FY 1-6
reactant A
reactant B
CAS
residencetime calc LT
1-8
LC1-8
CAS
ratiocalc
product
FT1-1
FC1-1
FC1-7
FT 1-7
CAS
TC1-3
TT1-3
product
PC1-5
FT 1-5
W
PT1-5
TT1-10
TC1-10
CTW
14
Liquid Reactants (Jacket CTW) Gas & Liquid Products Optimization
TT1-4
TC1-4
AT 1-6
LY 1-8
FY 1-6
FT1-2
FC1-2
reactant A
reactant B
residencetime calc
CAS
ratiocalc
AC1-6
makeup
return
LY 1-8
FY 1-6
reactant A
reactant B
CAS
residencetime calc LT
1-8
LC1-8
CAS
ratiocalc
product
FT1-1
FC1-1
FC1-7
FT 1-7
CAS
TC1-3
TT1-3
product
PC1-5
FT 1-5
W
PT1-5
TT1-10
TC1-10
ZC1-4
ZC1-10
ZC1-5
ZY 1-1
FC1-1CAS
ZY1-1OUT
low signalselector
ZC-10OUT
ZC-4OUT
ZC-5OUT
ZY-1IN1
ZY-1IN2
ZY-1IN3
CTW
ZC1-4, ZC-5, & ZC-10 are enhanced PID VPC
15
Liquid Reactants (Jacket CTW) Liquid Product with Recycle Basic Control
TT1-4
TC1-3
TC1-4
AT 1-6
LY 1-8
FY 1-6
FT1-2
FC1-2
reactant A
reactant B
residencetime calc
CAS
ratiocalc
AC1-6
makeup
return
LY 1-8
FY 1-6
reactant A
reactant B
residencetime calc LT
1-8TT1-3
LC1-8
CAS
ratiocalc
product
ventFT1-1
FC1-1
FC1-7
FT 1-7
PT1-5
PC1-5
FT 1-5
Recycle fromrecovery column distillate receiver
CAS CAS
CTW
A flow controller in the recyclepath from reactant in product back as recovered reactant is necessary to prevent a snowballing effect and divergenceof reactant concentration. In thiscase the flow controller is on thereactor discharge and recovery column distillate receiver level controller provides fresh makeup of recycled reactant as needed.
16
Liquid Reactants (Jacket CTW) Liquid Product with Recycle Optimization
TT1-4
TC1-3
TC1-4
AT 1-6
LY 1-8
FY 1-6
FT1-2
FC1-2
reactant A
reactant B
residencetime calc
CAS
ratiocalc
AC1-6
makeup
return
LY 1-8
FY 1-6
reactant A
reactant B
residencetime calc LT
1-8TT1-3CAS
ratiocalc
product
ventFT1-1
FC1-1
FC1-7
FT 1-7
PT1-5
PC1-5
FT 1-5
Recycle fromrecovery column distillate receiver
ZC1-4
LC1-8
CAS CAS
CTW
ZC1-4 is an enhanced PID VPC
The product discharge flowcontroller setpoint, whichsets reactor production ratecan be increased by a VPC to the maximum throttle limit of coolant system valve.
