1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online...

34
1 Chapter 2 Chapter 2 1. 1. Parametric Parametric Models Models

Transcript of 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online...

Page 1: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

11

Chapter 2Chapter 2

1.1. Parametric ModelsParametric Models

Page 2: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

22

Parametric ModelsParametric Models

The first step in the design of online parameter identification (PI) algorithms is to lump the unknown parameters in a vector and separate them from known signals, transfer functions, and other known parameters in an equation that is convenient for parameter estimation.

In the general case, this class of parameterizations is of the form

where is the vector with all the unknown parameters and

are signals available for measurement.

We refer it as the linear "static "parametric model (SPM).

Page 3: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

33

Parametric ModelsParametric Models

The SPM may represent a dynamic, static, linear, or nonlinear system.

Example:

where x, u are the scalar state and input, respectively, and a, b are the unknown constants we want to identify online using the measurements of x, u .

Page 4: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

44

Parametric ModelsParametric Models

Page 5: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

55

Parametric ModelsParametric ModelsAnother parameterization of the above scalar plant is

In the general case, the above parametric model is of the form

Page 6: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

66

Parametric ModelsParametric Models

Where are signals available for measurement and

is a known stable proper transfer function, where q is either

the shift operator in discrete time (i.e., q = z) or the differential

operator (q = s) in continuous time. We refer to this model as the

linear "dynamic"parametric model (DPM).

The importance of the SPM and DPM is that the unknown parameter

vector appears linearly.

So we refer to SPM and DPM as linear in the parameters

parameterizations.

Page 7: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

77

Parametric ModelsParametric Models

We can derive SPM from DPM if we use the fact that is a

constant vector and redefine to obtain

In a similar manner, we can filter each side of SPM and DPM using a

stable proper filter and still maintain the linear in the parameters

property and the form of SPM, DPM. This shows that there exist an

infinite number of different parametric models in the form of SPM,

DPM for the same parameter vector .

Page 8: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

88

Parametric ModelsParametric Models

In some cases, the unknown parameters cannot be expressed in

the form of the linear in the parameters models. In such cases the

PI algorithms based on such models cannot be shown to converge

globally. A special case of nonlinear in the parameters models for

which convergence results exist is when the unknown parameters

appear in the special bilinear form

bilinear static parametric model (B-SPM)

or

bilinear dynamic parametric model (B-DPM)

Page 9: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

99

Parametric ModelsParametric Models

bilinear static parametric model (B-SPM)

bilinear dynamic parametric model (B-DPM)

where are signals available for measurement

at each time t, and are the unknown parameters.

The transfer function is a known stable transfer function.

Page 10: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

1010

Parametric ModelsParametric Models

state-space parametric models (SSPM)

In some applications of parameter identification or adaptive control of plants of the form

whose state x is available for measurement, the following parametric model may be used:

where is a stable design matrix; are the unknown matrices; and are signal vectors available for measurement. The model may be also expressed in the form

Page 11: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

1111

Parametric ModelsParametric Models

state-space parametric models (SSPM)

It is clear that the SSPM can be expressed in the form of the

DPM and SPM.

Page 12: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

1212

Parametric ModelsParametric Modelsbilinear state-space parametric models (B-SSPM).

Another class of state-space models that appear in adaptive control is of the form

where B is also unknown but is positive definite, is negative definite, or the sign of each of its elements is known.

The B-SSPM model can be easily expressed as a set of scalar B-SPM or B-DPM.

Page 13: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

1313

Parametric ModelsParametric ModelsPI ProblemPI Problem

For the SPM and DPM:

Given the measurements , generate , the estimate

of the unknown vector , at each time t. The PI algorithm

updates with time so that approaches or converges

to . Since we are dealing with online PI, we would also expect

that if

changes, then the PI algorithm will react to such changes and

update the estimate to match the new value of .PIPI

Page 14: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

1414

Parametric ModelsParametric ModelsPI ProblemPI Problem

For the B-SPM and B-DPM:

Given the measurements generate estimates

respectively, at each time t the same way as

in the case of SPM and DPM.

PIPIl

z

z

( )

( )

t

t

*

*

( )

( )

t

t

Page 15: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

1515

Parametric ModelsParametric ModelsPI ProblemPI Problem

For the SSPM::

Given the measurements generate estimates

of , (and hence the estimates

, respectively)at each time t the same way as in the case of SPM

and DPM.

