mechatronics at the university of calgary: concepts and applications jeff pieper

Mechatronics at the University of Calgary:

Concepts and Applications

Jeff Pieper

Outline Mechatronics Past and Present Research Results Undergraduate Program Directions for the Future Summary

What is Mechatronics? Synergistic combination of mechanics,

electronics, microprocessors and control engineering Like concurrent engineering Does not take advantage of inherent

uniqueness available Control of complex electro-mechanical

systems Rethinking of machine component design by

use of mechanics, electronics and computation Allows more reconfigurablility

What is Mechatronics Design example: timing belt in

automobile Timing belt mechanically

synchronizes and supervises operation

Mechatronics: replace timing belt with software

Allows reconfigurability online

What is Mechatronics Second example: Active

suspension Design of suspension to achieve

different characteristics under different operating environments

Demonstrates fundamental trade-offs

What is Mechatronics Low tech, low cost, low performance: pure

mechanical elements: spring shock absorber Can use dynamic systems to find coefficients via

time or frequency domain mid tech, cost, performance: electronics: op amps,

RLC circuits, hydraulic actuators Typically achieve two operating regimes, e.g.

highway vs off-road High tech, cost , performance: microprocessor-

based online adjustment of parameters Infinitely adjustable performance

*

Research Projects Discrete Sliding Mode Control with Applications Helicopter Flight Control OHS Aircraft Flight Control Magnetic Bearings in Papermaking Systems Robust Control for Marginally stable systems Process Control and System ID Controller Architecture

*

Theme of research Control of

Uncertain Nonlinear Time-varying Industrially relevant Electro-mechanical systems

This is control applications side of mechatronics

Involves sensor, actuator, controller design

What is control? Nominal performance

Servo tracking Disturbance rejection

Low sensitivity Minimal effects of unknown aspects Non-time-varying

Feedback Control*

Discrete Time Sliding Mode Control

Comprehensive evaluation of theoretical methodology

Optimization, maximum robustness

Applications: Web tension system Gantry crane

Web Tension System

Gantry Crane*

Helicopter Flight Control Experimental Model Validation Model-following control design Experimental Flight Control

Multivariable Start-up and safety

Helicopter Flight Control

Helicopter Flight Control Model-following vs. stability

Conflicting Multi-objective

Controller Optimization Order reduction System validation

Helicopter Flight Control*

OHS Aircraft Flight Control Outboard Horizontal Stabilizer Non-intuitive to fly Developed sensors and actuators Model identification and validation Gain-scheduled adaptive control Reconfigurable controller based

upon feedback available

OHS Aircraft *

0 1 2 3 4 5

-4

-2

0

2

4V

eloc

ity (

m/s

)

0 1 2 3 4 5-25

-20

-15

-10

-5

0

5

10

15

20

25

Ang

le o

f Atta

ck (

deg)

0 1 2 3 4 5-30

-20

-10

0

10

20

30

Pitc

h (d

eg)

0 1 2 3 4 5

-100

-75

-50

-25

0

25

50

75

100

Pitc

h R

ate

(deg

/s)

0 1 2 3 4 5-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

Time (s)Time (s)

Alti

tude

(m

)

0 1 2 3 4 5-10

-8

-6

-4

-2

0

2

4

6

8

10E

leva

tor

Def

lect

ion

0 1 2 3 4

-4

-2

0

2

4A

irspe

ed (

m/s

)

0 1 2 3 4-25

-20

-15

-10

-5

0

5

10

15

20

25

Ang

le o

f Atta

ck (

deg)

0 1 2 3 4-30

-20

-10

0

10

20

30

Pitc

h (d

eg)

0 1 2 3 4-100

-75

-50

-25

0

25

50

75

100

Pitc

h R

ate

(deg

/s)

0 1 2 3 4-2.0

-1.5

-1.0

-0.5

0.0

0.5

1.0

1.5

2.0

Time (s)Time (s)

Alti

tude

(m

)

0 1 2 3 4-10

-8

-6

-4

-2

0

2

4

6

8

10E

leva

tor

Def

lect

ion

(deg

)

