2018-2019 Process Dynamic and Control (Covenant University)

PROCESS DYNAMIC AND
CONTROL
CHE 512
2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

02/08/2019 1
UNIVERSITY)
Recommended Textbooks
• Process Dynamics and Control: 3rd Edition, 2011, by Dale E. Seborg,
Thomas F. Edgar, Duncan A. Mellichamp, Francis J. Doyle III. John Wiley
& Sons Inc
• Process Dynamics and Control: Modeling for Control and Prediction.
2006, By Brian Roffel and Ben Betlem. John Wiley & Sons Ltd
• Chemical Process Control by George Stephanopoulos – Prentice Hall
International
• Instrumentation, measurement and feedback – Barry E Jones, McGraw Hill
Book Co. UK
• Process Control – Thomas E Marlin (McGraw Hill Book Co. UK)
• Advanced Engineering Mathematics – Erwin Kreyszig, by John Wiley &
Sons, New York

02/08/2019 2
UNIVERSITY)
Introduction to Process Control
• The continuous change of measurements in a chemical or biological
process leads to the conclusion that processes are dynamic.
• Process dynamics refer to an unsteady-state or transient behavior.
• Steady-state vs. unsteady-state behavior
Steady state: variables do not change with time
• Only with an understanding of transient behavior of physical systems can
an engineer design good processes.
• This is exactly what process control does: it provides the expertise needed
to design plants that function well in a dynamic environment.
• Process control is the regulation / manipulation of variables influencing
the conduct of a process in order to obtain a product of desired quality and
quantity in an efficient and economic manner.

02/08/2019 3
UNIVERSITY)
OBJECTIVES OF CONTROL:
A control system is required to perform either one or both task:
1. Maintain the process at the operational conditions and set points: Many
processes should work at steady state conditions or in a state in which it
satisfies all the benefits for a company such as budget, yield, safety, and
other quality objectives. In many real-life situations, a process may not
always remain static under these conditions and therefore can cause
substantial losses to the process.
2. Transition of the process from one operational condition to another:
• In real-life situations, engineers may change the process operational
conditions for a variety of different reasons, such as customer specifications
or environment specifications. Although, transitioning a process from one
operational condition to another can be detrimental to a process, it also can
be beneficial depending on the company and consumer demands.
02/08/2019 4
UNIVERSITY)
Common features in process control cases
• There is always a specific value (or range) as a desired value (referred to as
set point) for the controlled variable.
• The conditions of the system are measured; that is, all control systems use
sensors to measure the physical variables that are to be maintained near the
desired values.
• There is always a control calculation, or algorithm , which uses the
measured and desired values to determine the correction to the process
operation.
• The results of this calculation are implemented by adjusting some item of
equipment in the system, which is termed the final control element.

02/08/2019 5
UNIVERSITY)
• Input: input does not necessarily refer to material moving into the system.
In Process Control, input denotes the effect of the surroundings on the
chemical or biochemical process.
• Output: denotes the effect of the process on the surroundings.
• Controlled variable: it is the variable that needs to be maintained or
controlled at some desired value or range. Sometimes also referred to as
process variable.
• Set Point: it is the desired value of the controlled variable. Thus the job of a
control system is to maintain the controlled variable at its set point.
• Manipulated variable: is the variable used to maintain the controlled
variable at its set point.
• Disturbance: any variable that causes the controlled variable to deviate from
its set point. Also referred to as upset.

02/08/2019 6
UNIVERSITY)
Examples of why a process may be moved from
one operational set point to another
• 1. Economics
• 2. Product specifications
• 3. Operational constraints
• 4. Environmental regulations
• 5. Consumer/Customer specifications
• 6. Safety precautions

02/08/2019 7
UNIVERSITY)
Blending system (Example of how to control a
process)
• Notation:
• w1, w2 and w are mass flow rates
• x1, x2 and x are mass fractions of component A

02/08/2019 8
UNIVERSITY)
• Assumptions:
1. w1 is constant
2. x2 = constant = 1 (stream 2 is pure A)
3. Perfect mixing in the tank
• Control Objective:
• Keep x at a desired value (or “set point”) xsp, despite variations in x1(t). Flow rate
w2 can be adjusted for this purpose.
• Terminology:
• Controlled variable (or “output variable”): x
• Manipulated variable (or “input variable”): w2
• Disturbance variable (or “load variable”): x1

02/08/2019 9
UNIVERSITY)
x  xSP ?
Overall balance:
0  w1  w2  w (1-1)
Component A balance:
w1x1  w2 x2  wx  0 (1-2)
(The overbars denote nominal steady-state design values.)

• At the design conditions, x  xSP. Substitute Eq. 1-2, x  xSP and
x2  1 , then solve Eq. 1-2 for w2 :
xSP  x1
w2  w1 (1-3)
1  xSP
02/08/2019 10
UNIVERSITY)
• Equation 1-3 is the design equation for the blending system.
• If our assumptions are correct, then this value of w2 will keep x at xSP . But
what if conditions change?
• Control Question. Suppose that the inlet concentration x1 changes with time.
How can we ensure that x remains at or near the set point xSP ?
• As a specific example, if x1  x1 and w2  w2 , then x > xSP.
• Some Possible Control Strategies:
• Method 1. Measure x and adjust w2.
• Intuitively, if x is too high, we should reduce w2;

02/08/2019 11
UNIVERSITY)
CONTROL STRATEGIES
The objective of a control system is to keep the controlled variables at their
desired values (or set points). This is achieved by manipulating the
manipulated variables using a control algorithm.
CONTROL ALGORITHM
OPEN- LOOP SYSTEM (CONTROL)
An open –loop system is a system with only the input and the output but no
regulation / no control action at all.

