0% found this document useful (0 votes)
4 views47 pagesLagrange&Hamiltonian Module 1
This document introduces the concepts of Lagrangian and Hamiltonian formulations in classical mechanics, focusing on the mathematical formulation and solutions of dynamical systems. It discusses the classification of dynamical systems based on constraints and degrees of freedom, and introduces generalized coordinates to simplify complex mechanical problems. The principle of virtual work is also presented as a method to analyze systems in equilibrium without the direct influence of constraint forces.
Uploaded by
115 Christy Alex martinCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
0% found this document useful (0 votes)
4 views47 pagesLagrange&Hamiltonian Module 1
This document introduces the concepts of Lagrangian and Hamiltonian formulations in classical mechanics, focusing on the mathematical formulation and solutions of dynamical systems. It discusses the classification of dynamical systems based on constraints and degrees of freedom, and introduces generalized coordinates to simplify complex mechanical problems. The principle of virtual work is also presented as a method to analyze systems in equilibrium without the direct influence of constraint forces.
Uploaded by
115 Christy Alex martinCopyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
/ 47
CHAPTER 1
LAGRANGIAN AND.
HAMILTONIAN FORMULATIONS
OF CLASSICAL MECHANICS
INTRODUCTION
Mechanics is a branch of Physies, subdivided into statics and
dynamics, dealing respectively with physical objects at rest and in
‘motion under the influence of exteral and internal forees. Mechanics
had developed since ancient times based on the observation of the
motion of particles. Is knowa that the study ofa dynamieal system
involves two basie steps the frst one is is mathematical formulation,
fe. obtaining & tovering the motion of the system and the
Second one is finding its solution, Sir Isaac Newton was the first
petson to be suscessfl inthis by introducing his famous three laws of
motion that works well in most of the situations that we encounter.
Mechanics based on Newton's laws and their consequences is known
asthe Newtonian or vetoria siechancs. In this femaism, the quantities.
involved ave vectors like vslovty, ores, aoosleraton, ete. But, when
we encounter more complex physial situations involving unknown
ternal frees, a direct application ofthe Newton's equations becomes
reasingly dificult and practically impossible, In this case, a super
‘and a much mote powerful scheme was developed by D’Alembert,
Lasrange, Hamilton, et. known a8 the analytical mechanies, where
the quantities involved are scalars, like work, energy, etc, Vectarial
‘nd analytical mechanics together consti what is known as the classical
mechanies. In this chapter, we diseuss some basic aspects of the
Lagrangian and the Hamiltonian methods. But, before we proceed,
‘we will diseuss some basic framework necessary for the study.
REES OF FREEDOM
Consider a particle P moving freely ina three dimensional space. To deseribe its motion, we require a set of three coordinates like
(Cartesian) or r, 8, 6 (Sphercal polar) etc as illustrated in the jy?
pincceasinas fem
‘
’
fen2)
eee
4
?
Here the pati fre to ence
5 y, 0 independent
sufficient in numb
onfiguration ofa dynamical systann ca
necessary t0 specify the
Of freedom of the system,
ied henner af degaes
moving independly, each of them rec tet oF tree ents
vn eis ato tee
Seda) td 2, which ger cone sae
inates or six degrees of freedor Te pocia
‘can be extendec mths no he
snl ided to any nuinber of particles h cn retie
“cid neon, for syste
ment ina these-dimensonal space
ieee, But, in man
the motion of particles si
[erations rao
ined motion, which
h wich is discussed ney
uations of pra
ete
mW) ——— eee
CONSTRAINED MOTION
{A constrained motion is one which takes place in a restricted
manner. For example, the gas molecules in a container are constrained
by the walls to move only inside it. The beads of an abacus are
constrained to move in one dimension by the supporting wie. A rigid
body, where the distance hetween any two particles is # constant, is
“nother example ofa constrained system, In many cases, a constrained
motion is due to the surface of other bodies as in the ease of the
tmetion of a billiard bal ona table oF the movement of a train on the
tal. In situations where constraints ae involved the object in motion
fave so restricted that its coordinates and or velocity components are
required to satisfy certain relations at every instant of time called
‘Constant equations. Consraint equations ean be some equations oF
some inequalities.