17
Jacket CTWOutlet Temperature Control
TT1-4
TC1-4
makeup
return
CTW
18
Jacket CTWInlet Temperature Control
TT1-4
TC1-4
makeup
return
CTW
19
Jacket CTW with Steam Injection Outlet Temperature Control
TT1-4
TC1-4
TY1-4splitter
stea
m
steaminjectionheater CTW
makeup
return
20
Reactor Heat Exchanger Recirc Temperature Control
TT1-4
product
W
TT1-4
TC1-4
CTW
21
Reactor Heat Exchanger Recirc Temperature Bypass Control
TT1-4
TC1-4
product
W
CTW
22
Reactor Heat Exchanger Recirc Temperature Bypass Optimization
TT1-4
TC1-4
product
W
ZC1-4
CTW
ZC1-4 is an enhanced PID VPC
23
Jacket Steam & CTW Heat Exchanger Inlet Temperature Control
TT1-4
TC1-4
W
W
TY1-4
CTW
steam
splitter
24
Jacket CTW Heat ExchangerInlet Temperature Bypass Control
TT1-4
TC1-4
WCTW
25
Jacket CTW Heat Exchanger Inlet Temperature Bypass Optimization
TT1-4
TC1-4
Wchilledwater
ZC1-4
ZC1-4 is an enhanced PID VPC
26
Reactor Inferential Measurements ofHeat Transfer Rate and Conversion
TT1-4a
makeup
return
TT1-4bDT
SUB
FI 1-4MUL
SpecificHeat
CapacityCp
Heat Transfer Rate
INTIntegration
Period Total Heat Transferred (Inferential Measurement of Conversion)
DIVJacketVolumeJacketFlow
Integration Period is continuous reactor residence time or batch reactor cycle time
27
Batch Reactor ConcentrationProfile Slope Control
TT1-4
TC1-3
TC1-4
AT 1-6
LY 1-8
FY 1-6
FT1-2
FC1-2
reactant A
reactant B
CAS
residencetime calc
CAS
ratiocalc
AC1-6
makeup
return
LY 1-8
FY 1-6
reactant A
reactant B
residencetime calc LT
1-8TT1-3
LC1-8
CAS
ratiocalc
product
ventFT1-1
FC1-1
FC1-7
FT 1-7
CAS
PT1-5
PC1-5
FT 1-5
DT
SUB
DIV
LoopDeadtime
ΔPV
SlopeΔPV/Δt
2828
Innovative PID System to Optimize Ethanol Yield and Carbon Footprint
AT 1- 4
AC 1- 4
DC 2- 4
SC 1-4
FT 1- 5
FC 1- 5
AY 1- 4
Corn
NIR-T
Production RateEnhanced PID
DT 2- 4
Slurry SolidsEnhanced PID
DX 2- 4
Feedforward
FT 1- 6
FC 1- 6
Backset Recycle
Dilution Water
XC 1- 4
XY 1- 4
Average Fermentation TimeEnhanced PID
Fermentable Starch Correction
SlurryTank 1
SlurryTank 2
CoriolisMeter
setpoint
DY 2- 4
Lag and Delay
RCAS
Predicted Fermentable Starch
29
Bioreactor
VSD
VSD
TC 1-7
AT 1-4s2
AT 1-4s1
AT 1-2
AT 1-1
TT 1-7
AT 1-6
LT 1-8
Glucose
Glutamine
pH
DO
Product
HeaterVSD
VSD
Media
Glucose
Glutamine
Inoculums
VSDBicarbonate
AY 1-1
AC 1-1
Splitter
AC 1-2
AY 1-2
Splitter
CO2
O2
Air
0.002 g/L
7.0 pH
37 oC
MFC
MFC
AT 1-5x1
ViableCells
AT 1-5x2DeadCells
MFC - Mass Flow ControllerVSD - Variable Speed Drive
MFC
wireless
wireless
spargers
Mammalian Bioreactor WithEnhanced DO and pH Control
AT 1-9
Off-gas
Calc
OURMass Spec
AC1-1 and AC1-2are Enhanced PID
Product
30
Bioreactor
VSD
VSD
TC 1-7
AT 1-4s2
AT 1-4s1
AT 1-2
AT 1-1
TT 1-7
AT 1-6
LT 1-8
Glucose
Glutamine
pH
DO
Product
HeaterVSD
VSD
AC 1-4s1
AC 1-4s2
Media
Glucose
Glutamine
Inoculums
VSDBicarbonate
AY 1-1
AC 1-1
Splitter
AC 1-2
AY 1-2
Splitter
CO2
O2
AirProduct
0.002 g/L
7.0 pH
2.0 g/L
2.0 g/L
37 oC
MFC
MFC
AT 1-5x1
AT 1-5x2
MFC - Mass Flow ControllerVSD - Variable Speed Drive
MFC
wireless
wireless
spargers
Mammalian Bioreactor WithEnhanced Substrate Control
Substrate (Glutamine/Glucose) RatioControl with OUR Feedforward
AT 1-9
Off-gas
Mass Spec
Calc
FF
RatioOUR
AC1-1 and AC1-2are Enhanced PID
AC1-4s1 and AC1-4s2are Enhanced PID
ViableCells
DeadCells
31
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
32
Enhanced PID Algorithm Originally Developed for Wireless