15 m/s20 m/s

Magnetic Bearings in Papermaking Systems System identification for magnetic

bearings Nonlinear, unstable system Closed loop modeling Control Design using Sliding Mode

methods and state estimators Servo-Control of shaft position

Magnetic Bearings

Magnetic Bearings in Papermaking Systems Papermaking system modeling Use of mag bearings for tension

control actuation and model development

Control Design and implementation issues

Compare performance with standard roller torque control

Papermaking Tension Control

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.50.9

0.95

1

1.05

1.1

1.15

1.2

1.25

1.3

1.35

1.4The H-infinity norm of different order apprx system

H-in

finith

nor

m o

f tra

nsfe

r fun

ctio

n

Cross-sectional Area of web material

2nd order approx 4th order approx 6th order approx 10th order approx

0 0.05 0.1 0.15 0.2 0.25 0.3 0.35 0.4 0.45 0.50.8

1

1.2

1.4

1.6

1.8

2

2.2

2.4

2.6The H-infinity norm of transfer function of PID controller by changing velocity

The

H-in

finity

nor

m o

f tra

nsfe

r fun

ctio

n

Cross-sectional Area of web material(in.2)

*

Robust Control via Q-parameterization

Ball and beam application Stability margin optimization:

Nevanlinna-Pick interpolation

x

Experimental Results Large lag Non-minimum

phase behaviour Friction, hard

limits

Robust Stability Margin*

Delay Nominal Optimal0 -0.27 -0.300.1 -0.26 -0.280.2 -0.19 -0.220.3 -0.07 -0.110.4 0.06 0.010.5 0.17 0.100.6 0.26 0.180.7 0.34 0.240.8 0.39 0.290.9 0.44 0.331 0.48 0.371.1 0.51 0.401.2 0.54 0.421.3 0.56 0.441.4 0.58 0.461.5 0.60 0.47

Process Control and System ID: Quadruple-tank Process Diagram

PUMP1 PUMP2

Tank1 L1

Tank3 L3

Tank2 L2

Tank4 L4

Out1Out2 Out1Out2

G11

G12

G21

G22

System ID: Model Fitting

Tank2 process model with input u1

Tank4 process modelwith input u1

Control Techniques Classic PI control Internal Model Control (IMC) Model Predictive Control (MPC) Linear Quadratic Regulator

(LQR) Optimal Control

MPC Control Results

Different set points changes to test MPC controller performance

Tank2 response Tank4 response

Performance Comparison*

PI IMC MPC LQR

MP(%) 4 20 5 32

ts (sec) 31 99 50 210

ss e(%) 1.5 0 0 0

||e||2 2.7 0.8 0.45 0.3

||e|| 0.2 0.08 0.05 0.05

trc(sec) 34 102 47 76

ControllersParameters

Controller Architecture Given conflicting and redundant

information Design controller with best

practical behaviour Good nominal performance Robust stability Implementable

*

Undergraduate Program: Mechatronics Lab with Applications Developed lab environment for teaching and

research

Two courses: - linear systems, - prototyping

Hands-on work in: modeling

system identification

Sampled-data systems

Optical encoding

State estimation

Control design

Mechatronics Lab with Applications

Multivariable fluid flow control system Two-input-two-outputs Variable dynamics Model Predictive Control Chemical Process Emulation

Fluid Flow Control

Quanser Product

Mechatronics Lab with Applications

Ball and Beam system Hierarchical control system

Motor position servo Ball position

Solve via Q-parameterization Robust stability and optimal nominal

performance

Ball and Beam System

Quanser Product

Mechatronics Lab with Applications

Heat Flow Apparatus Unique design Varying deadtime for challenging

control Demonstrate industrial control

schemes PID controllers Deadtime compensators

Heat Flow Apparatus *

Quanser product

Directions for the Future Robust Performance

robust stability and nominal performance simultaneously guaranteed

Multiobjective Control Design Systems that meet conflicting, and disparate

objectives Performance evaluation for soft measures

Fuzzy systems

Bottom line: Mechatronics *

Summary Control applications in electro-

mechanical systems Emphasis on usability Complex problems from simple

systems Undergraduate program: hands-on

learning and dealing with systems form user to end-effector