Physically, any constained motion is realized by some forces
“which arise when the object in motion isin contact witha constraining
Surface ora curve, The forces involved are called constraint forces,
‘which ae usualy stiffelasie in nature and appear atthe surface of
contact because a motion initiated by an extemal forees is hindered
by the contact. The sole effect of a constraint foree is to keep the
constraint equations satisfied, The forces are so strong that they do
not allow the object under consideration to deviate even slightly from
the preseribed path
1m Newtonian formalism, all mechanical problems are dealt using
the Newton's second law. According to this la, for a system of NV
particles, which ae interacting, the equation of motion is given by
+R
o
where, mand 7 represents the mass and acceleration of the system,
Fis the exteral force on the particle and F isthe force on the
particle doe tothe particle, which infact isthe constraint fore.
‘Nevonian formalism requires a complete knowledge ofall the forces
acting on the system, external a§ well as intemal, for solving the
system. Systems for whic the forees acting are all external, which
fare known, in general, the above method works well, But, systems
Where the motions are restricted by some constraint forces, the use ea
“ *_cxsssent oxo CUNT Mechs
of Newton's equations bees limited the Fores of consi
inven pte ae nn th problem ca il Be sae at
the Newtonian procedure. Ht if the consraint forces ae ung
the Newtonian approach hevomes incapable and we have To gp
ltcate options Tor solving the sytem. These inestigitins jy
‘ese in Lagrangian and Halon foals in which he eg
‘of motion are ween without any reference toa patcularco-aeinge
sytem. The coordinate system independent approach bypasses he
Aificuis caused by the constants the sluion of mechancy
broblems. But, before we proeed, we give a classification of the
sitet type of cosas hat we ly come aos in mechani
systems, which led to a varity of dynamical systems as dscseg
elo
(CLASSIFICATION OF DYNAMICAL SYSTEM
Constraints are casted ina ferent ype based on for jg
titer, namely: whether they ar tine dependent or tne indepen
Whether they ar intepabi alpha lations among the coordinates
oF nonintgrable ons, weber they are conservative or dissipative
nd wheter they are alate equations or alsbrae inequalities
Depending om thi we hae an eight lait forthe deem
'ypes of dynamical systems. Note tht the classification nat an
sole one because tase afe many property that are common
mong the different sree tems
(0 Seteronomie system : In this ese, the onstins present
‘ot depend on tne explicitly A costa flaton in th ese of
the orm
Ber Pa Ponta) = 4 @)
where ri fo: 75. ae he soon of the pails and ei a
constant. An example of scleronomi system ia simple pendulum
witha rigid suppor. ln ths ease, the distance of @ mass Howie
point of suspension (F) rena constant in tie according to te
elation r= whore cms, .,
i) Rheonomic system + In this ease, the consiin relation
pend on time explicitly and canbe therefore expend ae
Heit Pot es ° A pendula of variable length in which the constant ation wil
be ofthe form i) 0. is eheonomi one
Gi) Motoromie system + Hooomic systems are tose which
se or can be mae indepenlent of velotes a ven by eution (2)
and (). A pendulum with righ support, «pendolum of variable
length anda egil body ave few examples of halonoic stems. The
Inolonomie constant arid body isexpressed
bond
where, 7 and stands for the coordinates of the Han partic
respectively and, ives forthe distance between them. which i @
(iv) Now-Lolonomie system : Non olonomi systems are tose
in which the constraint relations are not olonomic, Le they depend
fon elocies. An example ofa nonholonomic consi
FOR AD 6 “
In this case, the term 1, sands fora velocity component I the
seocty te absent or can be inerated to make it veloity fe,
Js sill hlononie Inthe case ofa spe rolling ona plone without
sling the constraint i nowholononse. Fora eye eng 09a
Feta for
(0) Conservative system + In this ease th total nechaical
energy is comerved while performing 3 constrained motion and we
iy tht the constant foros donot do ay work In 8 coserative
{ystems the forces are derabl fom a potetal energy fiction.
