100% found this document useful (1 vote)
15 views200 pages

Introduction To Model Predictive Control (MPC)

Uploaded by

haopengchen233
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
100% found this document useful (1 vote)
15 views200 pages

Introduction To Model Predictive Control (MPC)

Uploaded by

haopengchen233
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 200

Introduction to Model Predictive

Control (MPC)

Oscar Mauricio Agudelo Mañozca


Bart De Moor

Course :
Computergestuurde regeltechniek

ESAT - KU Leuven
May 11th, 2017
Basic Concepts
Control method for handling input and state constraints within an optimal control setting.

Principle of predictive control

  yref  y(k  i) 
2
min
u ( k ), ,u ( k  N 1)
Past Future i 1
subject to
Reference  yref  model of the process
 input constraints
Prediction of y (k )  output / state constraints
measurement

y (k ) Why to use MPC ?

u (k )  It handles multivariable interactions


time
 It handles input and state constraints
k 2 k 1 k k 1 k 2 k 3 kN
 It can push the plants to their limits
Prediction horizon
of performance.
 It is easy to explain to operators and
engineers

Introduction to Model Predictive Control Course: Computergestuurde regeltechniek 2


Some applications of MPC
Control of synthesis section of a urea plant

MPC strategies have been used for stabilizing and maximizing the throughput of the
synthesis section of a urea plant, while satisfying all the process constraints.

Urea plant of Yara at Brunsbüttel (Germany), where a MPC control system has been
set by IPCOS

Introduction to Model Predictive Control Course: Computergestuurde regeltechniek 3


Some applications of MPC
Control of synthesis section of a urea plant
Reaction 1: Fast and Exothermic

2NH3 + CO2  NH2COONH4


Carbon Ammonium
Ammonia
dioxide carbamate

Reaction 2: Slow and Endothermic

NH2COONH4  NH2CONH2 + H2O


Ammonium Urea Water
carbamate

Throughput increment of 11.81 t/h


thanks to the MPC controller
Results of a preliminary study done by

Introduction to Model Predictive Control Course: Computergestuurde regeltechniek 4


Some applications of MPC
Flood Control: The Demer

A Nonlinear MPC control strategy has been implemented (2016) for avoiding future
floodings of the Demer river in Belgium. Partners: STADIUS, Dept. Civil Engineering of KU
Leuven, IPCOS, IMDC, Antea Group, and Cofely Fabricom.

The Demer in Hasselt

Flooding events due to heavy rainfall:


1905, 1926, 1965, 1966, 1993-1994,
1995, 1998, 2002 and 2010.
The Demer and its tributaries in the south of
Control Strategy: PLC logic the province of Limburg
(e.g., three-pos controller)

Introduction to Model Predictive Control Course: Computergestuurde regeltechniek 5


Some applications of MPC
Flood Control: The Demer

DIEST

HASSELT

Flooded area during the flood event of 1998.


Control Strategy: PLC logic (e.g., three-position controller)

Introduction to Model Predictive Control Course: Computergestuurde regeltechniek 6


Some applications of MPC
Flood Control: The Demer

Upstream part of the Demer that is modelled


and controlled in a preliminary study carried
out by STADIUS

Maximal water levels for the


five reaches for the current
Notice: The MPC controller takes rain three-pos. controller and the
MPC controller together with
predictions into account! their flood levels (Flood event
2002) .

Introduction to Model Predictive Control Course: Computergestuurde regeltechniek 7


Some applications of MPC

In addition MPC, has been used


 in all sort of petrochemical and chemical plants,
 in food processing,
 in automotive industry,
 in the control of tubular chemical reactors,
 in the normalization of the blood glucose level of
critical ill patients,
 in power converters,
 for the control of power generating kites under
changing wind conditions,
 in mechatronic systems (e.g., mobile robots),
 in power generation,
 in aerospace,
 in HVAC systems (building control)
 …

Introduction to Model Predictive Control Course: Computergestuurde regeltechniek 8


Basic Concepts
Kinds of MPC

 Linear MPC : it uses a linear model of the plant x(k  1)  Ax(k )  Bu(k )
 Convex optimization problem
 Nonlinear MPC: it uses a nonlinear model of the plant x(k  1)  f  x(k ), u(k ) 
 Non-convex optimization problem

Remark: Since linear MPC includes constraints, it is a non-linear control strategy !!!

Linear MPC formulation (Classical MPC)

N N 1
min   x(k  i)  xref (k  i)  Q  x(k  i )  xref (k  i)     u(k  i)  u ref (k  i)  T R  u(k  i )  u ref (k  i ) 
T

x N ,u N
i 1 i 0

subject to
x(k  1  i)  Ax(k  i)  Bu(k  i), i  0,1, , N 1, Model of the plant

u(k  i)  umax , i  0,1, , N  1, Input constraints

x(k  i)  xmax , i  1, 2, , N, State constraints

with: x N   x(k  1); x(k  2); ; x(k  N ) , u N  u(k ); u(k  1); ; u(k  N 1)
Introduction to Model Predictive Control Course: Computergestuurde regeltechniek 9
MPC Algorithm
Typical MPC control Loop

xref (k )
u( k ) u(t )
u ref (k ) MPC ZOH Plant y (t )

xˆ (k )
Observer y (k )
Ts
Digital system
MPC Algorithm
At every sampling time:

 Read the current state of the process, x(k)


 Compute an optimal control sequence by solving the MPC optimization problem

Solution  u(k ), u(k  1), u(k  2), , u(k  N 1)

 Apply to the plant ONLY the first element of such a sequence  u(k)

Introduction to Model Predictive Control Course: Computergestuurde regeltechniek 10


LQR and Classical MPC
For simplicity, Let’s assume that the references are set to zero.

LQR CLassical Linear MPC


N N 1

min  x(k ) Q x(k )  u(k ) R u(k )
T T min  x(k ) Q x(k )   u(k ) T R u(k )
T
x N ,u N
x ,u k 1 ki 
=00
k 1

subject to subject to
x(k  1)  Ax(k )  Bu(k ), k  0,1, , N 1,
x(k  1)  Ax(k )  Bu(k ), k  0,1, ,
x(k )  xmax , k  1, 2, , N,
u(k )  umax , k  0,1, , N  1,
The optimal solution has the form: u(k )  Kx(k )
PRO PRO
 Explicit, Linear solution  Takes constraints into account
 Low online computational burden  Proactive behavior

CON CON

 Constraints are not taken into account  High online computational burden

 No predictive capacity  No explicit solution


 feasibility? Stability?

