Introduction to MATLAB

MATLAB Toolboxes
Computer Applications in CEE
Drs. Trani and Rakha
Civil and Environmental Engineering
Virginia Polytechnic Institute and State University

Spring 2000

Virginia Polytechnic Institute and State University

1 of 31

Sample Toolboxes

MATLAB has been extended over the years to respond to

the needs of various users

Several toolboxes exist to add to the power of the original

language. For example:
-

Simulink
Fuzzy Logic
Neural Networks
Optimization
Controls
C/C++ compiler library
Real-time workshop

Virginia Polytechnic Institute and State University

2 of 31

Simulink

Simulink is a powerful toolbox to solve systems of

differential equations

Simulink has applications in Systems Theory, Control,

Economics, Transportation, etc.

The Simulink approach is to represent systems of ODE

using block diagram nomenclature

Simulink provides seamless integration with MATLAB.

In fact, Simulink can call any MATLAB function

Simulink interfaces with other MATLAB toolboxes such

as Neural Network, Fuzy Logic, and Optimization
routines

Virginia Polytechnic Institute and State University

3 of 31

Simulink Building Blocks

Simulink has a series of libraries to construct models

Libraries have object blocks that encapsulate code and

behaviors

Connectors between blocks establish causality and flow

of information in the model

Virginia Polytechnic Institute and State University

4 of 31

Simulink Interface

The main application of Simulink is to model continuous

systems

Perhaps systems that can be described using ordinary

differential equations

Virginia Polytechnic Institute and State University

5 of 31

Typical Simulink Libraries

Shown are some typical Simulink libraries (Macintosh)

Virginia Polytechnic Institute and State University

6 of 31

Sample Simulink Library (Windows and UNIX)

The new Simulink interface in Windows/UNIX uses
standard OOP interfaces (Visual C++, Visual Basic)

Virginia Polytechnic Institute and State University

7 of 31

Simulink Example
The following example illustrates the use of Simulink to
solve a two vehicle car-following problem. This problem
has been studied for the past 40 years in traffic flow
theory

x( t + ) fc = k ( x( t ) lc x( t ) fc )
where: k is a gain constant of the response process

x( t + ) fc is the acceleration of the following vehicle

x( t ) lc is the speed of the leading vehicle
x( t ) fc is the speed of the following vehicle
Virginia Polytechnic Institute and State University

8 of 31

Set-Up of the Problem

Assume the velocity profile of the leading vehicle is

known (we drive this car to test the response of the
following vehicle)

Initially, assume no time lags in the acceleration response

of the following vehicle ( = 0 )

Test an emergency braking maneuver executed by the

first vehicle at 3 m/s2

Test a new scenario with a deterministic time lag

response time of 0.75 seconds

Verify that both cars do not collide

Virginia Polytechnic Institute and State University

9 of 31

Simulink Representation of the Car-Following

Problem
Car Following Problem
Mux
Speed Profiles

Mux1

Xdot

s
Integrator

s
Integrator1

Mux
Mux

Distance Profiles

Speed LTU
2
k (Xdot_ltu - Xdot_ftu)

Xdot_ltu - Xdot_ftu

Constant

1
s
Integrator2

Virginia Polytechnic Institute and State University

10 of 31

Car-Following Model Output

Velocity profiles for leading and trailing vehicles
Speed (m/s)

Time (s)

Virginia Polytechnic Institute and State University

11 of 31

Car-Following Model Output

The time space diagram below illustrates an emergency
braking maneuver for the leading vehicle
Space (m)

Time (s)

Virginia Polytechnic Institute and State University

12 of 31

Car-Following Model with Delay

A pure transport delay block is added to the original
model simulating the transport lag dynamics of a manmachine system
Mux
Speed Profiles

Mux1
Xdot
X

s
Integrator

s
Integrator1

Mux
Mux

Distance Profiles

Transport
Delay
Speed LTU
Xdot_ltu - Xdot_ftu
k (Xdot_ltu - Xdot_ftu)
1
Constant
1
s
Integrator2

Virginia Polytechnic Institute and State University

13 of 31

Output of the Car-Following Model with Delay

The output suggests that a collision occurs with

=0.75s.