the ese ofa shore rlling without sip, a he pi contact is
tut sliding the tonal forces do noo any work an te sytem fs
(i) Non-conserative system + Here, the constant Forces dre
istic work nl he ll echancal energy et conserved
“dry the motion If we consider the ease of simple penal with
ft siding a cil ack, the constant forces ine tion
iehich do word The penal loses its energy coatinuouly and the
‘usm ic dive Rhowonne stems ae genely aewenseratve veto re expres ne
Mee Fy at onl Eau
‘a ey ca. Ten 8
es ie ee
‘both the forward ane
the form (2). (3) ad (
deformable body (oppesce
int surf
fen] = 70
kono, bilateral and dsipatve
ras ne the ems eatin
. ies. The general form of a
isshconoms ok
ni) Vnlateral system
ane sapessed inthe Form of nea
trie conse is
Flirt tehponnns fo D2OL o
The cas of a gas molecule enclosed in a chamber isan example
of aatera stem,
«the consti of igi eonsenatve, seroma, holonemie
nba A single pendlum witha iid Soppor is another example
‘fa matem possessing a conservative, seleronomic, holonomie and
Bitrl contain, Butte constraint in pendulum of variable lent
‘Sdsiative heoome, holon and bilateral
Stems which ae seleronomi,holonomie and conservative are
cad sinple systems whl others are elle yeeral systems,
GENERALIZED COORDINATES
‘Based onthe sone dino, lata, conssinsinedice
‘wo pes offline sleton of mechanel problems. Fine
thy mike th sori, dependent, they connect coordinates
‘An immediate result 5 reduction inthe number of degsees of
Fiesdom ofthe system. Secondly in most of the eases tht ve ea
ih the esa oes re nomaly unknown: For example ai a
‘cs mentioned above th festa he wie exerts onthe Reno
!mabnes or te force tha he wal of he container exerts ana
Inclesuls are nko. Ths meas that, Newton
tela te nppopate abe aed here, Wha
fomardto a dierent itd of sporach in which ge
er exer on te pas
' equations, whic now the censtaint forces, nse bypass ts ffs in a erent
‘ay sing concept calle vial work
In the ese of holoomic systems, the Fist difficulty mentioned
hove is soved by inticing a set of new coordinates called the
seneaid cooediats, To epi the oigin a ature of eralized
oorinates We conser yxem of V partes ina thre dinero
Space having consent exuations ating om the syst, TH ste
is then represented by a set of SN goonies (ry, Ps P39)
‘with & contenint equations ating between hem We ure these
‘justo to eliminate Ko the 3 cordinaes so hat we ar et with
balance of n= 3k cordinat, which are all independent. The
Aegroes of feedom of the spem iz now reduced 1 3N-k The
‘limiation ofthe corns i estivey equivalent the notion
of a set oF SNK ne independent coordinates ff en de
‘which ae the generalized cordate, We may dine the peered
‘coordinates he independent cordial ne cape
ot speciying the configuration oa dynamica stem completly. A
iinporanee tthe generalized coordinate that they 40 86 depend
tna particular coordinate sytem. Also, they need ot hive the
timensons of ng ntl hey on be angi or sane ther eavenient
‘quantities.
Sine th two vs of cordate, i the generalized covdinates
(Gu do th cores Hecanah aml the Dependent cordists
rg cones tthe sae Sytem, the coordinates ithe
{vo sstems are incon of each other so ha we ea rte
hes FN Hen dD
Fy Pai ron te
rw de
« Aon 85 )
which ean be writen
= Enlapd 6 |
Equation (6 nfot epee he ansfomation of Seto var
‘roe new set g, We may saver io olan any 988 Turton
1 and ime mat be ated tat, he tasformation eis ge
Comsat in them impli
To ilastrate how genealzed cordiats are evolve inst,
of imterest. we considera particle moving onthe surface of «Spe
“The motion of the pales restrited by the constraint equation
4. where oi the ras ofthe sphere. We can ws this equation
liminater from the set of tee coordinates r, atl 30 that we
Se lett with 8 and, whch are the generalized coordinates ofthe
‘Siem. Another example can be «simple pend oscillating in
‘erucl plane, with he consin equation r= where i the lengh
‘ofthe pendulum Here the consrant equation reduces one degre
feeom and the coordinate remaiaing Hs which the generalizes
‘ow, coming book to the the cond diily introduce by the
constrains, Le. the Fores of constant are ot know api the
Staton is overcome by formulating the problem in such away that
the eects caused by the fores of onsaat disappear white saving
the system. This is done by inducing the concept of vir