If N∞, and constraints are not considered  the MPC and LQR give the same solution

Introduction to Model Predictive Control Course: Computergestuurde regeltechniek 11


MPC optimization problem – Implementation details
The following optimization problem,

N N 1
min   x(k  i)  xref (k  i)  Q  x(k  i )  xref (k  i)     u(k  i)  u ref (k  i)  T R  u(k  i )  u ref (k  i ) 
T

x N ,u N
i 1 i 0

subject to
x(k  1  i)  Ax(k  i)  Bu(k  i), i  0,1, , N 1,
u(k  i)  umax , i  0,1, , N  1,
x(k  i)  xmax , i  1, 2, , N,

with: x N    x(k  1); x(k  2); ; x(k  N ) , u N   u(k ); u(k  1); ; u(k  N 1)
n N x n N u

can be rewritten as a LCQP (Linearly Constrained Quadratic Program) problem in x as follows:

1 T
min x Hx  f T x
x 2
with:
subject to
 xN ; u N 
N  nx  nu 
Ae x  be x 
Ai x  bi

Introduction to Model Predictive Control Course: Computergestuurde regeltechniek 12


MPC optimization problem – Implementation Details

where
N times
Q   x ref (1) 
   
   
 Q   x (N ) 
 N  nx  nu   N  nx  nu  
f  x u    H  ref
N  n  n 
H   2  
 R   u ref (0) 
   
   
 R u ref ( N  1) 
N times

 x max 
 I nx  N   
    2N times
  I nx  N   x max 
Ai 
2 N  nx  nu   N  nx  nu   2 N  nx  nu 
 bi   
I nu  N   u 
  max

 I nu  N    2N times
  
 max 
u

nx = number of states, nu = number of inputs, N = prediction horizon

Introduction to Model Predictive Control Course: Computergestuurde regeltechniek 13


MPC optimization problem – Implementation Details

 I nx N times B   Ax(k ) 
   0 
  A I nx B 
A e  x  x u     
 N n  N  n  n 
b e  N nx
  
   
  A I nx N times B   0 

nx = number of states, nu = number of inputs, N = prediction horizon

Remarks:
 The problem is convex if Q and R are positive semi-definite

 The hessian matrix H, and the matrices Ae and Ai are sparse.

 Number of optimization variables: N(nx + nu). This number can be reduced to N·nu (Condensed
form of the MPC) through elimination of the states x(k) but at the cost of sparsity !!!

Matlab function for solving the LCQP optimization problem

x_tilde = quadprog (H, f, Ai, bi , Ae, be)

Introduction to Model Predictive Control Course: Computergestuurde regeltechniek 14


MPC with terminal cost
Large Prediction Horizon

What
happens?

We have a
winner !!!

Short Prediction Horizon

What
happens?
We have an
accident !!!

N should be large enough in order to keep the process under control

Introduction to Model Predictive Control Course: Computergestuurde regeltechniek 15


MPC with terminal cost
N 1 N 1
min   x(k  i)  xref (k  i)  Q  x(k  i)  xref (k  i)     u(k  i)  u ref (k  i)  T R  u(k  i)  u ref (k  i) 
T

x N ,u N
i 1 i 0

  x(k  N )  xref (k  N )  S  x(k  N )  xref (k  N ) 


T

subject to
x(k  1  i)  Ax(k  i)  Bu(k  i), i  0,1, , N 1,
u(k  i)  umax , i  0,1, , N  1,
x(k  i)  xmax , i  1, 2, , N,

x2 (k )
Main goal of the terminal cost:
x(k  2) x(k  1)
To include the terms for which i ≥ N in the cost
function (To extend the prediction horizon to
infinity) x( k )
x(k  N )
What do we gain?
Effect of the stabilizing
control law u(k) = -Kx(k)
“Stability”

x(k  ) x1 (k )
steady-state xref (k )  0

Introduction to Model Predictive Control Course: Computergestuurde regeltechniek 16


MPC with terminal cost
How to calculate S ?
By solving the discrete-time Riccati equation,

AT SA  S  (AT SB)(BT SB  R)1 (BT SA)  Q  0

where u(k) = -Kx(k) is the stabilizing control law. K   B SB  R   B SA  .


T 1 T

How to carry out this calculation in Matlab?

[K,S] = dlqr (A, B, Q, R )

Implementation details of the MPC with terminal constraint


The only change in the LCQP is the hessian matrix
N - 1 times
Q 
 
 
 Q 
 N  nx  nu   N  nx  nu    
H   2 S 
 R 
 
 
 R 
 N times

Introduction to Model Predictive Control Course: Computergestuurde regeltechniek 17


H0K03a : Advanced Process Control
Model-based Predictive Control 1 : Introduction

Bert Pluymers
Prof. Bart De Moor
Katholieke Universiteit Leuven, Belgium
Faculty of Engineering Sciences
Department of Electrical Engineering (ESAT)
Research Group SCD-SISTA

[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction


• Overview
• Motivating Example
Overview
• MPC Paradigm
• History
• Mathematical Formulation
• MPC Basics
• MPC 1 : Introduction
• MPC 2 : Dynamic Optimization
• MPC 3 : Stability
• MPC 4 : Robustness

• Industry Speaker : Christiaan Moons (IPCOS)


(november 3rd)

Signal processing
Identification

System Theory
Automation 1
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Overview
• MPC Paradigm
• History
• Mathematical Formulation
• MPC Basics
Lesson 1 : Introduction

• Motivating example
• MPC Paradigm
• History
• Mathematical Formulation
• MPC Basics

Signal processing
Identification

System Theory
Automation 2
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Motivating Example
• MPC Paradigm
• History
• Mathematical Formulation
• MPC Basics
Consider a linear discrete-time state-space model

called a ‘double integrator’.


We want to design a state feedback controller

that stabilizes the system (i.e. steers it to x=[0; 0])


starting from x=[1; 0], without violating the imposed
input constraints
Signal processing
Identification

System Theory
Automation 3
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Motivating Example
• MPC Paradigm
• History
• Mathematical Formulation
• MPC Basics
Furthermore, we want the controller to lead to a
minimal control ‘cost’ defined as

with state and input weighting matrices

A straightforward candidate is the LQR controller,


which has the form

Signal processing
Identification

System Theory
Automation 4
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Motivating Example
• MPC Paradigm
• History
LQR controller
• Mathematical Formulation
• MPC Basics 1

0.5

k,1
x
0

-0.5
0 50 100 150 200 250 300
k
10

0
k

-10
u

-20

-30
0 50 100 150 200 250 300
Signal processing
Identification k

System Theory
Automation 5
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Motivating Example
• MPC Paradigm
• History
LQR controller with clipped inputs
• Mathematical Formulation
• MPC Basics 1

0.5

k,1
0
x
-0.5

-1
0 50 100 150 200 250 300
k
0.15
0.1
0.05
k

0
u

-0.05
-0.1

0 50 100 150 200 250 300


Signal processing
Identification k

System Theory
Automation 6
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Motivating Example
• MPC Paradigm
• History
LQR controller with R=100
• Mathematical Formulation
• MPC Basics 1

0.5

k,1
x
0

-0.5
0 50 100 150 200 250 300
k
0.1

0.05
k

0
u

-0.05

-0.1
Signal processing 0 50 100 150 200 250 300
Identification k

System Theory
Automation 7
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Motivating Example
• MPC Paradigm
• History
• Mathematical Formulation waste gas
• MPC Basics
F T Cracking P
Furnace
Feed H condenser
EDC
L

EDC / VC / HCl

superheater evaporato
T r P

Fuel gas

Signal processing
Systematic way to deal with this issue… ?
Identification

System Theory
Automation 8
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
MPC Paradigm
• MPC Paradigm
• History
• Mathematical Formulation
• MPC Basics

Signal processing
Process industry in ’70s : how to control a process ???
Identification
and… easy to understand (i.e. teach) and implement !
System Theory
Automation 9
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
MPC Paradigm
• MPC Paradigm
• History → Modelbased Predictive Control (MPC)
• Mathematical Formulation
• MPC Basics

• Predictive : use model to optimize future input sequence


Signal processing
Identification • Feedback : incoming measurements used to compensate for
inaccuracies in predictions and unmeasured disturbances
System Theory
Automation 10
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
MPC Paradigm
• MPC Paradigm
• History MPC has earned its place in the control hierarchy…
• Mathematical Formulation
• MPC Basics

