Linear System and Random Process Final PhanLeNhatTan MEEIU23007

Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1of 11

111Equation Chapter 1 Section 1VIETNAM NATIONAL UNIVERSITY –

HOCHIMINH CITY
INTERNATIONAL UNIVERSITY
SCHOOL OF ELECTRICAL ENGINEERING

FINAL EXAM: LINEAR SYSTEM AND


RANDOM PROCESS

BY
PHAN LÊ NHẬT TÂN

A REPORT SUBMITTED TO THE SCHOOL OF ELECTRICAL ENGINEERING

HO CHI MINH CITY, VIET NAM


2024
TABLE OF CONTENTS

TURNITIN DECLARATION..............................................................................................i

TABLE OF CONTENTS....................................................................................................ii

LIST OF TABLES.............................................................................................................iii

LIST OF FIGURES............................................................................................................iv

ABBREVIATIONS AND NOTATIONS............................................................................v

ABSTRACT.......................................................................................................................vi

CHAPTER I INTRODUCTION.........................................................................................7

I.1. Background research................................................................................................7

I.2. Dataset......................................................................................................................8

CHAPTER II METHODOLOGY.....................................................................................10

II.1. Overview system...................................................................................................10

II.2. Preprocesing..........................................................................................................11

II.3. Segmentation.........................................................................................................11

II.4 Classification..........................................................................................................11

CHAPTER III RESULTS..................................................................................................13

III.1. Preprocessing.......................................................................................................13

III.2. Segmentation........................................................................................................13

III.3. Classification........................................................................................................14

CHAPTER IV CONCLUSION AND RECOMMENDATION........................................15

i
LIST OF TABLES

Table 1. Dataset open access on internet.............................................................................8


Table 2. Result accuracy score of these model (PCA)......................................................14

ii
LIST OF FIGURES

Figure 1. Section view of the heart......................................................................................7


Figure 2. The overview system..........................................................................................10
Figure 3. Heart sound cycle...............................................................................................11
Figure 4. Preprocessing signal...........................................................................................13
Figure 5. Segmentation signal...........................................................................................14

iii
ABBREVIATIONS AND NOTATIONS
F (x, t): moving source
H(∙): Heaviside step function
δ (∙ ) : Dirac delta function
v: source velocity
p(t): source magnitude
G[∙ ¿ : Greens’ function
u[∙ ¿: response function of linear systems
h[∙ ¿ : impulse response function
S(∙ ¿ : power spectral density
R(∙ ¿ : correlation function
E[∙ ¿ : expectation
2
σ : variance
ψ: standard deviation

iv
CHAPTER I

INTRODUCTION

Paper: Dynamic response of linear systems to moving stochastic sources


Authors: + L. Sun (Department of Civil Engineering, The university of Texas at Austin.
+ B.S. Greenberg (Department of Management Science & Information System,
The university of Texas at Austin.
Publish: Journal of Sound and Vibration (2000)
Problem:
22EQUATION SECTION (NEXT)
CHAPTER II

QUESTION FOR FINAL

What are the objectives of the paper?


 The objective of the paper is to provide a theoretical foundation for the study of the dynamic
response of linear systems to moving stochastic sources.
 The paper aims to analyze the stochastic characteristics of the response of linear systems to
moving stochastic sources based on a time-convolution expression.
 The authors aim to show that the random response of a linear system under a moving
stationary stochastic source becomes a non-stationary process, which requires a different
approach than commonly used spectral analysis.
 The paper introduces a follow-up spectral analysis procedure to overcome the limitations of
spectral analysis for non-stationary processes.
 The authors provide statistical characteristics of the dynamic response in fixed and follow-up
coordinates.
 The theory developed in the paper can be universally applied to the moving source problem
for linear systems.
Determine the Linear System proposed/used in the paper.
*With deterministic source:

v
- Suppose a linear differential operator describes the dynamic property of a physical
system. Different systems possess different dynamic properties and operators. For instance, the
operator of a specialized elastodynamic problem is given by the well-known Navier-Stoke's field
equations, Green's function is then defined as the fundamental solution of the system. In other
words, for the problem discussed in the paper. Green's function corresponds to the solution
of the system governing equations as the point source takes the form of a Dirac delta
function in both spatial and temporal domains.
According to the causality for a realistic system, Green's function G (x, t) = 0 for t < 0, we may
write. O[G(x−ξ , t−τ )]=δ (x−ξ )δ(t−τ)

Here, the initiation of the source is delayed by τ . Therefore, the physical law requires that for
Green's function, t ⩽ τ . The properties of the medium under consideration are accounted for in
the operator O , while approximate interface conditions relate the field quantities in layers with
different media. Equation (5) therefore applied to the classes of the configuration that are
mentioned in the beginning of this paper. Since the steady-state solution can be derived from the
transient solution, we initially need only to analyze the suddenly applied moving load.

We introduce a linear integral operator I as

t
I =∫ ❑∫ ❑ f (ξ , τ ) p (τ ) d ξ d τ
0 D

where the kernel f describes a spatial amplitude distribution within finite domain D in co-
ordinate ξ=(ξ ,η , ζ ), and where the kernel p is a transient source signature with p(t )=0 for t ⩽ 0
. The response of a linear system to a finite source distribution with arbitrary signature is
obtained by applying the integral operator to both sides of equation (5). Assuming that the
operator is bounded, we may interchange the order of the integral operator and the O . After
carrying out the integration we find

t
u(x , t)=∫ ❑∫ ❑ f (ξ , τ )G(x−ξ ,t−τ)d ξ d τ
0 D

vi
where u(x , t) represents the displacement field of the medium and the upper limit of the time
convolution is determined by the time of interest.

For more information here is two equation for transient response and the steady response:

and

t
u(x , t)= ṕ ∫ ❑ h ( x−vτ , y , z−z h , t−τ ) d τ
0

respectively.

t
u(x , t)= ṕ ∫ ❑ h ( x −vτ , y , z−z h , t−τ ) d τ
−∞

*With stochastic source condition.

Analyze the Linear System.

Results of the paper

vii
CHAPTER III

RESULTS

III.1. Preprocessing.

Figure 1. Preprocessing signal


Figure 4 show the picture subplot figure of the state preprocessing signal.
III.2. Segmentation
Figure 5 show the interval of the segmentation.

viii
Figure 2. Segmentation signal
III.3. Classification
Table 2 show the result of classification.
Table 1. Result accuracy score of these model (PCA)

Model 20 Features 10 Features (PCA)


DT 0.70 0.81
KNN 0.77 0.77
RD 0.79 0.78
MLPC 0.78 0.75
SVM 0.75 0.71
NB 0.63 0.63

ix
CHAPTER IV

CONCLUSION AND RECOMMENDATION

A plan to classification heartsound is challenge and based on the machine learning I have
to use in ski-learn without more tune coefficient the best score is DecisionTree with accuracy
score 0.81 with PCA.
In addition, this report is as quick as it can so the result can be improved more by change
the coefficient or change the method.

You might also like