Gsplacements. which willbe discussed next, We then have (0 det
‘rth only the known extra fortes as the sonsitnt forces dono
Come ie the pitre we solves syste
PRINCIPLE OF VIRTUAL WORK:
Consider a system of ' paris ina thse dimensional space
Latte yum be subjected oan hatin ep lsemen cnet
withthe forts and consaison Resse aay nat Asse
Sat ce change tthe cnigvaitn ofthe so tent auf
witha change of time Ths mesnetha te depscemen not acta
Sing which the eterna ce nthe const feces change
Ths diplocanent the vival acenen Sine thee sno seal
Jnution of he sat the lem wl tl ein equim and the
Torew on each price sro
fe. reo
Let br, denote the virwal displacement ofthe # particle. Since LAGRANGIAN AND HAMILTONIAN FORMULATIONS 8
the foree F, is zero, it is obvious that the work done by this force
during the virtual displacement 6; is also zero
ie, F,.6r, = 0 0)
Now, the sum of these vanishing products over all particles must
likewise be zero
ar
ie, 8
0 (8)
The resultant force F; acting on the i particle is composed of
‘two parts; an external force Ff and an internal constraint force fj
ie, R= Fey,
‘Substituting this result in equation (8), we get
LF + f).55 = 0
or LAF.85 +E F,.5n = 0 09)
Let us now restrict ourselves to systems for which the net virtual
work done by the forces of constraints is zero
ie, Lh87 = 0 (10)
Substituting ths result in equation (9), we get
Las
° ay
“Bauation (11) states that the virtual work done by the applied
force vanishes fora system in equilibrium, provided no frictional forces
are present. Ths statement is called the principle of Virwal work
To illustrate how the concept of @ virtual work is capable in
avoiding the effect of constraint force, we consider a particle, moving
fon the surface of a sphere that is smooth (no friction). In this case,
the constraint force is the reaction of the surface. The virtual
displacement is taken at right angles to the direction ofthe force, ie.
along the surface, so that the work done by the constraint foree
during the virtual displacement is zero. Remember, if frictional forces
are involved in the system, some work will be done by it because the
frictional forces act ina direction opposite to that of the displacement,
“This is the reason why we avoid fictional forces from our discussions. os PRINCIPLE,
a aver etre tc iia
omer in mechanics a dy
‘ymamic stem, Ft
ws scorn a. We hg
D’ALEMBE!
Equation (1
most of the
Spates. Let's ext
‘Bam system in ili
Ay
1 particle. Equation (y
i te ces in he yea le lb
the a fce Fas a vere
me tion (8) or ami sem
ahere, pis the momenta IME
simply ea
tinder a Force
forse =. We may now mod
Lens n
‘extemal and constraint Free,
ow, separating te force i
equation (13) bevomes
Ler fin onde =O
oe LAHSA DAS
We new restr oursles to ystems for whic the vital 9
ton tt Frcs of contains 1 2 Then, the above eon
DU po = ay
fe D* Alember's principle which i
al work applicable for a
Equation (14) is krown 3
the miifed form ofthe principle of virtua
yeamie system.
vere onwards, since we conse’ ony the effects caused By th
exe a orca, we drop the suprsi“€"fom the above equation
and write tas
ZU Ads 0 as)
30
“The D’Alember’ principles in terms ofthe vial displacements
es, We now transform it 10
lof the dependent set of coordi LAGRANGIAN AND HAMILTONIAN FORMULATIONS. nw
LAGRANGIAN AND HAMILTONIAN FORMULATIONS A.
expression involving virtual displacements of the generalized coordinates,
:Which’are independent of one another. The procédure then leads to a
set of new independent second order differential ‘equations called the
‘ Lagrange’s equations, which will be discussed next.
) LAGRANGE’S EQUATIONS FROM D’ALEMBERT’S
PRINCIPLE,
)
To begin with, we consider a system of N particles in a three
™ dimensional space with k constraint relations between the coordinates.
* The relations also involve time explicitly (rheonomic and holonomic
constraints). We are now left with m = 3N-k coordinates, which are
3) all independent. Let rand g, represent the dependent and the generalized
set of coordinates, respectively and m, the mass of the # particle
, The relation between 7; and gis given by (ecuation (6))
n= Lngyl
7
Now, using this equation, the velocity transformations can be
obtained as
a,
rk wt a
on
ae
= yay r
99) ar oy
4)
where, dj represents the time derivative of q;, which is called the
is generalized velocity.
