Maximizing the Return on Your Control Investment Part 1 of 2 Meet the Experts Sessions
Jan 28, 2015
Maximizing the Return on Your
Control InvestmentPart 1 of 2
Meet the Experts Sessions
Presenters
James Beall
Terry Blevins
Session Objective
Provide a roadmap that may be used to maximize the return on your control investment
Agenda
Assessment of control loop utilization and automatic control performance,
Identifying areas where it is possible to justified the cost of improving control performance.
Tools and techniques to achieve best control performance. Identifying and correcting field problems.
Evaluating Control System Utilization
Product quality and manufacturing efficiency may be impacted by variation in key parameters.
When production is process limited, then throughput may be increase by reducing process variation and operating closer to the limit.
The control must operate as designed to achieve these benefits.
Control Utilization
Control utilization is an indicator that can be used to quickly determine if control and measurement problems exist within a control system. Surveys indicate that the primary reasons for control not being fully utilized fall into two areas:
Field measurement or control element
Process or control design
An immediate improvement in control utilization will be achieved by addressing these problems. To achieve full utilization, improved communications between maintenance and operations is important.
Example – Case StudyAt one pulp and paper plant a snapshot of the control utilization was collected to quantify the state of the process control. This survey showed:
Control Normal Loops Mode Utilization
Bleach Plant 78 60 76%
Power House 185 130 70%
Pulp Mill 174 116 66%
Paper Mill 236 134 56%
An instrumentation team was formed to investigate loops that were not running in their normal design mode. This team was responsible for making sure measurement, control valve, and process problems were addressed in a timely fashion. The reduction in variability led to significant improvements in plant throughput and product quality
Example – Another Case Study
At refinery and petrochemical complex a snapshot of the control utilization was collected to quantify the state of the process control. This survey showed:
System Loops Utilization PX 471 67.3% APS&VPS、CLE、Sulfur Recovery 469 59.7% Refinery 478 60.9% IGCC\Auxiliary Boiler 946 52.7% Ethylene 1355 77.5% FCCU 475 48% C4 164 68.9%
Examining Control Utilization
Summarizes performance for System or by Area, Cell, or Units
Abnormal Control Conditions indicated for Problem Loops:
– Control Service Status:• Not in Normal mode• Limited control output• Bad/Uncertain input
– Control Performance Status:• Standard Deviation• Variability Index• Oscillation Index• Tuning Index
DeltaV InSight Control Performance Reports
Roadmap to Improved Control
Control Loop
in Normal
Mode?
Starting
Point
Use DeltaV Insight to
Determine Control
Utilization
No
Yes
Reason?
Transmitter Broken/UnreliableFix or Replace
Transmitter
Poor
Performance
Investigate
Tuning
Valve/Actuator Problem
Changing Process Gain
Process Dynamics
Loop Interaction
Use DeltaV Insight to
examine loop tuning
DeltaV Insight – On-demand Tuning
Allows control loop tuning to be quickly established
Examine Tuning Impact
DeltaV InSight - Adaptive Tuning
InSight automatically calculates dynamic models from operator changes
Model Quality and Learning Status
Tuning criteria and desired speed of response
Tuning Recommendation
No Testing Required
Provides Tuning Based on Normal Operator Changes
Before vs. After
Tips for Using On-Demand Tuning
Set step size large enough to insure the response to the PID output changing is larger that the measurement noise and influence of process disturbances.
Select “Integrating” if the process is not self-regulating.
If the process has low or high gain, then selected Default Process and set the process type to match the expected behavior.
Manual Tuning (Not a DeltaV System)
When commissioning PID control associated with a self-regulating process, this procedure may be quickly applied to both old and new control systems to determine the tuning for PI control.
The size of the step should be just large enough to easily distinguish the resulting change in the controlled process output.
Manual Tuning - (Not a DeltaV System) Alternate
When commissioning PID control associated with a self-regulating process, this procedure may be quickly applied to both old and new control systems to determine the tuning for PI control.
The size of the step should be just large enough to easily distinguish the resulting change in the controlled process output.
Tune as follows:– Place the PV and OUT on a trend.– Place the controller in Manual and allow the process to reach steady sate.– Impose a step change in OUT and observe the response– Set the RESET equal to ¼ of the time it takes to “almost” line out (98% of final
value)– Set the GAIN equal to 1/3 of process gain, Kp where– Kp = ((Δ%PV) / (Δ%OUT) (Be sure to convert to % of span!)– Place the loop in automatic and make small adjustments to the Setpoint and
observe the response. Adjust ONLY the GAIN to achieve the desired response.
Impact of Sticky Valve When a control loop is
placed in automatic control, it is easy to detect if a valve or damper is not responding to the control system by observing the response of the controlled parameter to control system changes in the PID output.
Cycling can not be eliminated through tuning.
Setpoint (SP)
Controlled Parameter (PV)
Implied Valve Position (OUT)
Stem Position)
Value
Time
Valve Positioner Recommendation
The most common problems in commissioning a control system can often be traced to the fact that a positioner has not been provided with the valve, or the positioner provided with the valve has not been properly installed or has malfunctioned.
– The rule of thumb is that to achieve best control performance, all regulating valves should be equipped with a positioner.
– Without a positioner, the control performance that may be achieved is very limited when a valve is sticking – which is inherent in most valves.
– The cyclic behavior caused by a sticky valve (with no positioner) cannot be eliminated through tuning. Changes in tuning will only impact the period of the cycle that develops. The only way to eliminate this type of behavior is to install a valve positioner.
Installed Characteristics.