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 valuePID reset and rate action are only computed when there is a new valueIf transmitter damping is set to make noise amplitude less than communication trigger level, valve packing and battery life is dramatically improvedEnhancement compensates for measurement sample time suppressing oscillations and enabling a smooth recovery from a loss in communications further extending packing -battery life
+
+
+
+
Elapsed Time
Elapsed Time
TD
Kc
Kc
TD
http://www2.emersonprocess.com/siteadmincenter/PM%20DeltaV%20Documents/Whitepapers/WP_DeltaV%20PID%20Enhancements%20for%20Wireless.pdf
Link to Enhanced PID White Paper
33
Broadley-James Corporation Wireless Bioreactor Setup
• Hyclone 100 liter Single Use Bioreactor (SUB)
• Rosemount WirelessHART gateway and transmitters for measurement and control of pH and temperature. (pressure monitored)
• BioNet lab optimized control system based on DeltaV
34
Wireless Bioreactor pHGain 30 Reset 1200
35
Gain 40 Reset 500
Output comes off high limit at 36.8 oC
0.30 oC overshoot
Wireless Bioreactor Temperature
36
Gain 40 Reset 5000
Output comes off high limit at 35.9 oC
0.12 oC overshoot
Wireless Bioreactor Temperature
37
Gain 40 Reset 10000
0.13 oC overshoot
Output comes off high limit at 36.1 oC
Wireless Bioreactor Temperature
38
Gain 40 Reset 15000
0.20 oC overshoot
Output comes off high limit at 36.4 oC
Wireless Bioreactor Temperature
39
Gain 80 Reset 15000
0.11 oC overshoot
Output comes off high limit at 36.1 oC
Wireless Bioreactor Temperature
Best Tuning: < 0.01 oC overshoot for Gain = 80 Reset = 5000
40
Bioreactor
VSD
VSD
TC 1-7
AT 1-4s2
AT 1-4s1
AT 1-2
AT 1-1
TT 1-7
AT 1-6
LT 1-8
Glucose
Glutamine
pH
DO
Product
HeaterVSD
VSD
AC 1-4s1
AC 1-4s2
Media
Glucose
Glutamine
Inoculums
VSDBicarbonate
AY 1-1
AC 1-1
Splitter
AC 1-2
AY 1-2
Splitter
CO2
O2
AirProduct
0.002 g/L
7.0 pH
37 oC
MFC
MFC
AT 1-5x2
AT 1-5x2
MFC - Mass Flow ControllerVSD - Variable Speed Drive
MFC
wireless
wireless
spargers
Mammalian Bioreactor WithViable Cell and Product Profile Control
AT 1-9
Off-gas
Mass Spec
Calc
FF
RatioOUR
AC1-1 and AC1-2are Enhanced PID
MPCCalc
Calc
ProductFormation
Rate
GrowthRate
ViableCells
DeadCells
AC1-4s1 and AC1-4s2are Enhanced PID
41
Model Predictive Control (MPC) Identified Responses for Profile
42
Product Formation Rate
Biomass Growth rate
Substrate
Dissolved Oxygen
Model Predictive Control (MPC) Profile Control Test
43
Batch Basic Fed-Batch APC Fed-BatchBatch
Inoculation Inoculation
Dissolved Oxygen (AT6-2)
pH (AT6-1)
Estimated Substrate Concentration (AT6-4)
Estimated Biomass Concentration (AT6-5)
Estimated Product Concentration (AT6-6)
Estimated Net Production Rate (AY6-12)
Estimated Biomass Growth Rate (AY6-11)
MPC in Auto
Model Predictive Control (MPC) Profile Control Cycle Time Reduction
44
Current Product Yield (AY6-10D)
Current Batch Time (AY6-10A)
Predicted Batch Cycle Time (AY6-10B)
Predicted Cycle Time Improvement (AY6-10C)
Predicted Final Product Yield (AY6-10E)
Predicted YieldImprovement (AY6-10F)
Batch Basic Fed-Batch APC Fed-BatchBatchInoculation
Inoculation
MPC in Auto
Predicted Final Product Yield (AY6-10E)
Predicted Batch Cycle Time (AY6-10B)
Model Predictive Control (MPC) Profile Control Improved Predictions
45
τp1 θp2 τp2 Kpθp1
τc1 τm2 θm2 τm1 θm1Kmθcτc2
Kc Ti Td
Valve Process
Controller Measurement
Kvτvθv
KLτLθLLoad Upset
Δ%PV
Δ%CO