Space (m)

Time (s)
Virginia Polytechnic Institute and State University

14 of 31

Brief on C/C++ Compiler

Converts MATLAB M-Files to C and C++ code

Using this toolbox all graphics and math code can be

converted to develop standalone applications

Typically requires MATLAB compiler and the MATLAB

C/C++ library

New features in version 5.3 also convert cell and

structure arrays

My limited experience with this toolbox suggest

substantial gains in speed if no vector operations have
been implemented in the code

Vector operations show very little improvement when the

code is compiled
Virginia Polytechnic Institute and State University

15 of 31

Neural Network Toolbox

Input
a3

P
W1
RxQ
Z1xR
1

a1
n1

W2

Z1xQ

F2
n2

Z2xZ1

, a2

W3

Z2xQ
Z3xZ2

F3
n3

Z3xQ

Z1xQ F1
b1
Z1x1

b2

Z2xQ
Z2X1

b3

Z3xQ

Z3x1

Provides 100+ functions to simplify the training,

generalization and implementation of artificial neural
networks

The functional implementation of the ANN toolbox is

just through a series of MATLAB functions

All ANN functions are execued from the command line

Virginia Polytechnic Institute and State University

16 of 31

Relevant Functions Provided

Layer initialization functions (Nguyen-Widrow)

Learning functions (gradients, perceptron, Widrow-Hoff)

Analysis functions (max. learning rate, error surface)

Line search functions (Golden section, backtraking)

Network functions (competitive, feed-forward wth

backpropagation)

Performance functions (mean absolute error, SEE)

Training functions (BFGS, Bayesion regularization,

Fletchel-Powell, Lavenberg-Marquardt, etc.)

Transfer functions (log signmoid, tangent sigmoid, etc.)

Virginia Polytechnic Institute and State University

17 of 31

Example of ANN Using MATLAB

Problem:

To estimate aircraft fuel consumption based on aircraft

performance parameters

Implementation on fast-time simulation models

(SIMMOD, RAMS, etc.)
Solution

Prototype model using MATLABs neural Network

Toolbox

Verify the accuracy of the model with the current stateof-the art

Virginia Polytechnic Institute and State University

18 of 31

Sample Data Presented in Flight Manual

Virginia Polytechnic Institute and State University

19 of 31

Modeling Approach
Training Data Set

Learning Module
Import Data

Data Normalization

Forward
Propagation

Input learning pairs

Compute
Error
Initialize connection
Weights

Configuration
Parameters

Change and
Update weights
YES

Error<
eg

NO
Number
of cycles > me
eg - maximum allowable error
me- maximum allowable iterations

NO

YES
End of learning

Virginia Polytechnic Institute and State University

20 of 31

Selected ANN Topology

n

1
1

sum
1
b
1
w
1,1

Target

f1

1
1

1,1

2
1

sum

f2

2
b

1
f1

sum

f2

sum

sum

f1

sum

f2

sum

f1

sum

f2

2
1

Altitude
Mach

sum
f1

f2

Initial
Weight

sum

ISA Cond.

sum

f1

sum

f2

sum

f1

sum

f2

1
w
8,4

sum

f1

sum
n

sum

1
8

1
8

2
8,8

3
1

Fuel burn

f3

3
1

2
8

2
8

f2

2
8

Where f1 - log-sigmoid transfer function

f2 - tansigmoid transfer function
f3 - purelin transfer function

Virginia Polytechnic Institute and State University

21 of 31

ANN Data
The data sets used to develop and train the ANN are
shown below. Generalization of the ANN was conducted
using another (random) data set
Flight Phase
Takeoff and Climbout
Climb to Cruise Altitude
Cruise
Descent

Number of Testing Points

Nor applicable (linear regression used instead)
850 (Fuel)
850 (Distance)
805
1210 (Fuel)
140 (Distance)

Virginia Polytechnic Institute and State University

22 of 31

Summary of Results (Fokker F100)

The results shown in the table illustrate the accuracy of
the model
Flight Phase
Climb

Distance

Fuel

Cruise Specific
Air Range

Mean
Error (%)

Standard
Deviation (%)

Null Hypothesis
(t-test at = 0.01 )

0.377
1.026

0.305
0.190

Accept
Accept

-0.034

0.334

Accept

Virginia Polytechnic Institute and State University

23 of 31

Flight Phase
Descent

Distance

Fuel

Mean
Error (%)

Standard
Deviation (%)

Null Hypothesis
(t-test at = 0.01 )