• Econ. Opt. : optimize profits using market and plant information (~day)
• MPC : steer process to desired trajectory (~minute)

Signal processing • PID : control flows, temp., press., … towards MPC setpoints (~second)
Identification

System Theory
Automation 11
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
History
• MPC Paradigm
• History
• Mathematical Formulation
• MPC Basics
Before 1960’s :
• only input/output models, i.e. transfer functions, FIR
models

• Controllers :
• heuristic (e.g. on/off controllers)
• PID, lead/lag compensators, …
• mostly SISO
• MIMO case : input/output pairing, then SISO control

Signal processing
Identification

System Theory
Automation 12
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
History
• MPC Paradigm
• History
• Mathematical Formulation
• MPC Basics
Early 1960’s : Rudolf Kalman
• Introduction of the State Space model :
• notion of states as ‘internal memory’ of the system
• states not always directly measurable : ‘Kalman’ Filter !
• afterwards LQR (as the dual of Kalman filtering)
• LQG : LQR + Kalman filter

• But LQG no real succes in industry :


• constraints not taken into account
• only for linear models
• only quadratic cost objectives
Signal processing

• no model uncertainties
Identification

System Theory
Automation 13
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
History
• MPC Paradigm
• History
• Mathematical Formulation
• MPC Basics
During 1960’s : ‘Receding Horizon’ concept

• Propoi, A. I. (1963). “Use of linear programming methods for synthesizing


sampled-data automatic systems”. Automatic Remote Control, 24(7), 837–844.

• Lee, E. B., & Markus, L. (1967). “Foundations of optimal control theory”. New
York: Wiley. :

“… One technique for obtaining a feedback controller synthesis


from knowledge of open-loop controllers is to measure the current
control process state and then compute very rapidly for the open-
loop control function. The first portion of this function is then used
during a short time interval, after which a new measurement of the
function is computed for this new measurement. The procedure is
then repeated. …”
Signal processing
Identification

System Theory
Automation 14
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
History
• MPC Paradigm
• History
• Mathematical Formulation
• MPC Basics
During 1960’s : ‘Receding Horizon’ concept

Signal processing
Identification

System Theory
Automation 15
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
History
• MPC Paradigm
• History
• Mathematical Formulation
• MPC Basics
1970’s : 1st generation MPC
• Extension of the LQR / LQG framework through combination with the
‘receding horizon’ concept
• IDCOM (Richalet et al., 1976) :
• IR models
• quadratic objective
• input / output constraints
• heuristic solution strategy

• DMC (Shell, 1973) :


• SR models
• quadratic objective
• no constraints
Signal processing
Identification • solved as least-squares problem

System Theory
Automation 16
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
History
• MPC Paradigm
• History
• Mathematical Formulation
• MPC Basics
Early 1980’s : 2nd generation MPC
• Improve the rather ad-hoc constraint handling of the 1st generation
MPC algorithms
• QDMC (Shell, 1983) :
• SR models
• quadratic objective
• linear constraints
• solved as a quadratic program (QP)

Late 1980’s : 3rd generation MPC


• IDCOM-M (Setpoint, 1988), SMOC (Shell, late 80’s), …
• Constraint prioritizing
Signal processing • Monitoring / Removal of ill-conditioning
Identification
• fault-tolerance w.r.t. lost signals
System Theory
Automation 17
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
History
• MPC Paradigm
• History
• Mathematical Formulation
• MPC Basics
Mid 1990’s : 4th generation MPC

• DMC-Plus (Honeywell Hi-Spec, ‘95), RMPCT (Aspen Tech, ‘96)


• Graphical user interfaces
• Explicit control objective hierarchy
• Estimation of model uncertainty

Currently (in industry) still …


• … no guarantees for stability
• … often approximate optimization methods
• … not all support state-space models
Signal processing
Identification • … no explicit use of model uncertainty in controller design

System Theory
Automation 18
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
MPC Formulation
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics
Basic ingredients :
• prediction model to predict plant response to future input
sequence
• (finite,) sliding window (receding horizon control)
• parameterization of future input sequence into finite number of
parameters
• discrete-time : inputs at discrete time steps
• continuous-time : weighted sum of basis functions
• optimization of future input sequence
• reference trajectory
• constraints

Signal processing
Identification

System Theory
Automation 19
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
MPC Formulation
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics

Signal processing
Identification

System Theory
Automation 20
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
MPC Formulation
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics
Assumptions / simplifications :

• no plant-model mismatch :
• no disturbance inputs
• all states are measured
(or estimation errors are negligible)
• no sensor noise

. . . seem trivial issues but form essential difficulties


Signal processing
in applications . . .
Identification

System Theory
Automation 21
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Theoretical Formulation (cfr. CACSD)
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics

• future window of length ∞


• impossible to solve, because . . .
• infinite number of optimization variables
• infinite number of inequality constraints
Signal processing
Identification
• infinite number of equality constraints
System Theory
Automation 22
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Formulation 1
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics

• still future window of length ∞, BUT


• quadratic cost function
• no input or state constraints
• linear model
Signal processing
Identification • Optimal solution has the form uk = -Kxk

System Theory
• Find K by solving Ricatti equation → LQR controller
Automation 23
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Formulation 1 : LQR
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics
PRO
• explicit, linear solution
• low online computational complexity

CON
• constraints not taken into account
• linear model assumption
• only quadratic cost functions
• no predictive capacity

Signal processing
Identification

System Theory
Automation 24
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Formulation 2
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics

• changes :
• keep all constraints etc.
• reduce horizon to length N

• solution obtainable through dynamic optimization


Signal processing
Identification • only u0 is applied, in order to obtain feedback at each k about xk

System Theory
Automation 25
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Formulation 2 : Classic MPC
• MPC Paradigm
• History
• Mathematical Form.
PRO
• MPC Basics
• takes constraints into account
• proactive behaviour
• wide range of (convex) cost functions possible
• also for nonlinear models (but convex ?)

CON
• high online computational complexity
• no explicit solution
• feasibility ?
• stability ?
• robustness ?

Signal processing
Identification

• In what follows, we will concentrate on this formulation.


System Theory
Automation 26
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Formulation 3
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics

• linear form of feedback law is enforced


• problem can be recast as a convex (LMI-based) optimization problem

Signal processing
Identification
more on this later . . .
System Theory
Automation 27
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview Summary
• Motivating Example
• MPC Paradigm
Formulation 1
• History
• infinite horizon
• Mathematical Form.
• MPC Basics • no constraints
• explicit solution
• → LQR
Formulation 2
• finite horizon
• constraints
• no explicit solution
• → Classic MPC
Formulation 3
• infinite horizon
• constraints
Signal processing • explicit solution enforced
Identification

• → has elements of LQR ánd MPC


System Theory
Automation 28
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview Open vs. closed loop control
• Motivating Example
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics
• Open loop : no state/output feedback : feedforward control
• Closed loop : state/output feedback : e.g. LQR

MPC is a mix of both :

• internally optimizing an open loop finite horizon control problem


• but at each k there is state feedback to compensate unmodelled
dynamics and disturbance inputs → closed loop control paradigm.
• has implications on e.g stability analysis
• Is of essential importance in Robust MPC, more on this later…
Signal processing
Identification