ΔMVΔPV
PID
Delay Lag
Delay Delay Delay
Delay
Delay
Delay
Lag Lag Lag
LagLagLag
Lag
Gain
Gain
Gain
Gain
LocalSet Point
ΔDV
First Order Approximation: θo ≅ θv + θp1 + θp2 + θm1 + θm2 + θc + τv + τp1 + τm1 + τm2 + τc1 + τc2(set by automation system design for flow, pressure, level, speed, surge, and static mixer pH control)
%
%%
Delay Dead TimeLag Time Constant
For integrating processes: Ki = Kv ∗ (Kp / τp2 ) ∗ Km
100% / span
Loop Block Diagram (First Order Approximation)
Hopefully τp2 is the largest lag in the loop
½ of Wireless Default Update Rate
46
Time (seconds)
% Controlled Variable (%PV) or
% Controller Output (%CO)
Δ%CO
Δ%PV
θo
Ko = Δ%PV / Δ%CO
0.63∗Δ%PV
%CO
%PV
Self-regulating process open loopnegative feedback time constant
Self-regulating process gain (%/%)
Response to change in controller output with controller in manual
observed total loopdeadtime
ideally τp2τo
Maximum speedin 4 deadtimesis critical speed
Open Loop Response ofSelf-Regulating Process
Noise Band
For CSTR τo >> θo processresponse appears to ramp for 10 θo and is termed a “near-integrating” process
For plug flow reactor andthe manipulation of feed θo >> τo process responseis a transport delay and istermed “deadtime dominant”
47
Time (seconds)θo
Ki = { [ %PV2 / Δt2 ] − [ %PV1 / Δt1 ] } / Δ%CO
Δ%CO
ramp rate isΔ%PV1 / Δt1
ramp rate isΔ%PV2 / Δt2
%CO
%PV
Integrating process gain (%/sec/%)
Response to change in controller output with controller in manual% Process Variable (%PV)
or% Controller Output (%CO)
observed total loopdeadtime
Maximum speedin 4 deadtimesis critical speed
Open Loop Response ofIntegrating Process
Wireless Trigger Level > Noise
Wireless DefaultUpdate Rate
Wireless default update rate must be fastenough that excursion for maximum ramprate is less than wireless trigger level that is set just larger than measurement noise
48
Response to change in controller output with controller in manual
Noise Band
Acceleration
Δ%PV
Δ%CO
1.72∗Δ%PV
Ko = Δ%PV / Δ%CO Runaway process gain (%/%)
% Process Variable (%PV) or
% Controller Output (%CO)
Time (seconds)observed total loopdeadtime runaway process open looppositive feedback time constant
For safety reasons, tests are terminated after 4 deadtimes
Maximum speedin 4 deadtimesis critical speed
Open Loop Response ofRunaway Process
Wireless presently notadvisable for runaway
τ’o must be τ’p2θo
Tests are terminatedbefore a noticeableacceleration leadingto characterization asan integrating process
49
COtPVMaxKKo
oi %/)/%( ΔΔΔ== τ
For “Near Integrating” gain approximation use maximum ramp rate divided by change in controller output
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 changein controller output from the proportional mode
Near Integrator Gain Approximation
The maximum ramp rate is found by passing filtered process variable (PV) through a deadtime (DT) block to create an old process variable. The deadtime block uses the total loop deadtime (θo) for the time interval (Δt ). The old process variable is subtracted from the new process variable and divided by the time interval to get the ramp rate. The maximum of a continuous train of ramp rates updated each module execution over A period of 3 or more deadtimes is selected to compute the near integrating process gain. For an inverseresponse or large secondary time constant, the computation may need to continue for 10 or more deadtimes.