4 Using equation (6), we ean also obtain an expression for the
virtual displacement as
he yoy ata
on = 25, 84+ a 1)
faq)
E Here, the second term on the right hand side vanishes because in
a vinual displacement there is no involvement of time
an
et ie. &, = Lz ba, (18)
an : 7a; tg
2 CLASSICAL AND QUANTUM wee,
uation (infact relates the virtua displacements in he
(and the generalised sets of coordinates (4). We now ste
‘equation (18) in the D*Alambert's principle, which gives
SEA) 5, EEA ge
ow EEA GL Ay -EEA GEM, ~0
te, EEAGE A = EDA
rial a
‘Note tha equation (19) ints of he vu displacement
eneralized coordinates. To proceed furter, we evaluate the Li
‘and HS. of the above equation separately.
us « BE 2
36
30.4, 4
were, EAE repent te? P ocean of i pl
foe is bos at just a he pri corts ne a
have the dimensions of engi, the generalized force Q ood the
the dimio o f Bu he guaty sl ve
inensio of work
‘valuting the RES. we get
nuts. = S55. a, LAGRANGIAN AND HAMILTONIAN FORMULATIONS
There exists deeper relationship between eyeic coordinate
andthe ppnical nature of motion ofthe sytem. When a coordinate >
rmrere ermine Leon
suerte nt sain on
eon emcees
pienso ieee ce
Same panacea sameny Irs mics ot
Scraper atc oom ote re
Stee anne erent
eee
APPLICATIONS OF LAGRANGE'S EQUATION sexe
‘Toillstrate the Lagrangian method, we discuss some fai,
cases lle linear Ramone olan simple pendulum an
Planetary matic, whose solaton are own bythe Newtonian mh
LINEAR HARMONIC OSCILLATOR
A burmonicosiliator sa simple system, that is sleroomis
‘onsevative and holon. near baronet, in it empl
form, consists ofa spring and a mase which can move on a sig)
line, say ais a shown
Lets mprsens the displacement, the velocity and m the ma
ofthe oscilla: The mass i atacted towards the eign by a esi
force F> kx, where isthe spring constant, The kine ener
— —omm-= LAGRANGIAN AND HAMILTONIAN FORMULATIONS: 19
LAGRANGIAN AND HAMILTONIAN FORMULATIONS ___ ag
the oscillator is given by
1
= me G9)
r
‘The potential energy is given by
Ve [rar ‘ite
0 é
12
= =k? +constant
2
If we choose the potential energy of the system at the origin as
zero, then L
Le
es
2 (40)
Now, using equations (39) and (40), the Lagrangian of the oscillator
is given by
12 dpe
ay = Ami? Le
L=T-V = pm? > 4D)
Here we have only a single co-ordinate x. The Lagrange's equation
forxis
a()-Gi)
age) (Ge) = 9 (42)
Using equation (41), we have
aL aL
ms and ~k
ae oe
Substituting these results inthe Lagrange’s equation, we get
d
Lomiytky = 0
aon)
or m+ kx =0
ek
which gives, ¥+->x = 0 (43)
which is the Lagrange’s equation of motion for a linear harmonic
oscillator. Here x represents the acceleration of the system, eS
AND QUANTUM HEC
standard
Comparing equation (83)
Aner erence Le, 20% = 96 gh ge
fork angle) ae
‘ B
i “4
‘Therefore, the frequen ofthe oscilla is given by
G
1
I> ele Ws
SIMPLE PENDULUM,
A smple pendula consists of «small but heavy bo suspend
ftom rigid support by snonextensble, ight and flexible sa
The distance between the conte of suspension of the endulm ee
the centre of mass of the bob is conideed asthe length of
endl «
Assume tht we ave simple pendulum of length J with bbl
masa shown nthe ire rao Let @ be the angular ipa
‘ofthe pendulum ram the uliom postion
From he igo th dsplacemcat ofthe ba even by
b= 10 LAGRANGIAN AND HAMILTONIAN FORMULATIONS 24
LAGRANGIAN AND HAMILTONIAN FORMULATIONS Wat
Therefore, the velocity of the bob is 10, where, 6 is the angular
velocity of the pendulum. Now, the kinetic energy of the pendulum ig
given by
Sa
T= 5m => mi (46)
The potential energy at the displaced condition is given by
V = mg.CA
mg (OA - OC)
mg (1-1 cos 8)
= mal (1 — cos 8) AAD)
Here, we have assumed that the potential energy is zero at the
‘equilibrium position.