02/08/2019 12
UNIVERSITY)
• Example of an open –loop system
U D In an open-loop system, the

control mechanism act without
the current information about the
Y status of the process.
Signal path is represented by
• Block diagram representation arrows, which show the direction
D
of information flow.
U(t) PROCESS Y(t)
Figure: A simplified open loop control system

02/08/2019 13
UNIVERSITY)
Feedforward Control (FFC) Algorithm
Feedforward control: a feedforward controller detects the disturbance
directly and takes an appropriate control action in order to eliminate its
effect on the process output.
• The disturbance can be measured

• We know how the disturbances affect the output
• We know how the control signal affects the output

02/08/2019 14
UNIVERSITY)
Feedforward Control Algorithm/Strategy
FF CONTROLLER
Disturbance (d)
CONTROL
EQUATION 
SP
PROCESS Y(t)
U(t)
Fig:B

02/08/2019 15
UNIVERSITY)
Block diagram representation of Fig: B
Disturbance (d)
MEASUREMENT
SP U(t)
CONTROLLER PROCESS Y(t)
Fig C. Typical Feedforward control systems

02/08/2019 16
UNIVERSITY)
Advantages and disadvantages of FF control loop
ADVANTAGES
i. Acts before the effect of a disturbance has been felt by the system.
ii.Is good for slow systems
iii.It does not introduce instability in the closed – loop response
DISADVANTAGES
i. Requires identification of all possible disturbances and their measurement.
ii.Cannot cope with unmeasured disturbance.
iii.Sensitive to process parameter variation.

02/08/2019 17
UNIVERSITY)
Example
• Consider the example of the stirred tank heater. The objective is to control
the temperature. The disturbance source is Ti.
T
C
I/
P
Feedforward control for a heated tank

02/08/2019 18
UNIVERSITY)
CLOSED – LOOP SYSTEM
A closed –loop system is a system with a total control action /

regulation around the entire process.
The total control action is called the feedback control. The controlled
variable is measured, and the measurement is fed to the controller,
thus the controller receiving information about how a control action
affects the output.

02/08/2019 19
UNIVERSITY)
Feedback Control Loop (FBC)
• Feedback control, A feedback loop measures a process variable and
sends the measurement to a controller for comparison to setpoint. If
the process variable is not at setpoint, control action is taken to return
the process variable to setpoint.
The basic elements of a feedback controller
• The process variable (PV). I.e. the variable that is to be maintained
under control.
• The set point (SP) which is the desired value of the process variable.
• The error (e), which is a measure of the difference between the PV and
the SP.
• The controller, whose control law and turning drive the corrective
action .

02/08/2019 20
UNIVERSITY)
• The final control element ( typically a valve)
• The manipulated variables (MV), these input variables are adjusted
dynamically to keep the controlled variables at their set-points.
Two types of feedback control loop
1. Negative feedback, this refers to the desirable situation where the

corrective action taken by the controller forces the controlled variable
towards the set point.
Error = reference – feedback signal
2 Positive feedback occurs when the controller forces the controlled
variables farther away from the set point.
Error = reference + feedback signal
02/08/2019 21
UNIVERSITY)
Advantages and disadvantages of FB control loop
ADVANTAGES
I. Does not require identification of all possible disturbances and their

measurements.
II. It is insensitive to modelling error.
III. It is insensitive to parameter changes.
DISADVANTAGES
I. It waits until the effect of disturbance is felt by the system.

II. It is unsatisfactory for slow system.
III. It may create instability in the closed – loop

02/08/2019 22
UNIVERSITY)
Example 1:
When in the shower.
Process variables
• The disturbances = Water from toilet flushing

• Manipulated variables = Temperature and the flow rate of the water
• Measured / controlled variables = Temperature of the water

02/08/2019 23
UNIVERSITY)
• Block diagram representation
Disturbance =
FINAL CONTROL flushing
ELEMENT Water
PROCESS temperature
Fw , T
MEASUREMENT
CONTROLLER
Set point

02/08/2019 24
UNIVERSITY)
Assignment
• Consider the example of the stirred tank heater. The objective is to
control the temperature. The disturbance source is Ti. Draw a feedback
control loop around this process

02/08/2019 25
UNIVERSITY)
• REVISIT THE BLENDING PROCESS: A CONBINATION OF
FEEDBACK – FEEDFORWARD CONTROL LOOPS
• These can be combine in several ways
AT AC I/P
Pure A I/P
W2
Mixture (A,B) X2=1
W1 AC
X1
AT
W (1) First way

X xsp

02/08/2019 26
UNIVERSITY)
A CONBINATION OF FEEDBACK – FEEDFORWARD
CONTROL LOOPS CONT
FFC
AT
x1 x2
w2
w1
AC sp
AT
(2) Second way w

02/08/2019 27
UNIVERSITY)
Analysis of Feedback control
• We introduce feedback control systems again by considering the stirred-tank
blending process discussed earlier.
• A schematic diagram of a stirred-tank blending process is shown in Fig. 8.1.

02/08/2019 28
UNIVERSITY)
• The control objective is to keep the tank exit composition x at the desired
value set point by adjusting w2, the flow rate of species A, via the control
valve. The composition analyzer-transmitter (AT) measures the exit
composition and transmits it as an electronic signal to the feedback
controller (AC). The controller compares the measured value Xm to the
desired value (set point) and calculates an appropriate output signal p, an
electronic signal that is sent to a current-to-pressure transducer (I/P) where
it is converted to an equivalent pneumatic (air) signal that is compatible with
the control valve.

02/08/2019 29
UNIVERSITY)
The example illustrates that the basic components in a feedback control loop
are:
• Process being controlled (blending system)
• Sensor-transmitter combination (AT)
• Feedback controller (AC)
• Current-to-pressure transducer (I/P)
• Final control element (control valve)
• Transmission lines between the various instruments (electrical cables and
pneumatic tubing)

02/08/2019 30
UNIVERSITY)
• A current-to-pressure (or voltage-to-pressure) transducer is required if the
control loop contains both electronic instruments and a pneumatic control
valve.
• The term final control element refers to the device that is used to adjust the
manipulated variable. It is usually a control valve but could be some other
type of device, such as a variable speed pump or an electrical heater.
• Most new control systems utilize digital technology with the control
algorithms implemented via digital computers and with digital signal
pathways (networks) used for data transmission. Consequently, we consider
digital control algorithms.