System Theory
Automation 29
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Standard MPC Algorithm
• MPC Paradigm
• History 1. Assume current time = 0
• Mathematical Form.
2. Measure or estimate x0 and solve for uN and xN :
• MPC Basics

3. Apply uo0 and go to step 1

Remarks :
• : Terminal state cost and constraint

Signal processing
• : some kind of norm function
Identification
• Sliding Window
System Theory
Automation 30
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
MPC design choices
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics
1.prediction model

2. cost function
• norm
• horizon N
• terminal state cost

3. constraints
• typical input/state constraints
• terminal constraint
Signal processing
Identification

System Theory
Automation 31
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Prediction Model
• MPC Paradigm
• History • Input/Output or State-Space ?
• Mathematical Form.
• MPC Basics • I/O restricted to stable, linear plants
• Hence SS-models

• Type of model determines class of MPC algorithm


• Linear model : Linear MPC
• Non-linear model : Non-linear MPC (or NMPC)
• Linear model with uncertainties : Robust MPC
• BUT : MPC is always a non-linear feedback law due to the
constraints

• Type of model determines class of involved optimization problem


• Linear models lead to most efficiently solvable opt.-problems
Signal processing
Identification
• Choose simplest model that fits the real plant ‘sufficiently well’
System Theory
Automation 32
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Cost Function
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics General Cost Function:

Design Functions and Parameters:


1.
2. Horizon N
3. F(x)
4. Reference Trajectory

Signal processing
Identification

System Theory
Automation 33
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Cost Function
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics

Signal processing
Identification

System Theory
Automation 34
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Cost Function
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics

Signal processing
Identification

System Theory
Automation 35
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Cost Function
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics

Signal processing
Identification

System Theory
Automation 36
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Constraints
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics

Signal processing
Identification

System Theory
Automation 37
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Constraints
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics

Signal processing
Identification

System Theory
Automation 38
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Constraints
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics

Signal processing
Identification

System Theory
Automation 39
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Constraints
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics

Signal processing
Identification

System Theory
Automation 40
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Terminal State Constraints
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics

Signal processing
Identification

System Theory
Automation 41
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Reference Insertion
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics

Signal processing
Identification

System Theory
Automation 42
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Reference Insertion
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics

Signal processing
Identification

System Theory
Automation 43
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Reference Insertion
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics

Signal processing
Identification

System Theory
Automation 44
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Reference Insertion
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics

Signal processing
Identification

System Theory
Automation 45
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
Exercise Sessions
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics
• Ex. 1 : Optimization oriented
• Ex. 2 : MPC oriented
• Ex. 3 : real-life MPC/optimization problem

Evaluation
• (brief !) report (groups of 2)
• oral examination → insight !

Signal processing
Identification

System Theory
Automation 46
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
• Overview
• Motivating Example
• MPC Paradigm
• History
• Mathematical Form.
• MPC Basics

Signal processing
Identification

System Theory
Automation 47
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 1 : Introduction
H0K03a : Advanced Process Control
Model-based Predictive Control 2 : Dynamic Optimization

Bert Pluymers
Prof. Bart De Moor
Katholieke Universiteit Leuven, Belgium
Faculty of Engineering Sciences
Department of Electrical Engineering (ESAT)
Research Group SCD-SISTA

[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Overview
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms
Lesson 2 : Dynamic Optimization

• Optimization basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

Signal processing
Identification

System Theory
Automation 1
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Notation
• Optimization Basics
• Convex Optimization
• Dynamic Optimization General form :
• Optimization Algorithms

Legend :

• : vector of optimization variables


• : objective function / cost function
• : equality constraints
• : inequality constraints
Signal processing
Identification • : solution to optimization problem
System Theory • : optimal function value 2
Automation

[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Gradient & Hessian
• Optimization Basics
• Convex Optimization
• Dynamic Optimization Gradient :
• Optimization Algorithms

(points in direction of
steepest ascent)

Hessian :

Signal processing
Identification (gives information about local curvature of )
System Theory
Automation 3
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Gradient & Hessian
• Optimization Basics
• Convex Optimization
• Dynamic Optimization Example :
• Optimization Algorithms

Gradients for different Eigenvectors of hessian


at the origin ( )

Signal processing
Identification

System Theory
Automation 4
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Unconstrained Optimality Conditions
• Optimization Basics
• Convex Optimization
• Dynamic Optimization Necessary condition for optimality of
• Optimization Algorithms

Sufficient conditions for minimum

positive definite

Classification of optima :
positive definite minimum
Signal processing
Identification indefinite saddle point
System Theory negative definite maximum 5
Automation

[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Lagrange Multipliers
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms
Introduction of Lagrange multipliers leads to Lagrangian :

with
Lagrange multipliers of the ineq. constraints
Lagrange multipliers of the eq. constriants

Constrained optimum can be found as

Signal processing
Identification

System Theory Minimization over but Maximization over !!!!


Automation 6
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Lagrange Multipliers
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms
Constrained optimum can be found as

First-order optimality conditions in

Gradient of Gradient of Gradient of


ineq. eq.
Signal processing
Identification
Interpretation ???
System Theory
Automation 7
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Lagrange Multipliers
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

Signal processing
Identification

System Theory
Automation 8
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Lagrange Multipliers
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

Signal processing
Identification

System Theory
Automation 9
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Lagrange Multipliers
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

Signal processing
Identification

System Theory
Automation 10
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Lagrange Duality
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

Signal processing
Identification

System Theory
Automation 11
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Karush-Kuhn-Tucker Conditions
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
From previous considerations we can now state
• Optimization Algorithms
necessary conditions for constrained optimality :

These are called the KKT conditions.


Signal processing
Identification

System Theory
Automation 12
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Optimization Tree
• Optimization Basics
• Convex Optimization
• Dynamic Optimization Optimization
• Optimization Algorithms

discrete continuous

constrained unconstrained

non-convex convex
optimization optimization

Signal processing
Identification
NLP QP SOCP SDP
LP
System Theory
Automation 13
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Convex Optimization
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
An optimization problem of the form
• Optimization Algorithms

is convex iff for any two feasible points :

• is feasible

This is satisfied iff


•`the cost function is a convex function
Signal processing
Identification
• the equality constraints or linear or absent
• the inequality constraints define a convex region
System Theory
Automation 14
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Convex Optimization
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms Importance of convexity :

• no local minima, one global optimum

• under certain conditions, primal and dual have


same solution

• efficient solvers exist


• polynomial worst-case execution time
• guaranteed precision
Signal processing
Identification

System Theory
Automation 15
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
From LP to SDP
• Optimization Basics generality
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

SDP Semi-Definite Programming

SOCP Second Order Cone Progr.