‘Now, using equations (46) and (47),
the Lagrangian of the pendulum
is given by
L
Tv
: pmta ~me_l(\ cos) 48)
Here, we have onl
ly a single coordinate @ The Lagrange’s equation
for 0 is given by
4 (ab)_ ae
ai\ad)~aa = © 49)
Using the expression for the Lagrangian (equation (48), we have
au au
ee =
a Pb and Fm meting
ee
Substituting these results in He Lagrange's equation, we get
a
Sime, sin
im 6) + mgl sin 8 = 0
or mP-6 + mgt sin 6 = 0
which gives, 6+£sing=
ich gives, B+ Esin d= 0 (50) ee
ves ap mnt cin ptm
as£e =0 43)
7
she 2
“av "
PLANETARY MOTION
“oe third example that we consider is the case of pact
‘mote. In planetary motion, «pate is moving under 4 ony
fore, whose magnitude is inversely peopotonl fo the square of
sina of the parle rom th cent offre (say Oh poh
a
reg “
where is constant, The sem i conservative andthe motion ¢
fice toa plane. Fr coneninc, we he plans poi eee
and Oto describe he motion ofthe particle ws shown Lat hone
o
Te Mois ew fhe pili planar cudnt ge
Tee
T= deseo) oa LAGRANGIAN AND HAMILTONIAN FORMULATIONS
LAGRANGIAN AND HAMILTONIAN FORMULATIONS 4 _23
‘The potential energy is given by
(55)
Using the values of T and V from equations (54) and (55),
respectively, we get the Lagrangian as
2)
ye (56)
Lig
= sm +r
L= ame
Since we have two variables here, iv., r and @ we have two
equations of motion, one for r and.the other for @.
‘The Lagrange’s equation for r is
4 (a) a
dt\ ar) ar
From the Lagrangian (equation (56)), we have
an.
oF
437)
me and mr
or
Substituting these results in equation (57), we get,
ae pak
Lom)-mr P+
a” ee
k
or mF-mr P+ = 0
r
‘The Lagrange’s equation for is
4 (28). og
dt 0
‘Agaia from the Lagrangian, we have
aL 26
So = id
og mre a i
24 CLASSICAL AND QUANTUM MECH4y
in the Lagrange’s equation, we get, tages
re prine
Gime) = 0 40
7 the e
or mr? §+2mrid = 0 that
which gives, 734286 = 0 aa!
‘The expression for the Lagrangian (equation (56)) and the equatig spac
for 8 (equation (59)) lead to an important conclusion, Here, we kno, for
that, the Lagrangian is independent of the coordinate 8 ie., g Mat!
cyclic. This requires that, the momentum conjugate to @ ie, th syste
angular momentum (m7? 0) of the particle about the centre of attracting
should be conserved, which is given by the equation (59).
CONFIGURATION SPACE,
The instantaneous configuration ofa sytem of particles can
represented by the values of m generalized coordinates qi qa.)
+ Gm in a Cartesian n-dimensional hyperspace called the
configuration space where the 4,’s form the m coordinate axes. AY
time passes, the state ofthe system changes, anda representa
point of the system, called the system point, moves in the configuration“
Space tracing outa curve nown asthe path of motion ofthe system
Note thatthe configuration space has no neeesary connection wi
any physical thee dimensional space. Similarly, the path of mtion
the configuration space hs no resemblance 1o/a pa in space of aah
of the system at an instant of time, ey
He
HAMILTON’S PRINCIPLE |
We have derived the Lagrange’s equations by considering the He
instantaneous state of a system and small virtual displacements about this we
instantaneous state, ie., from a “differential principle” known as the D
Alembert’s principle in a previous session. We can also obtain Lagrange’s
‘equations from a principle that considers the entire motion of the
system in the configuration space between two times f, and f andar
small virtual variations of the entire motion from the actual motiom ac
i.e., from an “integral principle”. The principle is called Hamilton's) th
eg ee ee oma LAGRANGIAN AND HAMILTONIAN FORMULATIONS 25
principle, which is discussed next.
Hamion’s principle is stated for monogenic systems for which
the external forces are derivable from a generalized scalar potential
that may be a function of coordinates, velocities and time. The principle
can be stated as, “of all possible paths, along which a dynamical
system may move from one point to another in the configuration
Space within a given interval of time, the actual path followed is that
for Which the time integral of the Lagrangian is an extremum”.