From a control perspective, it is highly desirable that the process gain be constant. If the process gain is constant, then the same proportional gain may be used over the entire operating range of the control loop.
If the valve characteristic was not been selected based on the process requirements, then the installed characteristic could be non-linear.
As illustrated in this example, the process gain varies from 0.5 to 4; that is, the process gain changes by a factor of eight.
7000
6000
5000
4000
3000
2000
0
Ga
s F
low
(S
CF
H)
0 20 40 60 80 100
Valve Position (%)
Impact of Non-linear Installed Characteristics
Non-linear Process Example Process gain and dynamics may change as a function of
operating conditions such as valve position or feed rate.
Linearizing Response
To compensate for the changes in process gain, a characterizer block may be installed between the PID and Analog Output blocks. Select the option to inverse the calculation done in the forward path for the PID back calculation to allow bumpless transfer.
Figure 12-8
100
90
80
70
60
50
40
30
20
10
0
Pro
cess
Out
put
(%
)
0 10 20 30 40 50 60 70 80 90 100
Valve Position (%)
Characterizer
IN OUT
0 0
5 1.5
10 3
15 5.5
20 8,5
25 10
30 11.5
35 13
40 17
45 20
50 22
55 26
60 31
65 36
70 42
75 48
80 54
85 63
90 74
95 86
100 100
Linear Relationship
Characterizer In-Out
Characterizer Setup The relationship between
the primary inputs and the output of the characterizer block may be defined by 21 x,y pairs over the final control element operating range.
Input values that fall between these points are automatically determined by the characterizer block using linear interpolation.
The curve defined by the characterizer points appears as the inverse of the plot of the final control element installed characteristic.
Identifying Model - DeltaV InSight
Last 200 Models automatically stored for each control loop in a model database
Various plot options to analyze impact of operating conditions on process models
Average of selected models may be utilized to establish the recommended tuning
Gain Process Insight with Model Analysis
Non-linear Process Models
DeltaV InSight – Adaptive Control
DeltaV InSight – Adaptive Control
Adaptive Control Enabled
Split Range Control
One of the most common ways of addressing multiple process inputs is known as split-range control.
The splitter block may be used to define a fixed relationship between the controller output and each manipulated process input – appearing as one valve to the PID block
The setup must account for the gain associated with each process input to achieve consistent control behavior.
Example – Steam Header Pressure
To allow the plant to continue operation if the turbine or generator fails and must be shut down, pressure reducing valves (PRVs) between the high pressure header and the lower pressure header may be adjusted to meet the lower pressure header steam demand and to maintain the header pressure constant.
This may be accomplished by using a splitter block in conjunction with a PID block to adjust the pressure reducing valves.
Steam Header – Splitter Characterization If the valve sizes or operating conditions for the valves are different, then it is
necessary to characterize the splitter to compensate for these differences. For example, if the flow ratings in thousand of pounds per hour, KPPH, of the
valves used in split-range control were as shown below:– Valve 1 flow rating = 50 KPPH– Valve 2 flow rating = 150 KPPH
Then the controller output range of adjustment associated with Valve 1 would be:
Interactive Loops
The fighting between interactive loops is most often addressed by simply detuning one of the control loops by reducing the proportional gain.
The valve associated with the detuned loop will change very slowly. Thus, the two loops will tend not to interact but at the expense of the detuned loop having slow response.
Mixing
ProcessAdditive
Valve(M2)
Feed
Valve(M1)
Composition
Controller
Flow
Controller
Outlet Flow
(C1)
Composition
(C2)
C2
M1
C1
M1
C2
M2
C1
M2
Roadmap to Improved Control
Normal
Mode?
Starting
Point
Determine
Control
Utilization
No
Yes
Reason?
Transmitter Broken/UnreliableFix or Replace
Transmitter
Poor
Performance
Investigate
Tuning
Valve/Actuator Problem
Changing Process Gain
Split range Setup incorrect
Process Dynamics
Loop Interaction
Variation on
Control
Monitor for
Change
Low/acceptable
High
Unacceptable
InvestigateChanging Process Gain
Process Disturbance
Process Dynamics
Loop Interaction
Changing Limit Condition
Utilize Process Capacity to Absorb Variability
Lambda
PV Back to SP in 6 x Lambda
Step change in load (inflow)
Controller Output changing outflowsmoothly!
PV
Setpoint
Inflow
Outflow
LIC
Outflow = inflow
Change in PV stopped
Utilize Process Capacity to Absorb Variability
Choose the arrest time “slow” enough to provide a variability sink yet maintain level within the allowable variation
Lambda = __2 * ALV___ Kp * MLD
– ALV = Allowable Level Variation
– Kp = Integrating process gain
– MLD = Maximum Load Disturbance (converted to % of level controller output scale)
Utilize Process Capacity to Absorb Variability
Level
Manipulated
Variable
Before
Level
Manipulated
Variable
After
Reducing Control Variation
When tuning is not sufficient to achieved the desired level of variation in critical control parameter or to maintain it at an operating limit, then multi-loop techniques may sometimes be applied to improve control. Three common multi-loop techniques are:
Feedforward Control Cascade Control Override Control
Summary
An on-line measurement of control utilization and variability is provided by DeltaV Insight.
Exploring the causes of poor utilization is the first step in resolving measurement, actuator or control issues.
When single loop control is not sufficient to achieve the desired level of control the multi-loops solutions should be explored.
Where To Get More Information
Many of the ideas discussed in this session are addressed “Control Loop Foundation – Batch and Continuous Processes”. Information on this book may be found at the book’s web site:
– http://controlloopfoundation.com/
Also, by going to this web site you can use your web browser to perform the 19 workshops that go with this book.