QP Quadratic Programming

LP Linear Programming

computational
efficiency

Signal processing
Identification

System Theory
Automation 16
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Linear Programming (LP)
• Optimization Basics
• Convex Optimization
• Dynamic Optimization General form :
• Optimization Algorithms

Remarks :

• always convex
• optimal solution always at a corner of ineq. constraints

Signal processing
Identification

• typically used in finance / economics / management


System Theory
Automation 17
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Linear Programming (LP)
• Optimization Basics
• Convex Optimization
• Dynamic Optimization Eliminating equality constraints :
• Optimization Algorithms

Reparametrize optimization vector :

Leading to
Signal processing
Identification

System Theory
Automation 18
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Quadratic Programming (QP)
• Optimization Basics
• Convex Optimization
• Dynamic Optimization General form :
• Optimization Algorithms

Remarks :

• convex iff

• LP is a special case of QP (imagine )

• Used in all domains of engineering

Signal processing
Identification

System Theory
Automation 19
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Second-Order Cone Programming (SOCP)
• Optimization Basics
• Convex Optimization
• Dynamic Optimization General form :
• Optimization Algorithms

SOC constraint
Remarks :

• Always convex
• Second-Order, Ice-Cream, Lorentz cone :

Signal processing
Identification •
• Engineering applications with sum-of-squares,
System Theory
Automation robust LP, robust QP 20
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Second-Order Cone Programming (SOCP)
• Optimization Basics
• Convex Optimization
• Dynamic Optimization QP as special case of SOCP :
• Optimization Algorithms

Rewrite this as

which is equivalent to

By introducing an additional variable we get the SOCP

Signal processing
Identification

System Theory
Automation 21
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Semi-Definite Programming (SDP)
• Optimization Basics
• Convex Optimization
• Dynamic Optimization General form :
• Optimization Algorithms

with

Remarks :

• means that should be positive semi-definite


• means that should be pos. semi-def.
• always convex :

• ineq. constraints called LMI’s : Linear Matrix Ineq.


Signal processing
Identification • LMI’s arise in many applications of
Systems & Control Theory
System Theory
Automation 22
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Semi-Definite Programming (SDP)
• Optimization Basics
• Convex Optimization
• Dynamic Optimization convexity :
• Optimization Algorithms

Easily verified :

hence

and therefore

Signal processing
Identification

which means that LMI’s are convex constraints.


System Theory
Automation 23
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Semi-Definite Programming (SDP)
• Optimization Basics
• Convex Optimization
• Dynamic Optimization Schur Complement :
• Optimization Algorithms

is equivalent with

More general :

Remarks :

• Originally developed in a statistical framework


Signal processing
Identification • Today widely used in S&C in order to reformulate
problems involving eigenvalues as an LMI.
System Theory
Automation 24
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Semi-Definite Programming (SDP)
• Optimization Basics
• Convex Optimization
• Dynamic Optimization SOCP as special case of SDP :
• Optimization Algorithms

is equivalent with (by using Schur complement) :

(exercise : apply Schur complement to LMI and


Signal processing
Identification reconstruct SOC constraint)
System Theory
Automation 25
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Convexity = SDP ?
• Optimization Basics
• Convex Optimization
• Dynamic Optimization Formulated as SDP Convex optimization
• Optimization Algorithms
Convex optimization formulatable as SDP

Example :

Signal processing
Identification

System Theory
Automation 26
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Convexity ≠ SDP !
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms convex optimization
convexity difficult to exploit
(computationally)

SDP

SOCP
QP
structure easily exploitable
LP (many toolboxes available)
→ significant efficiency gains !

Signal processing
Identification

System Theory
Automation 27
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
MPC Paradigm
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms
• At every discrete time instant , given information about
the current system state , calculate an ‘optimal’ input
sequence over a finite time horizon :

Signal processing
• Apply the first input to the real system
Identification

• Repeat at the next time instant , using new state


System Theory
Automation measurements / estimates. 28
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Dynamic Programming
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

• Finding the optimal input sequence is done by means


of Dynamic Programming

• Definition* :
“DP is a class of solution methods for solving sequential
decision problems with a compositional cost structure”

• Invented by Richard Bellman (1920-1984) in 1953

Signal processing
Identification

System Theory
Automation 29
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Dynamic Programming
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms Example 1 : The darts problem* :
“Obtain a score of 301 as fast as possible while
beginning and ending in a double.”

http://plus.maths.org/issue3/dynamic/

• Decision : next area towards which to throw the dart


Signal processing
Identification
• Cost : time
System Theory
Automation * D. Kohler, Journal of the Operational Research Society, 1982. 30
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Dynamic Programming
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms Example 2 : DNA sequence allignment :

G AAT T C A G T TA (sequence #1)


G GATC GA (sequence #2)

• Decisions : which nucleotides to match


• Cost : e.g. based on substitution / insertion prob.
Signal processing
Identification • Algorithms : Baum/Welch, Waterman/Smith, …
System Theory
Automation 31
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Dynamic Programming in MPC
• Optimization Basics
• Convex Optimization “Series of sequential decisions” :
• Dynamic Optimization
• Optimization Algorithms
measured

Optimization variables

Typical optimization problem :

Signal processing
Identification

→ standard QP formulation ?
System Theory
Automation 32
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Linear MPC as standard QP
• Optimization Basics
• Convex Optimization Optimization vector :
• Dynamic Optimization
• Optimization Algorithms

Cost function :

For convexity hence .


Signal processing
Identification

System Theory
Automation 33
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Linear MPC as standard QP
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms Equality constraints

with

Sparsity : many entries in equal to 0


Signal processing
Identification

System Theory
Automation 34
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Linear MPC as standard QP
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms Inequality constraints

with

Sparsity : many entries in equal to 0


Signal processing
Identification

System Theory
Automation 35
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Alternatives
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

• slightly faster to solve


• ‘chattering’ : optimal solution jumps around
Signal processing
Identification

System Theory
Automation 36
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Alternatives
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

Signal processing
Identification
Non-convex optimization in general : to be avoided !!!

System Theory
Automation 37
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
MPC and optimization
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

Signal processing
Identification

System Theory
Automation 38
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
MPC and optimization
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

Signal processing
Identification

System Theory
Automation 39
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Active Set Methods
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

Signal processing
Identification

System Theory
Automation 40
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Active Set Methods
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

Signal processing
Identification

System Theory
Automation 41
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Interior Point Methods
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms 1. Choose initial point
2. Linearize KKT conditions around current point
3. Solve Linearized KKT system to obtain search
direction
4. Calculate step length such that
5. Repeat from step 2, until convergence

Signal processing
Identification

System Theory
Automation 42
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Interior Point Methods
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

Signal processing
Identification

System Theory
Automation 43
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Comparison
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms
• ASM allow hot start
• reuse of active set, factorizations, …

• ASM has feasible intermediate results


• by construction

• IPM can exploit sparsity


• solution of KKT system by sparse solver

In industry currently mostly ASM due to first two


Signal processing
Identification advantages, but IPM under consideration…
System Theory
Automation 44
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Comparison
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

Signal processing
Identification

System Theory
Automation 45
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Comparison
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

Signal processing
Identification

System Theory
Automation 46
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Comparison
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

Signal processing
Identification

System Theory
Automation 47
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Comparison
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

Signal processing
Identification

System Theory
Automation 48
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
Conclusion
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

• convex optimization powerful tool !!!


• different optimization algorithm have pro’s and con’s
• try to avoid NLP’s !!!!

