Process Dynamics and

Control
Lecture 1 - Introduction
2018-2019
Costin Sorin Bildea
University “Politehnica” of Bucharest
Department of Chemical and Biochemical Engineering

Email: c.bildea@tudelft.nl
s_bildea@upb.ro

Room: E2.120
Phone:
Example of a
dynamic process
• Moving
• Unstable
• Disturbances
 Example of process
control
• Path set (in mind)
• Uses senses (speed, direction, position, etc)
• (Peripheral) nervous system transmits information
to the brain
• The brain processes the information and sends
commands to the body
• Actions are implemented
• pedaling (feet)
• steering (hands)
• balancing (body mass)
Other examples
• temperature control when showering
• thermostat in houses
• consumer products (CD, dish washers,.)
• airplanes
• chemical plants
Lecture 1 - Introduction
• Goals
• to provide an introduction to the contents of the course
• to introduce basic concepts

• Importance of Process Dynamics

• Levels of Process Control / Steps in designing control systems
The chemical process
• converts raw materials into products
• chemical, physical, mechanical changes
• units: reactors, distillation columns, evaporators, heat exchangers,
etc.
Objectives of operation
• Operate the processing units safely
• Maintain production rate and product quality
• In a profitable way

Need to:
• monitor &
• change process variables

Job of the control system

Process dynamics and control

… aspect of chemical engineering concerned with analysis,

design, and implementation of control systems that facilitate
the achievement of process safety, production rate,
and product quality
Process
dynamics
 Steady state F0(t)

• “Nothing is changing”
h(t)

F0(t) = F(t) = FS h(t) = hS

F(t)

• Time derivatives are zero

• Design:

liquid height in tank h

- height & diameter tank Flow rate

- diameter pipe

• Safety
- high / low level alarm
pipe diameter
- interlock
Dynamic behaviour F0(t)

- change F0 to a new value h(t)

F(t)

- what will happen with h(t), F(t) ?

F0

time

h
1
time
 Chemical Process

Deliver oil at a constant temperature, T*

Steam

Oil feed Heat exchanger

Trap
History of Process Control:
• Manual control of processes (< 1950):
? sensors and actuators operated by humans
 (Manual) process control
Steam Controller output
Manipulated variable
(valve opening) Set point
Control valve
(actuator)

Process variable
(temperature)

Sensor
Oil feed Heat exchanger (thermometer)

Trap
Performance of (manual) process control
Temperature

Manual control
T*
Setpoint

without control

time

Increase of the feed rate

History of Process Control :
• Manual control of processes (< 1950):
? sensors and actuators operated by humans
• Automatic control (~1950 - now):
? mechanical, electrical and digital controllers
? mainly one input - one output
• Model based control (~1985 - now)
? digital controllers with internal process model
? multiple inputs, multiple outputs
 Automatic process control
Set point
Steam

Control TC
valve Controller

Oil feed Heat exchanger

Sensor
Trap
Performance of (automatic) process
control
Temperature

Manual control
T*
Setpoint
Automatic control

without control

time

Increase of the feed rate

Levels of Process control

High-level decisions
• Process design
• Control structure
• Algorithms (type of controller)
• Controller tuning
Low-level decisions
 Level 0 - Process design

The driver controls both Control of direction is not needed Skillful driver required!
speed and direction
 Level 1 - Control structure
Set point
Steam
SP

OP
Control TC Feedback controller
valve

PV Takes action after the disturbance

has affected the process

Can we improve it ?

Oil feed Heat exchanger

Sensor
Trap
 Level 1 – Control structure
Feedforward controller

Feedforward controller Takes action when the disturbance

occurs, before the variable of interest
FFC is affected
Steam

Control Drawbacks:
- The disturbance must be measured
valve
- A good model of the process is
needed
- Does not deal with all disturbances
Oil feed Heat exchanger

Sensor
Trap
 Level 1 – Control structure
Combined feedback – feedforward control

FFC
Steam Set point
Control S TC But…
valve
The outlet temperature is determined by the
steam flow rate, which in turn depends on:
- Valve opening
Oil feed Heat exchanger - Steam pressure

Can we improve the control structure such

Sensor that it deals with variations of steam
Trap pressure?
 Level 1 – Control structure
S
Steam
FFC
Feed-forward T* Set point
controller Control TC
valve FC
Temperature controller
Flow controller
(master)
(slave)

