Homepage of the specialized course

Process Control

at Norwegian University of Life Sciences (NMBU)

Spring semester 2018

---

Syllabus

• Course level at NMBU: 400.
• Aim of the course: Students have knowledge about practical and theoretical process control concepts. They have skills in simulation for analysis and design of process control systems, and they are able to work scientifically with process control problems. The range of applications studied is broad, however with a special focus on applications in water resource recovery facilities (wrrfs).
• Teaching methods: Lectures (at NMBU); Self-study; Exercises - both voluntary and compulsory.
• Assessment: Oral exam with curriculum as listed under Literature and Powerpoint in the lecture plan below. However, the exact curriculum for the exam may adjusted (limited) somewhat in due time before the exam. To obtain a grade better than F for the course, it is necessary that the compulsory exercises have been passed (i.e. with approved results).
• Course responsible: Prof. Harsha Ratnawera, NMBU.
• Course teacher: Finn Aakre Haugen, PhD, TechTeach. E-mail: finnhaugen@hotmail.com. Tel.: 97019215.
• Participants:

Xiaodong Wang <xiaodong.wang@nmbu.no>
Nataliia Sivchenko <nataliia.sivchenko@nmbu.no>

Duo Zhang <duo.zhang@nmbu.no>

Aleksander Hykkerud <aleksander.hykkerud@nmbu.no>

Abhilash Muralidharan Nair <muralidharan.nair.abhilash@nmbu.no>

Dino Ratnaweera <dino@doscon.no>

Karla Sladka <Karla.Sladka@seznam.cz>

• Literature:
• Haugen, F. (2010). Basic Dynamics and Control (freely available):
• Textbook [referred to as B-T]
• Exercise book, which includes detailed solutions [B-E]
• Haugen, F. (2010). Advanced Dynamics and Control (freely available):
• Textbook [A-T]
• Exercise book, which includes detailed solutions [A-E]
• Lecture notes and Powerpoint presentations which will become available from the below lecture plan.
• Instructional videos from the TechVids library (freely available) are suggested for some of the lectures.
• Simulators from the SimView library (freely available). Simulators will be run by the teacher during lectures, and will be parts of the exercises.
• Software tools: Matlab/Simulink.
  In case not all of us will have access to Matlab/Simulink during the course: A free alternative to Matlab is Octave which is almost fully compatible with Matlab. However, there is no free alternative to Simulink that is compatible with Simulink (I am still looking for free block-diagram based simulation software that may be used as an alternative to Simulink).

---

Lecture plan

The dates for Lectures 2-6 will be discussed with the participants during the startup meeting/lecture on 21. Dec.