Signal processing
Identification

System Theory
Automation 49
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
• Overview
References
• Optimization Basics
• Convex Optimization
• Dynamic Optimization
• Optimization Algorithms

Signal processing
Identification

System Theory
Automation 50
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 2 : Dynamic Optimization
H0K03a : Advanced Process Control
Model-based Predictive Control 3 : Stability

Bert Pluymers
Prof. Bart De Moor
Katholieke Universiteit Leuven, Belgium
Faculty of Engineering Sciences
Department of Electrical Engineering (ESAT)
Research Group SCD-SISTA

[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability


• Introduction
Overview
• Example
• Stability Theory
• Set Invariance
• Implementations
Lecture 3 : Stability

• Introduction
• Example
• Stability Theory
• Set Invariance
• Implementations

Signal processing
Identification

System Theory
Automation 1
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
MPC Paradigm
• Example
• Stability Theory
• At every discrete time instant , given information about
• Set Invariance
• Implementations the current system state , calculate an ‘optimal’ input
sequence over a finite time horizon :

• Apply the first input to the real system


Signal processing
Identification • Repeat at the next time instant , using new state
measurements / estimates.
System Theory
Automation 2
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Optimality of input sequence
• Example
• Stability Theory
• Set Invariance to be applied at
• Implementations

computed at

Signal processing
Identification
Optimal input sequences
System Theory
Automation Input sequence applied to the system 3
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Stability Analysis
• Example
• Stability Theory
• Set Invariance
• Implementations
• Classical Way :
r u x
+
-

linear controller plant

Analyse poles/zeros of and associated


transfer functions.
• Modelbased Predictive Control :

r u x
+
-

MPC controller plant

Signal processing
Identification

System Theory
Automation
Lyapunov theory for stability. 4
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Inverted Pendulum
• Example
• Stability Theory
• Set Invariance
• Implementations

• 1 input :
• 4 states :
Signal processing
Identification
• open loop unstable system

System Theory
Automation 5
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Inverted Pendulum
• Example
• Stability Theory
• Set Invariance
• Implementations

Non-minimum phase
behaviour

Signal processing

• 4 different horizon lengths


Identification

System Theory
Automation
• 3 different MPC variants (to be defined later) 6
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Stability Theory
• Example
• Stability Theory
• Set Invariance
• Implementations • Explicit vs. Optimization-based controller
• Transfer functions → Lyapunov theory

Stability is obtained / proven in 2 steps :

1. Recursive feasibility
i.e. controller well-defined for all k
2. Lyapunov function construction
i.e. trajectories converge to equilibrium

Signal processing
Identification

System Theory
Automation 7
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Stability Theory
• Example
• Stability Theory
• Set Invariance
• Implementations Limited validity of MPC stability framework :

• only for ‘stabilization’ problems :


• initial state
• system steered towards
• no disturbances allowed (but extension possible)

• no general stability framework for ‘tracking’ problems

Signal processing
Identification

System Theory
Automation 8
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Stability Theory
• Example
• Stability Theory
• Set Invariance
• Implementations Recursive Feasibility
If the optimization problem is feasible for time , then it
is also feasible for time .
(and hence for all )

Feasible Region
The region in state space, defined by all states for
which the MPC optimization problem is feasible.

→ Recursive feasibility proven : all states within feasible


region lead to trajectories for which the MPC-controller is
Signal processing
Identification feasible and hence well-defined.
System Theory
Automation 9
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Stability Theory
• Example
• Stability Theory
• Set Invariance
→ Recursive feasibility proven : all states within feasible
• Implementations region lead to trajectories for which the MPC-controller is
feasible and hence well-defined.

feasible region

Signal processing
Identification

System Theory
Automation 10
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
MPC Stability Measures
• Example
• Stability Theory
• Set Invariance
• Implementations

Signal processing
Identification

System Theory
Automation 11
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
MPC Stability Measures
• Example
• Stability Theory
• Set Invariance
• Implementations

recursive feasibility
(terminal constraint) Lyapunov stability
(terminal cost)
Signal processing
Identification

System Theory
Automation 12
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Step 1 : Recursive Feasibility
• Example
• Stability Theory
Problem : given the optimal (and hence feasible) solution
• Set Invariance
• Implementations
to the optimization at time , construct a feasible solution
for the optimization at time .

Signal processing
Identification

System Theory
Automation 13
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Step 1 : Recursive Feasibility
• Example
• Stability Theory
• Set Invariance
• Implementations Given :

To be found :

Observe / Choose :

Signal processing

?
Identification

System Theory
Automation 14
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Step 1 : Recursive Feasibility
• Example
• Stability Theory
• Set Invariance Plant state at time predicted at time :
• Implementations

Real plant state at time :

Assumption : No plant model mismatch, i.e.

Signal processing Hence, reusing the overlapping part of the input sequence
Identification
will also result in an identical state sequence
System Theory
Automation 15
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Step 1 : Recursive Feasibility
• Example
• Stability Theory
• Set Invariance
• Implementations

OK OK ??? ???

Signal processing
Identification

System Theory OK OK OK ???


Automation 16
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Step 1 : Recursive Feasibility
• Example
• Stability Theory
• Set Invariance Condition 1 :
• Implementations

Satisfied if

Condition 2 :
Condition 3 :

How to choose
and ?
Signal processing
Identification

System Theory
Automation 17
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Step 1 : Recursive Feasibility
• Example
• Stability Theory
• Set Invariance How to choose and ?
• Implementations

Assume we know a locally stabilizing controller :

i.e. such that is locally stable.

Then choose

Signal processing
Identification

System Theory
Automation 18
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Step 1 : Recursive Feasibility
• Example
• Stability Theory
• Set Invariance
• Implementations
OK OK OK ???

OK OK OK ???

Condition 2 :
Condition 3 :
Signal processing
Identification

System Theory
Automation 19
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Step 1 : Recursive Feasibility
• Example
• Stability Theory
• Set Invariance
• Implementations
Condition 2 :
Condition 3 :

Since we know that …

Condition 2 is satisfied if

Condition 3 is satisfied if

Signal processing
Identification

System Theory
Automation 20
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Step 1 : Recursive Feasibility
• Example
• Stability Theory
• Set Invariance
• Implementations
Summary & Interpretation

Recursive feasibility is guaranteed if


Terminal constraint is
1) feasible w.r.t state constraints

Terminal constraint is
2) feasible w.r.t input constraints

Terminal constraint is
3) a positive invariant set w.r.t

Signal processing
Identification

System Theory
Automation 21
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Step 2 : Lyapunov stability
• Example
• Stability Theory •Lyapunov stability :
• Set Invariance
• Implementations If such that for some region around 0

then all trajectories starting within asymptotically


evolve towards 0.

•In the MPC context :


• is chosen as the feasible region,
• is chosen as the optimal cost value of the MPC
optimization problem for the given
Signal processing
Identification

Under which conditions is a Lyapunov function ???


System Theory
Automation 22
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Step 2 : Lyapunov stability
• Example
• Stability Theory
• Set Invariance
Under which conditions is a Lyapunov function ???
• Implementations

We have to prove that

Or in other words that

This is satisfied if

Signal processing
Identification

System Theory
Automation 23
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Step 2 : Lyapunov stability
• Example
• Stability Theory
• Set Invariance Special relationship between the two cost expressions :
• Implementations

should be < 0
Satisfied if

Signal processing
Identification
Condition 4 :
System Theory
Automation 24
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Summary & Interpretation
• Example
• Stability Theory Recursive feasibility and asymptotic stability is guaranteed if
• Set Invariance
• Implementations

Terminal constraint is
1) feasible w.r.t state constraints

Terminal constraint is
2) feasible w.r.t input constraints

Terminal constraint is
3) a positive invariant set w.r.t

4)
Lyapunov inequality

i.e. should ‘overbound’ cost of terminal controller

Iff the optimization problem is feasible at time !!!!!