02/08/2019 31
UNIVERSITY)
• As a simple example of feedback control, consider the flow control loop in
Fig. 8.2 where the flow rate of a process stream is measured and
transmitted electronically to a flow controller. The controller compares the
measured value to the set point and takes the appropriate corrective action
by calculating the controller output and transmitting it as an electronic
signal to the control valve.
• The block diagram for the feedback controller of Fig. 8.2 is shown in Fig.
8.3.

02/08/2019 32
UNIVERSITY)
• The set point is shown as a dashed line. For digital control systems, the set
point would be entered by an operator using a computer terminal.
• For an analog controller, the set point would be specified via a dial setting
on the equipment.
• For digital control systems, the input signals are first converted from
analog to digital form prior to the control calculations. Then, the calculated
value of the controller output is converted from a digital signal to an
analog signal for transmission to the control valve (or some other type of
final control element).

02/08/2019 33
UNIVERSITY)
TYPES OF FEEDBACK CONTROLLER
(1) Proportional Control
• In feedback control, the objective is to reduce the error signal to zero where
e(t) = ysp(t) - ym(t) (8.1)
e(t) = error signal
Ysp(t) = set point
Ym(t) = measured value of the controlled variable (or equivalent signal from
the sensor/transmitter)
• For proportional control, the controller output is proportional to the error
signal,

02/08/2019 34
UNIVERSITY)
• The key concepts behind proportional control are that
(1) the controller gain can be adjusted to make the controller output changes
as sensitive as desired to deviations between set point and controlled
variable.
(2) the sign of Kc can be chosen to make the controller output increase (or
decrease) as the error signal increases. For example, for the blending process
in Fig. 8.1, we want w2 to decrease as x increases; hence, Kc should be a
positive number.
02/08/2019 35
UNIVERSITY)
• For proportional controllers, bias can be adjusted, a procedure referred to
as manual reset. Because the controller output equals when the error is
zero, is adjusted so that the controller output, and consequently the
manipulated variable, are at their nominal steady-state values when the
error is zero.
• For example, if the final control element is a control valve, is adjusted so
that the flow rate through the control valve is equal to the nominal, steady-
state value when e = 0.
• The controller gain Kc is adjustable and is usually tuned (i.e., adjusted)
after the controller has been installed.
• For digital implementation, p and e are often expressed as numbers
between 0 and 100%. The latter representation is especially convenient for
graphical displays using computer control software.
02/08/2019 36
UNIVERSITY)
• In analyzing control systems it can be more convenient to express the error
signal in engineering units such as ⁰C or mol/L. For these situations, Kc
will not be dimensionless. As an example, consider the stirred-tank
blending system. Suppose that e [ =] mass fraction and p [ =] mA; then Eq.
8.2 implies that Kc [ =] mA because mass fraction is a dimensionless
quantity. If a controller gain is not dimensionless, it includes the steady-
state gain for another component of the control loop such as a transmitter
or control valve.
• Some controllers have a proportional band setting instead of a controller
gain. The proportional band PB (in %) is defined as:

02/08/2019 37
UNIVERSITY)
• This definition applies only if Kc is dimensionless. Note that a small
(narrow) proportional band corresponds to a large controller gain, whereas
a large (wide) PB value implies a small value of Kc.
• The ideal proportional controller in Eq. 8-2 and Fig. 8.4 does not include
physical limits on the controller output, p. A more realistic representation
is shown in Fig. 8.5, where the controller saturates when its output reaches
a physical limit, either Pmax or Pmin· In order to derive the transfer function
for an ideal proportional controller (without saturation limits), define a
deviation variable p'(t) as

02/08/2019 38
UNIVERSITY)
• Then Eq. 8-2 can be written as

02/08/2019 39
UNIVERSITY)
• Taking Laplace transforms and rearranging (8-5) gives the transfer function
for proportional-only control:
• An inherent disadvantage of proportional-only control is that a steady-state

error (or offset) occurs after a set-point change or a sustained disturbance.
• Fortunately, the addition of the integral control mode facilitates offset
elimination
(2) Integral Control
• For integral control action, the controller output depends on the integral of
the error signal over time,

02/08/2019 40
UNIVERSITY)
• where τI, an adjustable parameter referred to as the integral time or reset
time, has units of time. In the past, integral control action has been referred
to as reset or floating control, but these terms are seldom used anymore.
• Integral control action is widely used because it provides an important
practical advantage, the elimination of offset.
• In order for the controlled process to be at steady state, the controller
output p must be constant so that the manipulated variable is also constant.
• Equation 8-7 implies that p changes with time unless e(t*) = 0. Thus, when
integral action is used, p automatically changes until it attains the value
required to make the steady-state error zero.
• Although elimination of offset is usually an important control objective,
the integral controller in Eq. 8-7 is seldom used by itself, because little
control action takes place until the error signal has persisted for some time.
02/08/2019 41
UNIVERSITY)
• In contrast, proportional control action takes immediate corrective action
as soon as an error is detected. Consequently, integral control action is
normally used in conjunction with proportional control as the proportional-
integral (PI) controller:
• The corresponding transfer function for the PI controller in Eq. 8-8 is given
by
• The response of the PI controller to a unit step change in e(t) is shown in

Fig. 8.6. At time zero, the controller output changes instantaneously due to
the proportional action.