PIPIx

u

ˆ( )

ˆ

Tm

T

A A

B

Page 16: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

1616

Parametric ModelsParametric ModelsPI ProblemPI Problem

The online PI algorithms generate estimates at each time t, by

using the past and current measurements of signals. Convergence

is achieved asymptotically as time evolves.

For this reason they are referred to as recursive PI algorithms to

be distinguished from the non-recursive ones, in which all the

measurements are collected a priori over large intervals of time

and are processed offline to generate the estimates of the unknown

parameters.

Page 17: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

1717

Parametric ModelsParametric ModelsExample 1: Mass-Spring-Dashpot SystemExample 1: Mass-Spring-Dashpot System

Let us assume that M, f, k are the constant unknown parameters that we want to estimate online.

express in the form of SPM

Page 18: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

1818

Parametric ModelsParametric ModelsExample 1: Mass-Spring-Dashpot SystemExample 1: Mass-Spring-Dashpot System

Measurements:

To avoid of derivatives , we filter bothsides with the stable filter

Page 19: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

1919

Parametric ModelsParametric ModelsExample 1: Mass-Spring-Dashpot SystemExample 1: Mass-Spring-Dashpot System

Another possible parametric model is:

Page 20: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

2020

Parametric ModelsParametric ModelsExample 2: Cart with two inverted pendulumsExample 2: Cart with two inverted pendulums

1:y

Page 21: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

2121

Parametric ModelsParametric ModelsExample 2: Cart with two inverted pendulumsExample 2: Cart with two inverted pendulums

Page 22: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

2222

Parametric ModelsParametric ModelsExample 2: Cart with two inverted pendulumsExample 2: Cart with two inverted pendulums

To avoid of derivatives , we filter bothsides with the stable filter ,

Page 23: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

2323

Parametric ModelsParametric ModelsExample 2: Cart with two inverted pendulumsExample 2: Cart with two inverted pendulums

If in the above model we know that is nonzero, redefining the

constant parameters as we obtain the

following B-SPM:

where,

Page 24: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

2424

Parametric ModelsParametric ModelsExample 3: second-order autoregressive moving Example 3: second-order autoregressive moving average (ARMA) modelaverage (ARMA) model

This model can be rewritten in the form of the SPM as:

where,

Page 25: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

2525

Parametric ModelsParametric ModelsExample 3: second-order autoregressive moving Example 3: second-order autoregressive moving average (ARMA) modelaverage (ARMA) model

If one of the constant parameters, i.e., , is nonzero. Then we can obtain a model of the system in the B-SPM form as follows:

where,

Page 26: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

2626

Parametric ModelsParametric ModelsExample 4: nonlinear systemExample 4: nonlinear system

Filtering both sides of the equation with the filter , we can express the system in the form of the SPM

where,

Page 27: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

2727

Parametric ModelsParametric ModelsExample 5: Example 5: dynamical system in the transfer function formdynamical system in the transfer function form

where the parameter b is known and a, c are the unknown parameters. Rewrite it as:

where,

divide to

Page 28: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

2828

Parametric ModelsParametric ModelsExample 6: Example 6: DPM modelDPM model

If we want W(s) to be a design transfer function with a pole, say at we write:

Page 29: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

2929

Parametric ModelsParametric ModelsExample 6: Example 6: DPM modelDPM model

Page 30: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

3030

Parametric ModelsParametric ModelsExample 7: Example 7: SSPM modelSSPM model

where, 1 1 11 12 11 12

21 22 21 222 2

, , ,x u a a b b

x u A Ba a b bx u

where,

Page 31: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

3131

Parametric ModelsParametric ModelsExample 8: Example 8: n-n-th order-SISO LTI systemth order-SISO LTI system

where,

Page 32: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

3232

Parametric ModelsParametric ModelsExample 8: Example 8: n-n-th order-SISO LTI systemth order-SISO LTI system

Filtering by

where,

Page 33: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

3333

Exercises Exercises

From reference 1, chapter 2,

Choose 5 problems from 8 problems.

Page 34: 1 Chapter 2 1. Parametric Models. 2 Parametric Models The first step in the design of online parameter identification (PI) algorithms is to lump the unknown.

3434

END