Signal processing
Identification
• conditions are sufficient, but not necessary
System Theory • is only used implicitly !
Automation 25
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Set Invariance
• Example
• Stability Theory
• Set Invariance
• Implementations
“Given an autonomous dynamical system, then a set is
(positive) invariant if it is guaranteed that if the current state
lies within , all future states will also lie within .”

1.5 1.5

1 1

0.5 0.5

0 0

-0.5 -0.5

-1 -1

-1.5 -1.5
-1.5 -1 -0.5 0 0.5 1 1.5 -1.5 -1 -0.5 0 0.5 1 1.5

Signal processing not invariant invariant


Identification

System Theory
Automation 26
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Set Invariance
• Example
• Stability Theory
• Set Invariance
• Implementations
• Useful tool for analysis of controllers for constrained systems
• Example :
– linear system
– linear controller
– state constraints

1.5

0.5

-0.5

-1

Signal processing -1.5


-1.5 -1 -0.5 0 0.5 1 1.5
Identification

‘feasible region’ of closed loop system


System Theory
Automation 27
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Set Invariance
• Example
• Stability Theory
• Set Invariance
• Implementations Consider an autonomous time-invariant system as
defined previously

A set is …

… invariant iff
… feasible iff

Problem :
Given an autonomous dynamical system subject to state
Signal processing
Identification constraints, find the feasible invariant set of maximal size.
System Theory
Automation 28
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Invariant sets for LTI systems
• Example
• Stability Theory
(Gilbert et al.,1991, IEEE TAC)
• Set Invariance
• Implementations

Given an LTI system subject to linear constraints

then the largest size feasible invariant set can be found as

with a finite integer.

• is constructed by simple forward prediction


• can be proven to be the largest feasible invariant set
Signal processing
Identification
• is called the Maximal Admissible Set (MAS)

System Theory
Automation 29
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Implementations
• Example
• Stability Theory
• Set Invariance
• Implementations How to choose such that conditions are
satisfied ?

Different possibilities, depending of


• type of system (linear, non-linear)
• stability of the system
• presence of state constraints
• horizon length
• time constraints during design !

Signal processing
Identification

System Theory
Automation 30
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
• Example
Implementations
• Stability Theory
• Set Invariance
• Implementations

Signal processing
Identification

System Theory
Automation 31
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
• Example
Implementations
• Stability Theory
• Set Invariance
• Implementations

Signal processing
Identification

System Theory
Automation 32
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
• Example
Implementations
• Stability Theory
• Set Invariance
• Implementations

1
2

Signal processing
Identification

System Theory
Automation 33
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
• Example
Implementations
• Stability Theory
• Set Invariance
• Implementations

Signal processing
Identification

System Theory
Automation 34
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
• Example
Implementations
• Stability Theory
• Set Invariance
• Implementations

Signal processing
Identification

System Theory
Automation 35
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
• Example
Implementations
• Stability Theory
• Set Invariance
• Implementations

Signal processing
Identification

System Theory
Automation 36
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
• Example
Implementations
• Stability Theory
• Set Invariance
• Implementations

Signal processing
Identification

Terminal constraint set determines feasible region


System Theory
Automation 37
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Implementations
• Example
• Stability Theory
• Set Invariance
• Implementations

Signal processing
Identification

Solid : standard MPC, dashed : terminal cost, constraint, dotted : terminal equality constr.
System Theory
Automation 38
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
Conclusion
• Example
• Stability Theory
• Set Invariance
• Implementations • stability of standard MPC not guaranteed
• pole/zero analysis impossible
• recursive feasibility
• Lyapunov stability

• general stability framework for stabilization problems


• different implementations
• stability measures allow the use of shorter horizons

Signal processing
Identification

System Theory
Automation 39
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
• Introduction
References
• Example
• Stability Theory
• Set Invariance
• Implementations

Signal processing
Identification

System Theory
Automation 40
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 3 : Stability
H0K03a : Advanced Process Control
Model-based Predictive Control 4 : Robustness

Bert Pluymers
Prof. Bart De Moor
Katholieke Universiteit Leuven, Belgium
Faculty of Engineering Sciences
Department of Electrical Engineering (ESAT)
Research Group SCD-SISTA

[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness


• Example
Overview
• Robustness
• Robust MPC
• Conclusion

Lecture 4 : Robustness

• Example
• Robustness
• Robust MPC
• Conclusion

Signal processing
Identification

System Theory
Automation 1
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Example
• Robustness
• Robust MPC
• Conclusion
Linear state-space system of the form

with bounded parametric uncertainty

Aim : steer this system towards the origin from initial state
without violating the constraint

Signal processing
Identification

System Theory
Automation 2
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Example
• Robustness
• Robust MPC
• Conclusion Results for 4 different parameter settings :

Signal processing
Identification
• Recursive feasibility ?
System Theory
• Monotonicity of the cost ?
Automation 3
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Robustness
• Robustness
• Robust MPC
• Conclusion
Robust with respect to what ?

• Disturbances

Cause predictions of
‘nominal’ MPC to be inaccurate

• Model uncertainty

Signal processing
Identification

System Theory
Automation 4
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Robustness
• Robustness
• Robust MPC
• Conclusion

Main aims :
• Keep recursive feasibility properties, despite model errors,
disturbances
• Keep asymptotic stability (in the case without disturbances)

We need to have an idea about …


• the size of the model uncertainty
• the size of the disturbances

Signal processing
Identification

System Theory
Automation 5
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Uncertain Models
• Robustness
• Robust MPC
• Conclusion

Linear Parameter-Varying state space models with


polytopic uncertainty description

Signal processing
Identification

System Theory
Automation 6
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Uncertain Models
• Robustness
• Robust MPC
Linear Parameter-Varying state space models with
• Conclusion
norm-bounded uncertainty description

Signal processing
Identification

System Theory
Automation 7
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Bounded Disturbances
• Robustness
• Robust MPC
• Conclusion • Typically bounded by a polytope :
• Can be described in two ways

• Trivial condition for well-posedness :

Signal processing
Identification

System Theory
Automation 8
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Robust MPC
• Robustness
• Robust MPC
• Conclusion

Main aims :
• Keep recursive feasibility properties, despite model errors,
disturbances
• Keep asymptotic stability (in the case without disturbances)

Necessary modifications :
•Uncertain predictions (e.g predictions with all models within
uncertainty region)
• worst-case constraint satisfaction over all predictions
• worst-case cost over all predictions

•Terminal cost has to satisfy multiple Lyap. Ineq.


Signal processing
Identification
•Terminal constraint has to be a robust invariant set

System Theory
Automation 9
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Robust MPC
• Robustness
• Robust MPC
• Conclusion
Uncertain predictions :

model uncertainty
disturbances

Signal processing N
Identification

System Theory
Automation 10
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Uncertain Predictions
• Robustness
• Robust MPC
• Conclusion
Step 1) Robust Constraint Satisfaction

Observations :

• depends linearly on

• is a convex polytopic set


• is a convex set

Result : Sufficient to impose constraint only for vert. of :

Signal processing
Identification

System Theory
Automation 11
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Uncertain Predictions
• Robustness
• Robust MPC LTI LPV
• Conclusion
(L=1) (L>1, e.g. 2)

Signal processing
Identification

System Theory
Automation 12
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Uncertain Predictions
• Robustness
• Robust MPC
• Conclusion

Impose state constraints on all nodes


of state prediction tree

Signal processing
Identification
→ number of constraints increases expon. with incr. !!!
System Theory
Automation 13
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Worst-Case Cost Objective
• Robustness
• Robust MPC
• Conclusion
Step 2) Worst-Case cost minimization

Observations :
• depends linearly on

• is a convex polytopic set


• cost function typically convex function of

Signal processing
→ Also for objective function sufficient to make
Identification
predictions only with vertices of uncertainty polytope
System Theory
Automation 14
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Worst-Case Cost Objective
• Robustness
• Robust MPC
• Conclusion

states
inputs

Signal processing
Identification

System Theory
Automation 15
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Worst-Case Cost Objective
• Robustness
(1-norm)
• Robust MPC
• Conclusion

LP

Signal processing
Identification

System Theory
Automation 16
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Worst-Case Cost Objective
• Robustness
(2-norm)
• Robust MPC
• Conclusion

CVX ?