02/08/2019 42
UNIVERSITY)
• Integral action causes the ramp increase in p(t) for t > 0. When t = τI, the
integral term has contributed the same amount to the controller output as the
proportional term. Thus, the integral action has repeated the proportional
action once.
• One disadvantage of using integral action is that it tends to produce
oscillatory responses of the controlled variable.
02/08/2019 43
UNIVERSITY)
• it reduces the stability of the feedback control system. A limited amount of
oscillation can usually be tolerated, because it often is associated with a
faster response.
(3) Derivative Control
• The function of derivative control action is to anticipate the future behavior
of the error signal by considering its rate of change. The derivative mode is
given by:
• where τD, the derivative time, has units of time. Note that the controller
output is equal to the nominal value as long as the error is constant (that
is, as long as de/dt = 0).

02/08/2019 44
UNIVERSITY)
• Consequently, derivative action is never used alone; it is always used in
conjunction with proportional or proportional-integral control. For
example, an ideal PD controller has the transfer function
• Derivative control action also tends to improve the dynamic response of the
controlled variable by the settling time, the time it takes reducing to reach
steady state.
• Unfortunately, the ideal proportional-derivative control algorithm in Eq. 8-
11 is physically unrealizable because it cannot be implemented exactly
using either analog or digital controllers. For analog controllers, the transfer
function in (8-11) can be approximated by

02/08/2019 45
UNIVERSITY)
• where the constant a typically has a value between 0.05 and 0.2, with 0.1
being a common choice. In Eq. 8-12 the denominator term serves as a
derivative mode filter (or a derivative filter) that reduces the sensitivity of
the control calculations to noisy measurements.
Proportional-Integral-Derivative Control
• Elimination of Derivative and Proportional Kick
• Most complicated to tune (Kc, I, D) .
• Better performance than PI
• No offset
• Derivative action may be affected by noise

02/08/2019 46
UNIVERSITY)
Parallel Form of PID Control
• The parallel form of the PID control algorithm (without a derivative filter)
is given by
• The corresponding transfer function is
• Figure 8.8 illustrates that this controller can be viewed as three separate
elements operating in parallel on E(s).

02/08/2019 47
UNIVERSITY)
02/08/2019 48
UNIVERSITY)
Series Form of PID Control
• The series form of PID control without a derivative filter is shown in Fig.
8.9. Commercial versions of the series-form controller have a derivative
filter that is applied to either the derivative term, as in Eq. 8-12, or to the
PD term, as in Eq. 8-15:
• The parallel-form PID controller with and without a derivative filter are
shown in Table 8.1.

02/08/2019 49
UNIVERSITY)
02/08/2019 50
UNIVERSITY)
Expanded Form of PID Control
• The expanded form of PID control is:
• The controller parameters for the expanded form are three "gains," Kc, KI
and KD, rather than the standard parameters, Kc, τI, and τD· The expanded
form of PID control is used in MATLAB. This form might appear to be
well suited for controller tuning, because each gain independently adjust the
influences only one control mode.
• Equations 8-2 through 8-16 describe how controllers perform during the
automatic mode of operation.
• However, in certain situations the plant operator may decide to override the
automatic mode and adjust the controller output manually.

02/08/2019 51
UNIVERSITY)
Elimination of Derivative and Proportional Kick
• One disadvantage of the previous PID controllers is that a sudden change
in set point (and hence the error, e) will cause the derivative term
momentarily to become very large and thus provide a derivative kick to the
final control element.
• To illustrate the elimination of derivative kick, consider the parallel form
of PID control in Eq. 8-13. Replacing de/dt by –dym/dt gives
• This method can be implemented quite easily by placing the PD element in

the feedback path, as shown in Fig. 8.10.

02/08/2019 52
UNIVERSITY)
• A more flexible PID control algorithm can be obtained by weighting the
set point in both the proportional and the derivative terms. This
modification eliminates the proportional kick that also occurs after a step
change in set point. For this modified PID algorithm, a different error term
is defined for each control mode:

02/08/2019 53
UNIVERSITY)
• where β and γ are nonnegative constants. This control algorithm is
known as the parallel PID controller with proportional and derivative
mode weighting, or the beta gamma controller.
• To eliminate derivative kick, γ is set to zero; to eliminate proportional
kick, β is set to zero.

02/08/2019 54
UNIVERSITY)
ON-OFF CONTROLLERS
• On-off controllers are simple, inexpensive feedback controllers that are
commonly used as thermostats in home heating systems and domestic
refrigerators.
• They simply turn an output device ON or OFF depending on the value of
the e. Most home thermostats are ON/OFF controllers. In the heating
mode, a thermostat will turn the furnace ON when the error is negative
and OFF when the error is positive.
• They are also used in noncritical industrial applications such as some level
control loops and heating systems.
• However, on-off controllers are less widely used than PID controllers,
because they are not as versatile or as effective.
• For ideal on-off control, the controller output has only two possible values:

02/08/2019 55
UNIVERSITY)
• where pmax and pmin denote the on and off values, respectively (for
example, for a typical digital computer implementation, pmax = 100% and
pmin = 0%; for a current-based electronic controller, pmax = 20 mA and
pmin = 4 mA).
• The disadvantages of on-off control are that it results in continual cycling
of the controlled variable and produces excessive wear on the control
valve (or other final control element).

02/08/2019 56
UNIVERSITY)
TRANSMISSION
• Electronic controllers (digital or analog) can be located relatively far from
their instruments with little concern for the impedance of the intervening
transmission lines or for the time of transmission, which for all practical
purposes is instantaneous.
• Most transmitter analog signals are in the form of current rather than
voltage, because voltage is affected by wire and connector resistances,
which change with wire length, temperature and aging.
• Pneumatic pressure signals between instruments are transmitted by means
of tubing, usually 1/4- or 3/8 inch diameter.
• Signals from digital instruments and controllers are usually transmitted in
digital format as a sequence of on-off pulses.