1.760
1.423

1.860
1.177

Accept
Accept

Virginia Polytechnic Institute and State University

24 of 31

Sample Results (Climb Fuel)

The results of training an ANN with 3 layers are shown

below
6000
Actual Data

5000

Climb Fuel (lb)

Estimated Data

4000
3000
2000
1000
0

10

15

20

25

30

35

40

Pressure Altitude (kft)

Virginia Polytechnic Institute and State University

25 of 31

Absolute Error of ANN

The errors of the ANN are very small as depicted in this
graph for SFC
160
Complete flight envelope

140

0.60 < Mach < 0.75

Frequency

120

10,000 ft < altitude < 37,000 ft

58,000 lb < weight < 98,000 lb

100

805 data points

80
60
40
20
0
-1.5

-1

-0.5
0
0.5
Specific Range Error (%)

Virginia Polytechnic Institute and State University

1.5

26 of 31

Application of the Model to Complete Flight Plans

The ANN model developed has been applied to a generic
flight trajectory generator with very accurate results
Constant heading
segments
Flight plan way-points

Altitude (kft)

30

Climb

Descent

20
DFW

10

29.5

Pseudo-globe circle route

0
80 MIA

28
27.5

85
90

Longitude (deg)

29
28.5

27
26.5

95
100

Latitude (deg)

26

Virginia Polytechnic Institute and State University

27 of 31

Complete Flight Plan Correlation

Cruise Flight
Level (FL)

Distance
(nm) /
Time (hr)

Flight Manual
Fuel Burn
(lb)

Neural Net
Fuel Burn (lb)

Percent
Difference (%)

ROAaMDWb

280

448 / 1:08

6,457

6,546

1.37

310

448 / 1:10

6,360

6,330

0.46

MIAc-DFWd

310

972 / 2:24

11,851

11,865

0.12

350

972 / 2:13

11,510

11,544

0.29

290

352 / 0:57

5,298

5,260

0.71

330

352 / 0:58

5,343

5,429

1.61

290

518 / 1:20

6,990

7,047

0.80

330

518 / 1:21

7,009

7,082

1.04

Flight

ROA-LGAe
ATLf-MIA

a. ROA - Roanoke Regional Airport (Virginia)

b.MDW - Midway Airport (Illinois)
c. MIA - Miami International (Florida)
d.DFW - Dallas-Forth Worth International (Texas)
e. LGA - Laguardia Airport (New York)
f. ATL - Atlanta Hartsfield International Airport (Georgia)

Virginia Polytechnic Institute and State University

28 of 31

Sample Source Code

% Initialize Weights and Biasis
nns = 8; % Number of Neurons in each layer
nns2 = 8;
%********************************
%
For Climb Distance **
%********************************
[W31_cb_d,b31_cb_d,W32_cb_d,b32_cb_d,W33_cb_d,
b33_cb_d ]=initff(P1_cbd,nns,'logsig',nns2 ...
,'tansig',T1_cbd,'purelin');
% Taining of the neural networks using LavenbergMarquardt Alogrithm

Virginia Polytechnic Institute and State University

29 of 31

[W31_cb_d,b31_cb_d,W32_cb_d,b32_cb_d,W33_cb_
d,b33_cb_d ]= trainlm(W31_cb_d,b31_cb_d,'logsig' ...
,W32_cb_d,b32_cb_d,'tansig',W33_cb_d,b33_cb_d,'pure
lin',P_cbd,Ta_cbd,tp);
%********************************
%
For Climb Fuel
**
%********************************
[W31_cb_f,b31_cb_f,W32_cb_f,b32_cb_f,W33_cb_f,b3
3_cb_f ]=initff(P1_cbf,nns,'logsig',nns2,'tansig' ...
,T1_cbf,'purelin');
% Taining of the neural networks using LavenbergMarquardt Alogrithm
[W31_cb_f,b31_cb_f,W32_cb_f,b32_cb_f,W33_cb_f,b3
3_cb_f ]=
trainlm(W31_cb_f,b31_cb_f,'logsig',W32_cb_f ...
Virginia Polytechnic Institute and State University

30 of 31

,b32_cb_f,'tansig',W33_cb_f,b33_cb_f,'purelin',P_cbf,
Ta_cbf,tp);

Virginia Polytechnic Institute and State University

31 of 31