Lesson no.	Topic	Teaching activity	Powerpoint present- ations	Literature	Videos tutorials	N/A	Problems (voluntary). Solutions are included in the refs.
Introduction (“INT”):
INT1	Introduction to the course	Lecture 1 (L1) 21.12.17, 1015-1400 SU-113, NMBU	-	-	-	-	-
INT2	Introduction to process control	L1 (21.12)	PPT: Intro to process control	B-T: Ch. 1 B-T: App. A	Feedback control	-	B-E: 1.1, 1.2, 1.3, 1.4, 1.5.
INT3	Some Internet resources for process control	L1 (21.12)	-	Some Internet resources for process control	-	-	-
Instrumentation (“INS”):
INS1	Instrumentation of control systems: - Controllers - Sensors - Actuators	Self-study	PPT: Controllers PPT: Sensors PPT: Actuators	B-T: 9 Literature: The PPT-files (at left)	-	-	B-E: 9.1, 9.3. Additional problems for INS1. Solutions.
Basic systems theory (“BST”):
BST1	Differential equation models, incl. block diagram representation (of diff eqs) and state space model form	L1 (21.12)	-	B-T: 2	-	-	B-E: 2.1, 2.2, 2.3, 2.4.
BST2	Mathematical mechanistic (first principles) modeling	L1 (21.12)	-	B-T: 3.1, 3.2, 3.3, 3.4.	-	-	B-E: 3.1, 3.2, 3.3.
BST3	From diff eq to simulation algorithm (ready for programming)	L1 (21.12)	-	A-T: 6, 7, 8.1, 8.2.	-	-	A-E: 6.1, 7.1, 7.2, 7.3, 8.1, 8.2.
BST4	Laplace transform	L1 (21.12)	-	B-T: 4.	-	-	B-E: 4.1, 4.2.
BST5	Transfer functions	L1 (21.12)	-	B-T: 5.	-	-	B-E: 5.1, 5.2, 5.3, 5.4, 5.6.
BST6	Process dynamics	L1 (21.12)	PPT: Process dynamics	B-T: 6.	Time-constant and integrator dynamics	-	B-E: 6.1, 6.3, 6.4, 6.6, 6.7.
Learning/reviewing relevant computer programming:
CP1	Matlab	Self-study	-	Tutorials at MathWorks (explore by yourself)	Matlab Quickie	-	-
CP2	Matlab Control System Toobox	Self-study	-	Tutorial for Control System Toolbox for Matlab (a little old, but still relevant)	-	-	-
CP3	Simulink	Self-study	-	Tutorials at MathWorks (explore by yourself)	Simulink Quickie	-	-
Compulsory exercise for Lecture 1. Deadline: Tuesday 9.1 2018, 16:00. Solutions by students: Abhilash. Aleksander. Duo. Karla. Xiaodong. Discussion of student solutions takes place in beginning of Lecture 2 (11.1.18).
PID control (“PID”):
Lesson no.	Topic	Teaching activity	PPT	Literature	Videos tutorials	N/A	Problems (voluntary)
PID1	Measurement filtering, incl. discrete-time filter algorithm	Lecture 2 (L2) Date: 11.1.18. Time: 1015-1500. Room: BT3A11 (Bio- teknologi- bygget)	PPT: Filtering of measure- ment signals	B-T: 7.2.2 A-T: 8.3	Time-constant lowpass filter	-	B-E: 7.1. A-E: 8.3.
PID2	Feedback control with PID controller, incl. discrete-time PID control algorithm	L2	PPT: PID control PPT: Reverse or direct action in the PID- controller PPT: Anti windup in PID controllers	B-T: 7.1-7.4 A-T: 8.4.1, 8.4.2	Feedback control Reverse or direct action in the PID controller? Anti windup in a PI(D) controller	-	B-E: 7.4, 7.5, 7.6, 7.7, 7.9, 7.10. A-E: 8.4, 8.6, 8.7.
PID3	Feedback control with On/off-controller	L2	PPT: On-off control	The PPT-files (at left)	-	-	-
PID4	Experimental performance analysis of control systems: The IAE index.	L2	-	Ch. 2.8 in PID Control (text-book, 2004)	-	-	-
PID5	Experimental stability analysis (in terms of gain margin and phase margin) of control systems	L2	PPT: Experimental stability analysis	Section 3.2.2 in the article “Relaxed Ziegler-Nichols Closed Loop Tuning of PI Control- lers”	-	-	Problem to Lesson PID5
PID6	Tuning of PID controllers: - Ziegler-Nichols method - Auto-tuning - Good Gain method - Skogestad method - Gain scheduling	L2	PPT: Ziegler-Nichols' method of PID controller tuning PPT: Skogestad method of PI controller tuning	B-T: 10 PPT-file: Ziegler-Nichols method (at left) PPT-file: Skogestad method (at left)	PID controller tuning with Ziegler-Nichols' oscillations method Gain scheduling	-	B-E: 10.1, 10.2, 10.3, 10.4, 10.6, 10.7, 10.8, 10.9.
PID7	How a control system may become unstable, and how to recover stability by controller retuning	L2	PPT: How a control system may become unstable - and how to recover stability by controller retuning	B-T: 7.5	How a control system may become unstable	-	B-E: 7.11, 7.12, 7.13.
Compulsory exercise for Lecture 2. Deadline: Tuesday 30.1 2018, 16:00. Solutions by students: Abhilash. Aleksander. Duo. Karla. Xiaodong. Discussion of student solutions takes place in beginning of Lecture 3 (1.2.18).
Control structures (“CS”):
Lesson no.	Topic	Teaching activity	PPT	Literature	Videos tutorials	N/A	Problems