50
SRoo
oNI TT ∗∗+
∗=
τθθ4
3
The near integrating test time (3 deadtimes) as a fraction of the self-regulating test (time to steady state is taken as 98% response time TSR = T98 = θo + 4 το ) is:
If the process time constant is greater than 6 times the deadtime
oθτ ∗≥ 6oThen the near integrating tuning test time is reduced by > 90%:
SRNI TT ∗≤ 1.0
For example:
sec100o =τ sec4o =θThe near integrator tuning time is reduced by 97%!
SRNI TT ∗≤ 03.0
Reduction inIdentification Test Time
51
4
PVSUB
First Principle Parameters = f (Ki)
CO0Initial Controller Output at time 0
∆COθo
Ko
Ki∆PV Switch
ODE (Ki)
∆PV
∆PV
Sum
PV0Initial Controlled Variable at time 0
Ko = PV0 / CO0 process gain approximationτo = Ko / Ki negative feedback time constantτ’o = Ko / Ki positive feedback time constant
Methodology extends beyond loops to anyprocess variable that can be measured
and any variablethat can be changed
CO
τo
Ko τ’o
1
2
3
∆PV
For the manipulation of jacket temperature to control vessel temperature, the near integrator gain is
)(/)( opi MCAUK ∗∗=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 usefulordinary differential equation (ODE) for a first principle model (1).
Rapid Process Model Identification and Deployment Opportunity
52
The observed deadtime (θo ) and integrator gain (Ki) are identified after a change in any controller output (e.g. final control element or setpoint) or any disturbance measurement. The identification of the integrator gain uses the fastest ramp rate over a short time period (e.g. 2 dead times) at the start of the process response.
The models are not restricted to loops but can be used to identify the relationship between any variable that can be changed and any affected process variable that can be measured.
The models are used for processes that are have a true integrating response or slow processes with a “near integrating” response (τo > 5∗θo ). The process deadtime and integrating process gain can be used for controller tuning and for plant wide simulations including but not limited to the following types of models:
Model 1: Hybrid ordinary differential equation (ODE) first principle and experimental model Model 2: Integrating process experimental modelModel 3: Slow self-regulating experimental modelModel 4: Slow non-self-regulating positive feedback (runaway) experimental model
Patent disclosure filed on 3-1-2010
Rapid Process Model Identification and Deployment Opportunity
53
ooo
ox EE ∗+
=)( τθ
θ
ooo
oi EE ∗+=
)(
2
τθθ
Peak error is proportional to the ratio of loop deadtime to 63% response time(Important to prevent SIS trips, relief device activation, surge prevention, and RCRA pH violations)
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)
Loop Performance Ultimate Limit
Total loop deadtimethat is often set byautomation design
Largest lag in loopthat is ideally set by
large process volume
Wireless default update rate affects ultimate performance limitbecause ½ of default update rate is additional loop deadtime
54
oco
x EKKE ∗
∗+=
)1(1
oco
fxii EKK
tTE ∗
∗++
=τΔ
Peak error decreases as the controller gain increases but is essentially the open loop error for systems when total deadtime >> process time constant
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
Open loop error forfastest and largestload disturbance
ocffi
r SPKSPCOKSPT θ+
∗+=
)|,min(|( max ΔΔΔ
Rise time (time to reach a new setpoint) is inversely proportional to controller gain
Loop PerformancePractical Limit
55
oo
oc K
Kθ
τ∗
∗= 4.0 oiT θ∗= 4 1d pT τ=
For runaway processes:
For self-regulating processes:
oic K
Kθ∗
∗=15.0 oiT θ∗= 4 1d pT τ=
oic K
Kθ∗
∗=16.0
oiT θ∗= 40 1d 2 pT τ∗=
For integrating processes:
oo
oc K
Kθ
τ∗
∗='6.0
oic K
Kθ∗
∗=14.0
Near integrator (τp2 >> θo):
oiT θ∗= 5.0
Near integrator (τ’p2 >> θo):
Deadtime dominant (τp2 Process Response Time !