02/08/2019 57
UNIVERSITY)
• The gain of the measurement element Km for any linear instrument is:
• Many sensor-transmitters have overdamped dynamics and exhibit

monotonic responses to a step change in the variable being measured.
Thus, it is reasonable to model this type of measurement dynamics as a
first order transfer function between the actual value y and the measured
value ym:
• where Km is the gain given by Eq. 9-1 and τm is the measurement time
constant. For the temperature transmitter example, the units of Km are
mA/⁰C.

02/08/2019 58
UNIVERSITY)
Final control elements
• Every process control loop contains a final control element (or actuator),
the device that enables a process variable to be manipulated.
• For most chemical and petroleum processes, the final control elements
(usually control valves) adjust the flow rates of materials-solid, liquid, and
gas feeds and products-and, indirectly, the rates of energy transfer to and
from the process.

02/08/2019 59
UNIVERSITY)
• Pneumatic control valves are to be specified for the applications listed
below. State whether an A-0 or A-C valve should be specified for the
following manipulated variables, and give reason(s):
(a) Steam pressure in a reactor heating coil.
(b) Flow rate of reactants into a polymerization reactor.
(c) Flow of effluent from a wastewater treatment holding tank into a river.
(d) Flow of cooling water to a distillation condenser.
Solution
(a) A-O (fail-close) to make sure that a transmitter failure will not cause the
reactor to overheat, which is usually more serious than having it operate at
too low a temperature.
02/08/2019 60
UNIVERSITY)
(b) A-O (fail-close) to prevent the reactor from being flooded with
excessive reactants.
(c) A-O (fail-close) to prevent excessive and perhaps untreated waste
from entering the stream.
(d) A-C (fail-open) to ensure that overhead vapor is completely
condensed before it reaches the receiver.

02/08/2019 61
UNIVERSITY)
White Box versus Black Box Models
• White box models are based on physical and chemical laws of
conservation, such as mass balance, component balance, momentum
balance and energy balance. They are also called first principles or
mechanistic models. The models give physical insight into the process and
explain the process behavior in terms of state variables and measured
variables.
• The state variable of the model is the variable whose rate of change is
described by the conservation balance. These models can already be
developed when the process does not yet exist.
• The dynamic equations are supplemented with algebraic equations
describing heat and mass transfer, kinetics, etc. Developing these models is
very time consuming.
02/08/2019 62
UNIVERSITY)
• Black box models or empirical models do not describe the physical
phenomena of the process, they are based on input/output data and only
describe the relationship between the measured input and output data of
the process. These models are useful when limited time is available for
model development and/or when there is insufficient physical
understanding of the process. Mathematical representations include time
series models (such as ARMA, ARX, Box and Jenkins models, recurrent
neural network models, recurrent fuzzy models).
• It could be that much physical insight is available, but that certain
information or understanding is lacking. In those cases, physical models
could be combined with black box models; the resulting models are called
gray box or hybrid models.
• Examples are physical models combined with neural network or fuzzy
logic models
02/08/2019 63
UNIVERSITY)
Simulation of Dynamic models
• A dynamic model of the process allows alternative control strategies to
be evaluated. For example, a dynamic model can help identify the
process variables that should be controlled and those that should be
manipulated. For model-based control Strategies, the process model is
part of the control law.
• A dynamic model can be used to characterize the transient behavior of
a process for a wide variety of conditions.

02/08/2019 64
UNIVERSITY)
General principles for the development of dynamic models
• It is important to remember that a process model is nothing more than a
mathematical abstraction of a real process.
• The model equations are at best an approximation to the real process as
expressed by the adage that "all models are wrong, but some are useful."
Consequently, the model cannot incorporate all of the features, whether
macroscopic or microscopic, of the real process.
• The model should incorporate all of the important dynamic behavior
while being no more complex than is necessary. Thus, less important
phenomena are omitted in order to keep the number of model equations,
variables, and parameters at reasonable levels.

02/08/2019 65
UNIVERSITY)
• The failure to choose an appropriate set of simplifying assumptions
invariably leads to either
• (1) rigorous but excessively complicated models or
• (2) overly simplistic models. Both extremes should be avoided.
• A systematic procedure for developing dynamic models from first
principles is summarized in Table 2.1.
• The degrees of freedom analysis in Step 7 is required in model
development for complex processes. Because these models typically
contain large numbers of variables and equations, it is not obvious whether
the model can be solved, or whether it has a unique solution.
• Dynamic models of chemical processes consist of ordinary differential
equations (ODE) and/or partial differential equations (PDE), plus related
algebraic equations.
02/08/2019 66
UNIVERSITY)
02/08/2019 67
UNIVERSITY)
Conservation Laws
• Theoretical models of chemical processes are based on conservation laws
such as the conservation of mass and energy. Consequently, we now
consider important conservation laws and use them to develop dynamic
models for representative processes.
• Conservation of Mass
• Conservation of Component i

02/08/2019 68
UNIVERSITY)
• The last term on the right-hand side of (2-7) represents the rate of
generation (or consumption) of component I as a result of chemical
reactions. Conservation equations can also be written in terms of molar
quantities, atomic species, and molecular species.
• Conservation of Energy
• The general law of energy conservation is also called the First Law of
Thermodynamics. It can be expressed as

02/08/2019 69
UNIVERSITY)
• The total energy of a thermodynamic system, Utot, is the sum of its internal
energy, kinetic energy, and potential energy:
• For the processes and examples considered here, it is appropriate to make

two assumptions:
• 1. Changes in potential energy and kinetic energy can be neglected,
because they are small in comparison with changes in internal energy.
• 2. The net rate of work can be neglected, because it is small compared to
the rates of heat transfer and convection.
• For these assumptions, the energy balance in Eq. 2-8 can be written as

02/08/2019 70
UNIVERSITY)
• where Uint is the internal energy of the system, is the enthalpy per unit
mass, w is the mass flow rate, and Q is the rate of heat transfer to the
system. The a operator denotes the difference between outlet conditions
and inlet conditions of the flowing streams. Consequently, the term
represents the enthalpy of the inlet stream(s) minus the enthalpy of the
outlet stream(s).
• The analogous equation for molar quantities is
• where Is the enthalpy per mole and w is the molar flow rate. Note that
the conservation laws of this section are valid for batch and semi-batch
processes, as well as for continuous processes.