Mathematically, the principle can be expressed as, the motion of the
system from time ¢; to time 1 is such that the line integral
J= fia (60)
js an extremum for the actual path of the motion. Note that, in most of
the applications in dynamics, one obtains a minimum condition. The
integral given by equation (60) is known as the action integral
In terms of a variation of the fine integral at an instant of time
called Svariation (to be discussed next), the principle can also be
stated as
(61)
with zero variations at ¢, and fp.
DERIVATION OF LAGRANGE’S EQUATIONS FROM
HAMILTON'S PRINCIPLE,
We can show that Lagrange’s equations follow directly from the
Hamilton’s principle. For this, we consider a conservative system in
‘which the Lagrangian is not an explicit function of time
ie, L = L(4j4;) (2)
We construct an n-dimensional configuration space for the system
and discuss the motion of the system from 1 = ¢, to *=4, along the
actual and a varied path as shown in the following figure. Note that,
the end points of all varied paths are coterminous in coordinates and 4
26 CLASSICAL AND QUANTUM MECHAy,
ime. The d-variation is illustrated in the figure drawn, ;
'
)
‘The S-variation in L can be writen as :
de
aL at
8L = Sy 8aj 425-54; (63)
F245 7 74;
Integrating both sides of the above equation from #= ¢, to f=
we get
in hie
a 45 ab
bid = fe sade + fy Baar
fou = Testa te Btona |
64) ag
The L.H.S. ofthis equation vanishes on account of the Hamilton's
principle (equation 60)) so that we get yee
But
ig gg ea ge pet
Jagp dears Seonao 065) 8
EE et Lew fore
Integrating the second term by pats, we get ihe
5 the
Ah aap Hae
2 (a dt of
[ ae a4, JO (66) FB ‘RgRANGIAN AND HAMATONIAN FORMULATIONS
But at the end points of the path, the S-variation of the co-
dinates vanishes, ie. 6ig(th) = (ta) = 0. This means that, the
‘st term in the RHLS. of the above equation vanishes so that equation
6) becomes
aL,
“pd ( aL
= shee 8g jdt (67)
a a3; j' (67)
‘Substituting this result in equation (65), we get
4 OL 4 a( ab
sae Hala a
Pe
Since: q's are all independent, the variations 6's are also
independent and the coefficients of 6) vanish separately so that we
43) 2an write the integrand to be equal to zero
‘iz:
4 ji
x
Th
“ af aL) ae
2 se. 24 aq) =0 69)
which are the Lagrange’s equations for a conservative system.
HAMILTONIAN FUNCTION
4), the formulation of the tas of mechanics with the help of
Lagrangian and the equations of motion following from it it has been
1's persumed thatthe mechanical state ofa system is completely defined
nce its generalized coordinates and generalized velocities are specified.
But this is only one of the possible ways in which the motion of @
mechanical system can be specified. There is another mode of deseripton,
which is much more powerful and advantageous than the Lagrangian
Formalism, known as the Hamiltonian formalism, where the state of
‘the system is described in terms of the generalized coordinates and
ihe generalized momenta, Here, we define another function called the
Hlemiltonian H, having the functional form H (q» pj) ie. @ function
6) of generalized coordinates, generalized momenta and time, The transition
5) fiom the Lagrangian tothe formulation con
shang of variables fom (8) 10 (2 ~
HAMILTONIAN CONSERVATION THEOREM,
We ave sey sen that th translation nd ttn
of mechanical systems lead to the conservation of the lng
angular moment, respectively. Theres anther conservation
‘which seated to energy oresponding the symmeuy’ wy
tothe translation ofa system along tne called the conserng
amittonian, This sys that, if he Lagrangian of system dy
depend on tine excl, but only ipl trough he varie
and he traslation of the ter long tine leaves the Lap
‘eta then, te Laggan il bf he fm =
a
aod 3-0 |
Hera derativ ofthe Lagangian wt ini ven by
Hyd yy emy |
a Taya aay
too 3 ing te Lge’ etn ve
epi tinte 2 w oe
+ nnd) i
This means that the quantity in the bracket is constant ie?
ey Eq,
ee ence .