Wireless default update rate affects fastest controller tuningbecause ½ of default update rate is additional loop deadtime
56
Effect of Wireless MeasurementUpdate Time and Interval on Performance
o
omS E
S τθ ∗∗= 5.0
ovwo
x ETE ∗++=
63
θθθo
vwoi ET
E ∗++=63
2)( θθθowoT τθθ ++=63
),( STw Min θθθ Δ= wT TΔΔ ∗= 5.0θmax)/%(
5.0tPV
SmS ΔΔ
∗=θ
)/()/%( max ooi KEKtPV ∗=ΔΔo
oi
KKτ
=o
oE
tPVτ
=max)/%( ΔΔ
57
Additional Deadtime fromValve Stick-Slip, Resolution, or Deadband
ox
oovv EK
KS∗
∗∗∗=
θθ 5.0
max)/%(5.0
tCOSv
v ΔΔ∗
=θ maxmax )/%()/%( tPVKtCO c ΔΔΔΔ ∗=
oo
x
KEK
tCO oθ∗
∗=max)/%( ΔΔ
[ ]⎥⎦⎤
⎢⎣
⎡−∗
∗=
002.0),(max,min
mm
v
oo
oxc SN
SKKK
θτ
o
oE
tPVτ
=max)/%( ΔΔ
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
58
ΔCV = change in controlled variable (change in process variable in % of scale)ΔCO = change in controller output (%)Kc = controller gain (dimensionless)Ki = integrating process gain (%/sec/% or 1/sec)Kp = process gain (dimensionless) also known as open loop gainDV = disturbance variable (engineering units)MV = manipulated variable (engineering units)PV = process variable (engineering units)ΔSP = change in setpoint (engineering units)SPff = setpoint feedforward (engineering units)Δt = change in time (sec)Δtx = execution or update time (sec)θo = total loop dead time (sec)τf = filter time constant or well mixed volume residence time (sec)τm = measurement time constant (sec) Τp2 = primary (large) self-regulating process time constant (sec) τ’p2 = primary (large) runaway process time constant (sec) τp1 = secondary (small) 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)λ = Lambda (closed loop time constant or arrest time) (sec)λf = Lambda factor (ratio of closed to open loop time constant or arrest time)
Nomenclature(Process Dynamics & Performance)
59
Ei = integrated error for unmeasured load disturbance (% sec)Ex = peak error for unmeasured load disturbance (%)Eo = open loop error (loop in manual) for unmeasured load disturbance (%)Ki = near integrator process gain (% per % per sec)Ko = open loop gain (product of valve, process, and measurement gains) (dimensionless)Kx = detuning factor for controller gain (dimensionless)Nm = measurement noise (%)Δ%CO/Δt = rate of change in PID % controller output (% per sec)Δ%PV/Δt = rate of change in PID % process variable (% per sec)ΔTw = wireless default update rate (update time interval) (sec)Sm = wireless measurement trigger level (threshold sensitivity) (%)Sv = valve stick-slip, resolution, or deadband (%)T63 = 63% process response time (sec)θo = original loop deadtime (sec)θΔt = additional deadtime from default update rate (sec)θs = additional deadtime from wireless trigger level (sec)θv = additional deadtime from valve (sec)θw = additional deadtime from wireless measurement (sec)τo = self-regulating open loop time constant (largest time constant in loop) (sec)τ’o = runaway open loop time constant (largest time constant in loop) (sec)
Nomenclature(Wireless Dynamics & Performance)
60
Key Insights
• A liquid or solids phase reaction without a continuous liquid or solids discharge flow is the distinguishing characteristic of batch and fed-batch.