02/08/2019 71
UNIVERSITY)
• For example, in batch processes, there are no inlet and outlet flow rates.
Thus, w = 0 and = 0 in (2-10) and (2-11).
• In order to derive dynamic models of processes from the general energy
balances in Eqs. 2-10 and 2-11, expressions for Uint and or are
required, which can be derived from thermodynamics.

02/08/2019 72
UNIVERSITY)
The Blending Process Revisited
• The dynamic model of the blending process in Eqs. 2-2 and 2-3 can be
simplified and expressed in a more appropriate form for computer
simulation.
• For this analysis, we introduce the additional assumption that the density
of the liquid, ρ, is a constant.

02/08/2019 73
UNIVERSITY)
• Rate of mass accumulation is given as
• Where w1, w2 and w are the mass flowrate

• For the perfect mixing assumption, the rate of accumulation of component
A is d(vρx)/dt, where x is the mass fraction of A. The unsteady state
component balance is
• Eqs. 2-2 and 2-3 provide an unsteady state model for the blending system.
• The corresponding steady state model It also can be obtained by
setting the accumulation terms in Eqs. 2-2 and 2-3 equal to zero,

02/08/2019 74
UNIVERSITY)
• where the nominal steady-state conditions are denoted by
and so on. In general, a steady-state model is a special case of an
unsteady-state model that can be derived by setting accumulation terms
equal to zero.
• A dynamic model can be used to characterize the transient behavior of
a process for a wide variety of conditions.
• The dynamic model of the blending process in Eqs. 2-2 and 2-3 can be
simplified and expressed in a more appropriate form for computer
simulation. For this analysis, we introduce the additional assumption
that the density of the liquid, ρ, is a constant.
• This assumption is reasonable because often the density has only a weak
dependence on composition. For constant ρ, Eqs. 2-2 and 2-3 become

02/08/2019 75
UNIVERSITY)
• Equation 2-13 can be simplified by expanding the accumulation term using
the "chain rule" for differentiation of a product:
• Substitution of (2-14) into (2-13) gives
• Substitution of the mass balance in (2-12) for pdV/dt in (2-15) gives

02/08/2019 76
UNIVERSITY)
• After cancelling common terms and rearranging (2-12) and (2-16), a more
convenient model form is obtained:
• The dynamic model in Eqs. 2-17 and 2-18 is quite general and is based on
only two assumptions: perfect mixing and constant density. For special
situations, the liquid volume Vis constant (that is, dV/dt = 0), and the exit
flow rate equals the sum of the inlet flow rates, w = w1 + w2. For example,
these conditions occur when
1. An overflow line is used in the tank as shown in Fig. 1.3.
2. The tank is closed and filled to capacity.
3. A liquid-level controller keeps V essentially constant by adjusting a flow
rate.
02/08/2019 77
UNIVERSITY)
• In all three cases, Eq. 2-17 reduces to the same form as Eq. 2-4, not
because each flow rate is constant, but because w = w1 + w2 at all times.
• The dynamic model in Eqs. 2-17 and 2-18 is in a convenient form for
subsequent investigation based on analytical or numerical techniques.
• In order to obtain a solution to the ODE model, we must specify the inlet
compositions (x1 and x2) and the flow rates (w1, w2 and w) as functions of
time.
• After specifying initial conditions for the dependent variables, V(0) and
x(0), we can determine the transient responses, V(t) and x(t). The derivation
of an analytical expression for x(t) when V is constant is illustrated in
Example 2.1.

02/08/2019 78
UNIVERSITY)
02/08/2019 79
UNIVERSITY)
• EXAMPLE2.1
• A stirred-tank blending process with a constant liquid holdup of 2 m3 is
used to blend two streams whose densities are both approximately 900
kg/m3. The density does not change during mixing.
(a) Assume that the process has been operating for a long period of time
with flow rates of w1 = 500 kg/min and w2 = 200 kg/min, and feed
compositions (mass fractions) of x1 = 0.4 and x2 = 0.75. What is the steady-
state value of x?
(b) Suppose that w1 changes suddenly from 500 to 400 kg/min and remains
at the new value. Determine an expression for x(t) and plot it.
(c) Repeat part (b) for the case where w2 (instead of w1) changes suddenly
from 200 to 100 kg/min and remains there.
(d) Repeat part (c) for the case where x1 suddenly changes from 0.4 to 0.6.
(e) For parts (b) through (d), plot the normalized response xN(t),
02/08/2019 80
UNIVERSITY)
• SOLUTION
(a) Denote the initial steady-state conditions by: and so on. For the
initial steady state, Eqs. 2-4 and 2-5 are applicable. Solve (2-5) for
• (b) The component balance in Eq. 2-3 can be rearranged (for constant V
and ρ) as
• where In each of the three parts, (b)-(d), τ = 3 min and the right side
of (2-19) is constant for this example. Thus, (2-19) can be written as

02/08/2019 81
UNIVERSITY)
• where
• The solution to (2-20) can be obtained by applying standard solution

methods
• For case (b):
• Substituting C* into (2-22) gives the desired solution for the step change
in w1

02/08/2019 82
UNIVERSITY)
(c) For the step change in w2,
• and the solution is
(d) Similarly, for the simultaneous changes in x1 and w2, Eq. 2-21 gives C*
= 0.625. Thus, the solution is
(e) The individual responses in (2-22)-(2-24) have the same normalized

response:

02/08/2019 83
UNIVERSITY)
The responses of (b)-( e) are shown in Fig. 2.2.