Signal processing
Identification

System Theory
Automation 17
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Worst-Case Cost Objective
• Robustness
(2-norm)
• Robust MPC
• Conclusion

Constraints of the form :


CVX ?

SOC
Signal processing
Identification

System Theory
Automation 18
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Worst-Case Cost Objective
• Robustness
(2-norm)
• Robust MPC
• Conclusion

SOCP

Signal processing
Identification

System Theory
Automation 19
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Robust MPC
• Robustness
(2-norm)
• Robust MPC
• Conclusion By rewriting we now get

SOCP

Signal processing
Identification
Terminal constraint
System Theory
Automation
Terminal cost 20
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example Robust Terminal Cost
• Robustness
• Robust MPC
• Conclusion
“non-robust” stability condition for terminal cost:

In case of…
• LPV system with polytopic uncertainty
• linear feedback controller
• quadratic cost criterion
• quadratic terminal cost

… this becomes :

Signal processing or equivalent :


Identification

System Theory
Automation 21
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example Robust Terminal Cost
• Robustness
• Robust MPC
• Conclusion
Robust stability condition for terminal cost:

Observations :

• inequality is convex and linear in and (i.e. LMI in )


• is a convex polytopic set

Hence, inequality satisfied iff

Signal processing
Identification

System Theory
Automation 22
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example Robust Terminal Cost : Design
• Robustness
• Robust MPC
• Conclusion
1. Find a robustly stabilizing controller

2. Find a terminal cost satisfying

by solving the following optimization problem :

Minimization of
eigenvalues of SDP
Signal processing
Identification
optimization variables

System Theory
Automation 23
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example Robust Terminal Constraint
• Robustness
• Robust MPC
• Conclusion

Reminder : nominal case

Recursive feasibility is guaranteed if


remain unchanged
Terminal constraint is
1) feasible w.r.t state constraints

Terminal constraint is
2) feasible w.r.t input constraints

Terminal constraint is
3) a positive invariant set w.r.t

Has to be modified in order to


Robust positive invariance
Model uncertainty into account
Signal processing
Identification

System Theory
Automation 24
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example Robust Terminal Constraint
• Robustness
• Robust MPC
• Conclusion
Consider linear terminal controller ,

then the resulting closed loop system is :

Robust positive invariance :

Again : sufficient to satisfy inclusion


Signal processing
Identification

System Theory
Automation 25
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example Robust Terminal Constraint
• Robustness
• Robust MPC
• Conclusion

Reminder : invariant sets for LTI systems

Given an LTI system subject to linear constraints

then the largest size feasible invariant set can be found as

with a finite integer.

Signal processing
Comes down to making forward predictions using
Identification

System Theory
Automation 26
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example Robust Terminal Constraint
• Robustness
• Robust MPC
• Conclusion
LTI LPV
(L=1,n=2) (L>1, e.g. 2, n=2)

Signal processing
Identification

System Theory
Automation 27
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Ellipsoidal invariant sets for LPV systems
• Robustness
(Kothare et al.,1996, Automatica)
• Robust MPC
• Conclusion

1
X
0.8 S
0.6

0.4

0.2

-0.2

-0.4

-0.6

-0.8

-1

-0.8 -0.6 -0.4 -0.2 0 0.2 0.4 0.6 0.8

• Constructed by solving semi-definite program (SDP)


• Conservative with respect to constraints
Signal processing
Identification

System Theory
Automation 28
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Polyhedral invariant sets for LPV systems
• Robustness
• Robust MPC
• Conclusion
Reformulate invariance condition :

A set is invariant with respect to a system defined


by iff

with

Sufficient condition :

Signal processing
Identification

Also necessary condition


System Theory
Automation 29
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Polyhedral invariant sets for LPV systems
• Robustness
• Robust MPC
• Conclusion Algorithm :

• Initialize
• iteratively add constraints from to until

Advantages :

• in step 2 only ‘significant’ constraints are added to :

Signal processing
Identification
significant insignificant
System Theory
Automation 30
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Polyhedral invariant sets for LPV systems
• Robustness
• Robust MPC
• Conclusion Algorithm :

• Initialize
• iteratively add constraints from to until

Advantages :

• prediction tree never explicitly constructed

• given a polyhedral set , it is straightforward


to calculate :

Signal processing
Identification

System Theory
Automation 31
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Polyhedral invariant sets for LPV systems
• Robustness
• Robust MPC
• Conclusion Example
Initialization

Signal processing
Identification

System Theory
Automation 32
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Polyhedral invariant sets for LPV systems
• Robustness
• Robust MPC
• Conclusion Example
Iteration 10

Signal processing
Identification

System Theory
Automation 33
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Polyhedral invariant sets for LPV systems
• Robustness
• Robust MPC
• Conclusion Example
Iteration 10 + garbage collection

Signal processing
Identification

System Theory
Automation 34
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Polyhedral invariant sets for LPV systems
• Robustness
• Robust MPC
• Conclusion Example
Iteration 20

Signal processing
Identification

System Theory
Automation 35
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Polyhedral invariant sets for LPV systems
• Robustness
• Robust MPC
• Conclusion Example
Final Result

Signal processing
Identification

System Theory
Automation 36
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Polyhedral invariant sets for LPV systems
• Robustness
• Robust MPC
• Conclusion Example
Final Result

Signal processing
Identification

System Theory
Automation 37
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Polyhedral invariant sets for LPV systems
• Robustness
• Robust MPC
• Conclusion Example

Signal processing
Identification

System Theory
Automation 38
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Recursive feasibility, stability guarantee ?
• Robustness
• Robust MPC
• Conclusion
Open loop Closed loop
optimal input sequence optimal input sequence

Signal processing
Identification

System Theory
NO recursive feasibility !!! Recursive feasibility
Automation 39
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Example revisited…
• Robustness
• Robust MPC
• Conclusion Results for 4 different parameter settings :

Signal processing
Identification
• Recursive feasibility ?
System Theory
• Monotonicity of the cost ?
Automation 40
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Example revisited…
• Robustness
• Robust MPC
• Conclusion

Signal processing
Identification

System Theory
Automation 41
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness
• Example
Conclusion
• Robustness
• Robust MPC
• Conclusion

• Robustness w.r.t a) bounded model uncertainty


b) bounded disturbances
• necessary modifications :
• worst-case constraints satisfaction
• worst-case objective function
• terminal cost
• terminal constraint
• “open-loop” vs. “closed-loop” predictions

→ currently hot research topic !

• convex optimization but problem size impractical


Signal processing
→ currently hot research topic !
Identification

System Theory
Automation 42
[email protected] H0k03a : Advanced Process Control – Model-based Predictive Control 4 : Robustness

You might also like