• There is generally no level control except perhaps in terms of a high level override or high level shutdown of feeds in batch and fed-batch reactors.
• In batch reactors, the reactants are fed sequentially and shutoff when charge tank weight or flow totals indicate the total charged is complete.
• In fed-batch operation, the reactants are fed simultaneously under flow control at a rate determined by a control system.
• Many of the same controls used for continuous reactors are applicable to fed-batch except typically there is no level or residence time control.
• There is a profile of temperature, physical properties, and composition with respect to length for a plug flow reactor and with respect to batch time for a fed-batch vessel. In both types of reactors opportunities exist for profile control and optimization. The temperature may be controlled at various setpoints depending upon length and time. Since composition generally goes in one direction only with length and time, the slope is controlled at points in length and time for composition profile control.
61
Key Insights
• Gas phase reactors are generally continuous with a short tight residence time.• The process deadtime from transportation delay of gas reactants is small
compared to the lags from catalyst heat capacity and thermowell design.• Fast temperature control is possible by manipulation of gas reactant flows.• Mature high capacity products (e.g. oil, gas, and petrochemicals) tend to use
continuous reactors whereas new high value processes (e.g. specialty chemical and biological) primarily use batch and fed-batch reactors.
• The fastest and simplest implementation is batch with quantities charged sequentially mimicking lab experiments. As knowledge is gained batch reactors can become fed-batch reactors and eventually continuous reactors if there is enough demand and chemistry permits a variable residence time.
• Candidates for continuous reactors are products with a low profit margin, high volume requirement, fast reaction, minimal adverse reactions, preventable buildup of inhibitors and inactive components, and an extensive R&D history.
• Candidates for batch reactors are products with a high profit margin, low volume requirement, slow reaction, significant side effects, and minimal R&D.
62
Key Insights
• Reactors have 3 types of dynamic responses observed when the PID is put in manual and a step change is made in PID output with no disturbances.
– If the response lines out at a steady state, the process is “self-regulating.” – If the response continues to ramp (no steady state), the process is “integrating.” – If the response continues to accelerate, the process is “runaway.”
• Inverse response can occur in any of the above responses when the initial response is in the opposite direction of eventual response.
– Temperature control by manipulation of cold feed can exhibit an inverse response• A CSTR has a slow self-regulating response.• A batch and fed-batch reactor has a slow integrating response. • A plug flow and gas phase reactor has a fast self-regulating response.• All of these reactors can develop a runaway response when the increase in
reaction heat release with temperature exceeds the cooling capability.
63
Key Insights
• Reaction rate depends upon temperature and composition. • To prevent an excess or deficiency of reactants, the reactant concentration
must be in the ratio set by the stoichiometric equation for the reaction.• High capacity products such as petrochemicals and intermediates greatly
depend upon the added value of chromatographs because even a fractional percent increase in production rate is millions of dollars.
• Biological reactions utilize a wide variety of composition measurements including potentiometric and amperiometric electrodes for substrates and waste products, and dielectric spectroscopy and digital imagery for viable cell concentration, and near infrared (NIR) spectroscopy for compounds.