02/08/2019 84
UNIVERSITY)
02/08/2019 85
UNIVERSITY)
Degree of Freedom
• The number of degrees of freedom, NF can be calculated from the
expression
• where Nv is the total number of process variables and NE is the number of

independent equations. A degrees of freedom analysis allows modeling
problems to be classified according to the following categories:
1. NF = 0: The process model is exactly specified. If NF = 0, then the number
of equations is equal to the number of process variables and the set of
equations has a solution. (However, the solution may not be unique for a set
of nonlinear equations.)

02/08/2019 86
UNIVERSITY)
2. NF > 0: The process is underspecified. If NF > 0, then Nv > N £,so there
are more process variables than equations. Consequently, the NE equations
have an infinite number of solutions, because NF process variables can be
specified arbitrarily.
3. NF < 0: The process model is over specified. For NF < 0, there are fewer
process variables than equations, and consequently the set of equations has
no solution.
• Note that NF = 0 is the only satisfactory case. If NF > 0, then a sufficient
number of input variables have not been assigned numerical values. Then
additional independent model equations must be developed in order for the
model to have an exact solution.

02/08/2019 87
UNIVERSITY)
Modelling of Dynamic and static behaviour of chemical
process systems
• In a static model, the most recent output or dependent variables depend on
the most recent values of the input or independent variables. In dynamic
models, the state variables describe the change in dependent variable as a
function of independent variables and time, the response is called the
system transient.
• In chemical engineering, static models are, among others, used in
optimization and process design, whereas dynamic models find their
application in process control and prediction of the values of future process
variables.

02/08/2019 88
UNIVERSITY)
Dynamic vs. Steady-state (Static)
Steady-state (Static)
• Variables not a function of time
• useful for design calculation
Dynamic
• Variables are a function of time
• Control requires dynamic model

02/08/2019 89
UNIVERSITY)
State variables and state equations for a chemical process
• A state variable is one of the set of variables that are used to describe the
mathematical "state" of a dynamical system
• The state equations describe the behavior of the process with time.
Selection of Independent Model Variables
• The first thing to do is to make a distinction between process input
variables and state variables. This also determines the hierarchy of the
variables in the model. Figure 19.2 show the hierarchy in the modeling
exercise

02/08/2019 90
UNIVERSITY)
• The process input variables (usually flows) affect the process state
variables (usually pressure, level, temperature, concentration) which in
turn have an impact on the measured process output such as a quality (for
example viscosity).
• Furthermore, the state variables are usually not mutually independent, for
example temperature, pressure and composition. This means that a model
that includes the process inputs as well as the state variables as additional
inputs, models the effect of the true process inputs multiple times.
Therefore a proper selection of the inputs and output(s) of the model is
very important.
02/08/2019 91
UNIVERSITY)
• To describe a process system we need a set of variables that characterize
the system and a set of relationships that describe how these variables
interact and change with time. The variables that characterize a state, such
as density, concentration, temperature, pressure and flow rate, are called
state variables. They can be derived from the conservation balances for
mass, component, energy and momentum.
• Additional elements of mathematical model

• mathematical model contains four major elements. These are
• Governing equations
• Defining equations
• Constitutive equations
• Constraints

02/08/2019 92
UNIVERSITY)
• The governing equations of a mathematical model describe how the
values of the unknown variables (i.e. the dependent variables) will change.
The change of the value of a variable with respect to time may be explicit,
in that a governing equation includes a derivative with respect to time, or
implicit, such as when a governing equation has velocity or flux as an
unknown variable.
• A constitutive equation or constitutive relation is a relation between
two physical quantities (especially kinetic quantities as related to
kinematic quantities) that is specific to a material or substance, and
approximates the response of that material to external stimuli, usually as
applied fields or forces.

02/08/2019 93
UNIVERSITY)
Feedforward control systems
• The basic concept of feedforward control is to measure important
disturbance variables and take corrective action before they upset the
process. In contrast, a feedback controller does not take corrective action
until after the disturbance has upset the process and generated a nonzero
error signal.

02/08/2019 94
UNIVERSITY)
Feedforward control has several disadvantages:
1. The disturbance variables must be measured online. In many
applications, this is not feasible.
2. To make effective use of feedforward control, at least an approximate
process model should be available. In particular, we need to know how the
controlled variable responds to changes in both the disturbance variable
and the manipulated variable. The quality of feedforward control depends
on the accuracy of the process model.
3. Ideal feedforward controllers that are theoretically capable of achieving
perfect control may not be physically realizable. Fortunately, practical
approximations of these ideal controllers can provide very effective
control.

02/08/2019 95
UNIVERSITY)
FEEDFORWARD CONTROLLER DESIGN BASED ON
STEADY-STATE MODELS
• Distillation column shown in Fig. 15.8, which is used to separate a binary mixture.
• In Fig. 15.8, the symbols B, D, and F denote molar flow rates, while x, y, and z are
the mole fractions of the more volatile component. The objective is to control the
distillate composition y despite measurable disturbances in feed flow rate F and
feed composition z, by adjusting distillate flow rate D. It is assumed that
measurements of x and y are not available.

02/08/2019 96
UNIVERSITY)
• The steady-state mass balances for the distillation column can be written as
• Solving (15-4) for B and substituting into (15-5) gives
• Because x and y are not measured, we replace these variables by their set
points to yield the feedforward control law:
• Thus, the feedforward controller calculates the required value of the

manipulated variable D from measurements of the disturbance variables, F
and z, and knowledge of the composition set points xsp and ysp· Note that the
control law is nonlinear owing to the product of F and z .