Poté ye quantity is known as the Hamiltonian of the system H, given by
eAT2)
“nq, Equation (71) can therefore be wri
Mey aH
the, rie 473)
a4 which states that, if the Lagrangian is not an explicit function of time,
Aly ut depends only implicitly through the time variation of q and 4,, the
log jlamiltonian is @ constant quantity for the system. The statement i
tion, known as the conservation theorem for the Hamiltonian of a system.
‘gray Equation (72) also shows that Hamiltonian is a function of generalized
Ug, coordinates, generalized momenta and time.
We will now show that the Hamiltonian represents the total energy
of the system when potential energy is a function of coordinates only
y (WV (qs Gees a conservative system) and the kinetic energy is a
quadratie function of velocities (T= 7 (4j)). The statement that the
Kinetic energy is a quadratic function of velocities is equivalent to
saying that the transformation equations connecting the dependent
and the generalized coordinates do not depend on time explicitly (equation
nO
Since the potential energy does not depend on velocity or time,
wwe can write the generalized momentum as
(74)
Substituting this result inthe expression for the Hamiltonian (equation
(72), we get
or
ne Lagenk 05
Las 35, rc)
In terms of the Cartesian components, the kinetic energy is given
1
= pmi (76) 20 CLASSICAL AND OU)
When the transformation equations connecting the
the generalized coordinates do not depend on time ex
viene
equation (16), the expression for the velocity is
an. L
ay,
= 4,
mo Fay +
Substituting this result in the expression for the kinetic xg
(equation (76)), we get ,
1 on. or r
= XimrsiG. rig |
To Ey kg UF
an, Br,
UL aie og
a2 Bean, any
Now, 4
Multiplying the above equation by 4, and summing over all,
340 3m(220)]
7 My
Now, comparing equations (78) and (80), we get
ar
Lag, =r
24, 4
Substituting this result in equation (75), we get
Hn
=T+V¥ A
which is the total energy £ of the system, which is a const
The result shows that the Hamiltonian represents the total energy
system when the potential energy is independent of velocities ®
time and the kinetic energy is a quadratic function of only veloc
This is the law of conservation of energy for a conservative syst®
having timegndependent constraints a
rach ome i VAUMTONAN FORMATIONS 31
ly, HAMILTON'S ‘CANONICAL EQUATIONS OF MOTION
In Lagrangian formulation, we have used generalized coordinates
and generalized velocities and time in formulating dynamical problems,
s.gwhich result in a set of n = 3N-k linear second order differential
equations. In a new formulation called the Hamiltonian formulation,
© eng We use the function Hamiltonian, which is a function of the generalized
coordinates, generalized momenta and possibly time. No new physical
concept is introduced in this formulation but is only a different tool to
work with the mechanical problems. The resulting equations are called
Hamilton’s canonical equations of motion, which are 2m linear first
order differential equations. The Hamiltonian and and the Hamilton’s
«ug eauations of motion are applicable only for holonomic conservative
systems.
‘To arrive at the Hamilton’s canonical equations of motion, we use
(9 equation (72)
i
7 H(ajsPyjot) = La;Pp- May. 9j)
i
‘Taking differential of the above equation, we get
a dH a(zan }-a
oH oH oH
ie, Ls—dgj+L5—dp; +—=—dt
a Jj gutsy f Pla ay 1.2
aL al aL
= Lay byt Ly 4) 5 aj - 50) Ft (83)
ere 4; 04, ar
Now, from the definition of the generalized momentum, we have
a.
G a,
onsta# and using the Lagrange’s equations for a conservative system, we
hs eB ee
“ey of) have
o
ies
ies a
syst! 24, CLASSICAL AND QUANTUM
2
Subrng zea inegton we at
BH gy ap ol t
SE gj + ey GO
Fag ay a ‘
thy
Bg Eyl) 04 Pap \
tact i
a ‘
= Lipa) 4-7 t
‘Now, equating coeffcins of day dy and dl 09 both sgn
uation 4), we get,
Ht. -»,
my " f
an
Bea 1
an 2
“ ae 4
ustios (85) and (86) ae the Hamilton canonical equi
motion, which oosise a set of 2 fst order teat aie
‘unin. Ne tha, Hamitons equations ae sumerc ing
‘xcept fora mins sign in the st et. This means tht te
Sindy cin bo trv interchanged by making change ofS
enctlied coordinates and. generalized moment are