CS1	Feedforward control	Lecture 3 (L3) Date: 1.2.18 Time: 1015-1500 Room: MU68 (Meieri- bygget)	PPT: Feedforward control	B-T: 8	Feedforward control	-	B-E: 8.2.
CS2	Cascade control	L3	PPT: Cascade control	B-T: 11.1	Cascade control	-	B-E: 11.1.
CS3	Ratio control	L3	PPT: Ratio control	B-T: 11.2	-	-	B-E: 11.4.
CS4	Split-range control	L3	PPT: Split-range control	B-T: 11.3	-	-	B-E: 11.5.
CS5	Averaging level control of buffer tanks (equalization magazines)	L3	PPT: Averaging level control	B-T: 11.4 Lecture notes	-	-	B-E: 11.6.
CS6	Plantwide control - basic principles	L3	PPT: Plantwide control - basic principles	B-T: 11.5	-	-	B-E: 11.7, 11.8.
CS7	Sequential control	L3	PPT: Sequential control	B-T: 12	Sequential control	-	B-E: 12.1.
Compulsory exercise for Lecture 3. Deadline: Tuesday 20.2 2018, 16:00. Solutions by students: Abhilash. Aleksander. Duo. Karla. Xiaodong. Discussion of student solutions takes place in beginning of Lecture 4 (22.2.18).
Advanced systems theory (“AST”):
Lesson no.	Topic	Teaching activity	PPT	Literature	Videos tutorials	N/A	Problems (voluntary)
AST1	Continuous-time state-space models, incl. linearization	Lecture 4 (L4) Date: 22.2.18 Time: 1015-1500 Room: MU68 (Meieri- bygget)	-	A-T: 1	-	-	A-E: 1.1, 1.2, 1.3.
AST2	Stability analysis of continuous-time dynamic systems	L4	-	A-T: 4	-	-	A-E: 4.1, 4.2, 4.3.
AST3	Stability analysis of continuous-time feedback systems (pole/eigenvalue based analysis)	L4	-	A-T: 5.1, 5.2	-	-	A-E: 5.1.
AST4	Discrete-time state-space models	L4	-	A-T: 9.	-	-	A-E: 9.1, 9.2, 9.3, 9.4.
AST5	The z-transform	L4	-	A-T: 10.	-	-	A-E: 10.1, 10.4.
AST6	z-transfer functions	L4	-	A-T: 11.1, 11.2, 11.3, 11.6, 11.7.	-	-	A-E: 11.1, 11.2, 11.5, 11.6.
AST7	Stability analysis of discrete-time state-space models	L4	-	A-T: 13, except 13.2.	-	-	A-E: 13.1, 13.2, 13.3, 13.4.
AST8	Stochastic signals	L4	-	A-T: 15. SimView simulators: - White and Coloured Noise - Correla- tion	-	-	A-E: 15.1.1, 15.1.2, 15.1.3, 15.3, 15.4, 15.5.
AST9	Introduction to optimization methods (used in later lessons for optimal estimation and optimal control)	L4	-	Lecture notes. Matlab scripts: - Grid search - Newton’s method - NLP with fmincon() in Matlab	-	-	Exercise included in the compulsory exercises for this lecture.
Compulsory exercise for Lecture 4. Deadline: Tuesday 13.3 Wednesday 14.3 2018, 16:00. Solutions by students: Abhilash. Aleksander. Duo. Karla. Xiaodong. Discussion of student solutions takes place in beginning of Lecture 4 (1516.3.18).
Estimation of model parameters (“EMP”):
Lesson no.	Topic	Teaching activity	PPT	Literature	Videos tutorials	N/A	Problems
EMP1	Estimation of model parameters using Least Squares method	Lecture 5 (L5) Date: 1516.3.18 Time: 1015-1500 Room: MU68 (Meieri- bygget)	-	A-T: 16, except 16.3.5	-	-	A-E: 16.1, 16.2, 16.4, 16.6, 16.7.
EMP2	Estimation of model parameters using nonlinear optimization	L5	-	Lecture notes: Section 1.3.3	-	-	-
EMP3	Estimation of transfer function models using Subspace identification	L5	-	A-T: 16.3.5 Matlab script with subspace id. of DC motor using these experimental data.	-	-	A-E: 16.8.
Estimation of state variables (“ESV”):
ESV1	State estimation with Observer	L5	-	A-T: 17	-	-	A-E: 17.1.
Compulsory exercise for Lecture 5. Deadline: Tuesday 10.4 2018, 16:00. Solutions by students: Abhilash. Aleksander. Duo. Karla. Xiaodong. Discussion of student solutions takes place in beginning of Lecture 4 (12.4.18).
Model-predictive control (“MPC”):
Lesson no.	Topic	Teaching activity	PPT	Literature	Videos tutorials	N/A	Problems
ESV2	State estimation with Kalman Filter	Lecture 6 (L6) Date: 12.4.18 Time: 1015-1700 Room: MU68 (Meieri- bygget)	Kalman Filter	A-T: 18	Kalman Filter	-	A-E: 18.1.
ESV3	State estimation with Moving Horizon Estimator (MHE)	L6	MHE	Lecture notes: Section 1.3.4 Matlab script presented in Example 1.11 the lecture notes	-	-	-
MPC1	Model-based Predictive Control (MPC)	L6	MPC Simulations run in the lecture (to appear):	A-T: 22 Lecture notes: Section 1.3.5	-	-	-
Compulsory exercise for Lecture 6. Deadline: Thursday 26.4 2018, 16:00. Solutions by students: Abhilash. Aleksander. Duo. Karla. Xiaodong. Solution to Problem 3 - MHE for estimation of K (in stead of d). Unfortunately, there is no time for discussion of student solutions of this exercise (but comments to the handins are provided to each student).
Exam:
2.5.17	Exam (oral)	TF102	-	-	-	-	-

---

Updated 29 April 2018 by Finn Aakre Haugen, course teacher.