Oil feed Heat exchanger

Sensor
Trap
Level 2 – Control
F
algorithm
in

Water feed

Ac h

Fout

Water outlet
Dynamic models
dh Water feed

Ac  Fin  Fout
dt
Fin = Fout = Fin,S = Fout,S Ac h

Fout
hS
Water outlet

d  h  hs 
Ac   Fin  Fin , S    Fout  Fout , S 
dt

h  hs  y
dy 1
Fin  Fin , S  d  d  u 
dt Ac
Fout  Fout , S  u
Level 2 – Control algorithm
u0
d 0
dy 1
 d d
dt Ac y t
Ac

t
Level 2 – Control algorithm

Fout  Fout , S  K   h  hS 
Water feed

h
LC hS

Fout

Water outlet
 Level 2 – Control algorithm
Fout  Fout , S  K   h  hS  Water feed

h hS
LC

Control law: uKy F out

dy 1 Water outlet

 d  K  y 
dt Ac
increasing
d  K 
y t   1  exp    t   K
K  A
 c 
 y

d
lim y t  
t  K t
Level 2 – Control algorithm
Question:

Can you find a better control law

u(t) = some function of y(t)

such that

limy t   0
t 
 Level 3 – Controller tuning
uKy Water feed

h hS
LC
How to choose the parameter K ?
F out

d
y t  
Water outlet

lim
t  K

Apparently, larger K gives better control increasing

K
but… y

The model used for design assumes that

- The level measurement is instantaneous
- The valve opens instantaneously t
Process variables
• Inputs
• manipulated variables u
• disturbances d
• Outputs (controlled) y
• State variables x Disturbances
• Set points (references) r Manipulated Controlled
variables System

Controlled
variables

Controller

Reference
variables
The process control system
• monitoring outputs (sensor)
• making decisions about how to change the manipulated inputs
(controller)
• to obtain desired output behaviour (set point)
• and implementing the decisions (control elements)
• when the process is affected by disturbances
Overview of control system design
• Assess the process & define objectives
• control needed?
• another alternative ? (design)
• performance
• Select process variables
• variables of interest (outputs)
• disturbances
• inputs available
• Select control structure
• connections input / output
• Design the controller (s)
• control algorithm, algorithm parameters (tuning)
“Laws” of process control
1. The best control system is the simplest one that will do the job

2. You must understand the process before you can control it

3. Liquid levels must always be controlled (“Safety first”)

Other concepts
• open-loop
• closed – loop

• manual
• automatic

• regulatory control (disturbance rejection)

• servo-control (set-point tracking)

• Use of computers (digital signals, sampling,…)

Languages of process control
• Time domain (differential equations)
• Laplace domain (transfer functions)
• Frequency domain (Bode, Nyquist plots)
• State variable (matrices)
• Sampled-data systems (z-domain)
Conclusions
• The dynamic response of a process is important, and must be
considered in the process design

• The process itself places restrictions on the achievable performance,

that can not be overcome by controller complexity

• Control system structure, type of controller, and tuning of the

controller are important engineering decisions
Knowledge, abilities and skills
• Develop mathematical function models of dynamic processes
• Use of Laplace transform and transfer functions
• Analyse stability & dynamic response
• Determine process dynamics from data
• Understand feedback controllers
• Analyse and tune PID controllers for performance
• Read P&ID
• Design feedforward and cascade control
• Analyse multivariable systems (interactions)
• Process control problem based on a flowsheet
Contents of the course
• Introduction to Process Control
• PART I: Process Modelling and Dynamics
• Model Development
• Procedure; Formulation; Degree of Freedom Analysis; Numerical Solution and Model
Validation
• Model Analysis
• Operating Window; Linearisation; Analysis Tools
• Dynamic Behaviour of Process Systems
• Basic Building Blocks - first-order; second order; time delay systems
• Series and Parallel Structures - non-interacting and interacting systems; higher order
systems; inverse response
• Recycle Structures
Contents
• PART II: Implementation
• Conventional control
• Feedback Control
• PID Control Algorithm
• Tuning Rules - goals and techniques
• Advanced control systems
• Cascade control; feedforward control; override control; split-
range control; anti reset wind-up
• Plantwide control
Contents
• PART III: Advanced topics
• Multivariable processes
• Process identification & parameter estimation
• Computer control
 Textbook

Thomas Marlin - Process Control, Designing Processes and

Control Systems for Dynamic Performance, 2nd Edition

http://www.pc-education.mcmaster.ca/Book_Links.htm
Textbooks
• C.A. Smith, Automated Continuous Process Control, John Wiley & Sons, 2002.
(typical dynamic behaviour, development of process control structures)

B.A. Ogunnaike and W. H. Ray, Process Dynamics, Modelling and Control,

Oxford University Press, 1994. (dynamic models, analysis tools, controller
design)

W.L. Luyben, and M.L. Luyben, Essentials of Process Control, McGraw-Hill,

1997. (dynamic models, analysis tools, controller design)

A.C. Dimian, C.S. Bildea, A.A. Kiss, Integrated Design and Simulation of
Chemical Processes, Elsevier, Amsterdam, 2014. (dynamic simulation,
plantwide control)