02/08/2019 97
UNIVERSITY)
Blending system is shown in Fig 15.9
• The aim is to design a feedforward control scheme to maintain exit

composition x at a constant set point xsp, despite disturbances in inlet
composition, x1. Suppose that inlet flow rate w1 and the composition of the
other inlet stream x2 are constant.
02/08/2019 98
UNIVERSITY)
• It is assumed that x1 is measured but x is not. If x were measured, then
feedback control would also be possible. The manipulated variable is inlet
flow rate w2. The flow-head relation for the valve on the exit line is given
by
• Note that the feedforward controller has two input signals: the x1
measurement x1m, and the set point for the exit composition xxp·
• The starting point for the feedforward controller design is the steady-state
mass and component balances
• where the bar over the variable denotes a steady-state value. These
equations are the steady-state version of the dynamic model in Eqs. 2-12
and 2-13. Substituting Eq. 15-8 into Eq. 15-9 and solving for gives:
02/08/2019 99
UNIVERSITY)
• In order to derive a feedforward control law, we replace
• Note that this feedforward control law is based on physical variables rather
than deviation variables.
• The feedforward control law in Eq. 15-11 is not in the final form required
for actual implementation, because it ignores two important
instrumentation considerations:
• First, the actual value of x1 is not available, but its measured value x1m is.
Second, the controller output signal is p rather than inlet flow rate, w2.
02/08/2019 100
UNIVERSITY)
• Thus, the feedforward control law should be expressed in terms of x1m
and p, rather than x1 and w2. Consequently, a more realistic feedforward
control law should incorporate the appropriate steady-state instrument
relations for the w2 flow transmitter and the control valve, as shown
below.
• Composition Measurement for x1
• Suppose that the sensor/transmitter for x1 is an electronic instrument with
negligible dynamics and a standard output range of 4-20 mA. If the
calibration relation is linear, it can be written as
• where (x1)0 is the zero of this instrument and K1 is its gain.

02/08/2019 101
UNIVERSITY)
• where St is the span of the instrument.
Control Valve and Current-to-Pressure Transducer
• Suppose that the current-to-pressure transducer and the control valve
operate as linear devices with negligible dynamics. Then its input-output
relationship can be written as
• where Kv and K1P are the steady-state gains for the control valve and l/P
transducer, respectively, while (w2)0 is the w2 flow rate that corresponds to
the minimum controller output signal of 4 mA.
• This value also corresponds to the minimum signal of 3 psi from the l/P
transducer.

02/08/2019 102
UNIVERSITY)
• Note that all of the symbols in Eqs. 15-8 through 15-14 denote physical
variables rather than deviation variables.
• Rearranging Eq. 15-12 gives
• Substituting (15-14) and (15-15) into (15-11) and rearranging the resulting
equation provides a feedforward control law that is suitable for
implementation:

02/08/2019 103
UNIVERSITY)
• The blending and distillation column examples illustrate that feedforward
controllers can be designed using steady-state mass and energy balances.
The advantages of this approach are that the required calculations are quite
simple, and a detailed process model is not required.
• However, a disadvantage is that process dynamics are neglected, and
consequently the control system may not perform well during transient
conditions.

02/08/2019 104
UNIVERSITY)
FEEDFORW ARD CONTROLLER DESIGN BASED ON
DYNAMIC MODELS
• Consider the block diagram shown in Fig. 15.11. The feedback control has a
signal path through Gt. The disturbance transmitter with transfer function Gt
sends a measurement of the disturbance variable to the feedforward controller
Gf. The outputs of the feedforward and feedback controllers are then added
together, and the sum is sent to the control valve.

02/08/2019 105
UNIVERSITY)
• In contrast to steady-state feedforward control, the block diagram in Fig.
15.11 is based on deviation variables.
• The closed-loop transfer function for disturbance changes in Eq. 15-20 can
be derived using the block diagram algebra
• Ideally, we would like the control system to produce perfect control, where
the controlled variable remains exactly at the set point despite arbitrary
changes in the disturbance variable, D. Thus, if the set point is constant
(Ysp(s) = 0), we want Y(s) = 0, even though D(s) ≠ 0
• This condition can be satisfied by setting the numerator of (15-20) equal to
zero and solving for Gf:

02/08/2019 106
UNIVERSITY)
• Figure 15.11 and Eq. 15-21 provide a useful interpretation of the ideal
feedforward controller. Figure 15.11 indicates that a disturbance has two
effects:
• It upsets the process via the disturbance transfer function Gd; however, a
corrective action is generated via the path through GtGfGvGp.
• Ideally, the corrective action compensates exactly for the upset so that
signals Yd and Yu cancel each other and Y(s) = 0.

02/08/2019 107
UNIVERSITY)
Example 15.2

02/08/2019 108
UNIVERSITY)

2018-2019 Process Dynamic and Control (Covenant University)

Загружено:

Сведения о документе

Авторское право

Доступные форматы

Поделиться этим документом

Поделиться или встроить документ

Параметры публикации

Этот документ был вам полезен?

Это неприемлемый материал?

Авторское право:

Доступные форматы

2018-2019 Process Dynamic and Control (Covenant University)

Загружено:

Авторское право:

Доступные форматы

PROCESS DYNAMIC AND

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

(The overbars denote nominal steady-state design values.)

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

OPEN- LOOP SYSTEM (CONTROL)

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

U D In an open-loop system, the

U(t) PROCESS Y(t)

Figure: A simplified open loop control system

• The disturbance can be measured

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

Fig C. Typical Feedforward control systems

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

Feedforward control for a heated tank

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

A closed –loop system is a system with a total control action /

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

1. Negative feedback, this refers to the desirable situation where the

I. Does not require identification of all possible disturbances and their

I. It waits until the effect of disturbance is felt by the system.

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

• The disturbances = Water from toilet flushing

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

W (1) First way

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

(2) Second way w

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

• An inherent disadvantage of proportional-only control is that a steady-state

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

• The response of the PI controller to a unit step change in e(t) is shown in

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

• The corresponding transfer function is

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

• This method can be implemented quite easily by placing the PD element in

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT

2017/2018 PROCESS DYNAMIC AND CONTROL (COVENANT