• The average amount of time reactants stay in contact is called residence time in continuous operations and is simply volume divided by flow rate.
• For fed-batch and batch, batch cycle time is used instead of residence time.• In batch reactors, the batch cycle time is often longer than necessary.• The process dynamics for vessels offer incredibly tight concentration, level,
pH, pressure, and temperature control.
64
Resource
Advanced Temperature Measurement and Control - 2nd Ed has a lot of detailson how to improve temperature sensoraccuracy and temperature control of heat exchangers, reactors, and kilns
Biological and Chemical Reactor Controlis an ISA book whose kernel is the ISAAutomation Week tutorial. The book will beavailable in December just in time for the holidays. Surprise your spouse with the bookas a “gift that keeps on giving.” Your spousewill never let you forget it.
Biological-Chemical Reactor ControlPresenterTop Ten Things You Don’t Want to �Hear in a Project Definition MeetingIntroductionReactor Type�Dynamics and ControlLiquid Reactants (Jacket CTW) �Liquid Product Basic ControlLiquid Reactants (Jacket CTW) �Liquid Product OptimizationLiquid Reactants (Jacket BFW) �Liquid Product Basic ControlLiquid Reactants (Jacket BFW) �Liquid Product OptimizationGas Reactants (Jacket BFW) �Gas Product OptimizationGas & Liquid Reactants (Jacket CTW) �Liquid Product End Point ControlGas & Liquid Reactants (Jacket CTW) �Gas Product End Point ControlLiquid Reactants (Jacket CTW) �Gas & Liquid Products Basic ControlLiquid Reactants (Jacket CTW) �Gas & Liquid Products OptimizationLiquid Reactants (Jacket CTW) �Liquid Product with Recycle Basic ControlLiquid Reactants (Jacket CTW) �Liquid Product with Recycle OptimizationJacket CTW�Outlet Temperature ControlJacket CTW�Inlet Temperature ControlJacket CTW with Steam Injection �Outlet Temperature ControlReactor Heat Exchanger �Recirc Temperature ControlReactor Heat Exchanger �Recirc Temperature Bypass ControlReactor Heat Exchanger �Recirc Temperature Bypass OptimizationJacket Steam & CTW Heat Exchanger �Inlet Temperature ControlJacket CTW Heat Exchanger�Inlet Temperature Bypass ControlJacket CTW Heat Exchanger �Inlet Temperature Bypass OptimizationReactor Inferential Measurements of�Heat Transfer Rate and ConversionBatch Reactor Concentration�Profile Slope ControlInnovative PID System to Optimize �Ethanol Yield and Carbon FootprintMammalian Bioreactor With�Enhanced DO and pH ControlMammalian Bioreactor With�Enhanced Substrate ControlTop Ten Reasons Why an Automation Engineer �Makes a Great Spouse or at Least a Wedding Gift Enhanced PID Algorithm �Originally Developed for WirelessBroadley-James Corporation �Wireless Bioreactor SetupSlide Number 34Gain 40 Reset 500Gain 40 Reset 5000Gain 40 Reset 10000Gain 40 Reset 15000Gain 80 Reset 15000Mammalian Bioreactor With�Viable Cell and Product Profile ControlSlide Number 41Slide Number 42Slide Number 43Slide Number 44Loop Block Diagram �(First Order Approximation)Open Loop Response of�Self-Regulating Process Open Loop Response of�Integrating Process Open Loop Response of�Runaway Process Slide Number 49Slide Number 50Slide Number 51Slide Number 52Loop Performance �Ultimate Limit �Slide Number 54Fastest Controller Tuning �(Reaction Curve Method*)Effect of Wireless Measurement�Update Time and Interval on PerformanceAdditional Deadtime from�Valve Stick-Slip, Resolution, or DeadbandNomenclature�(Process Dynamics & Performance)Nomenclature�(Wireless Dynamics & Performance)Key InsightsKey InsightsKey InsightsKey InsightsResource