5 views

Uploaded by Laxman Naidu N

save

- Numerical Linear Algebra Biswa Nath Datta PART I _Ch1-6
- CFD - Assignment
- 613_1393682715_cs_mathematics
- LECT06 - MDOF Part 2 [Compatibility Mode]
- NC Final
- A Graphical Introduction to Survey Adjustment777
- Population Models Using Leslie Matrices
- Linear Algebra Review
- Complex Modes
- Final
- Markov Sample2 (1)
- Jonathan Tennyson and Steven Miller- A Program Suite for the Calculation of Ro-Vibrational Spectra of Triatomic Molecules
- 2006_IMPS_Non Graphical Solutions for the Cattell’s Scree Test PRESENTATION
- ComputerScienceandInformationTechnology_62
- Further Math Syllabus
- Power Sysetm Simulation Lab Manual_opt
- bookstat171_20170119
- Jordan
- Kekre’s Wavelet Transform for Image Fusion and Comparison with Other Pixel Based Image Fusion Techniques
- Dynamic Analysis of Base-Isolation Highway Bridge
- Syllabus f.maths
- 15 Positioning
- water-06-03864.pdf
- Von Newman theory
- Complex Eigenvalues best++++
- Rigid Body Dynamics
- Ch4QM09
- Cambridge A Level Further Mathematics Syllabus code 9231
- sepvar
- Google Eigenvector
- 168641787 Power System Analysis Notes
- 168641787 Power System Analysis Notes
- 10c
- Electrical Interview Questions Part 3
- Electrical Interview Questions Part 4
- Electrical Interview Questions Part 7
- Generator Protection Siprotec
- Electrical Interview Questions Part 10
- 45896562 Power System Protection
- 100590177 Electrical Machines Formula Sheet for Electrical Engineers
- Electrical Interview Questions Part 11
- 21657085 Jk Lecture Notes on Electric Power Systems
- Computer Networks Lecture Notes
- Burdens & Current Transformer Requirements of MiCOM Relays
- Quantz Tricks 1
- Computer Networks Lecture Notes
- Unit_2.3
- Electrical Interview Questions Part 6
- Aptitude Formula
- Unit_2.2
- Electrical Machines
- Electrical Interview Questions Part 2
- Electrical Inteview Questions Part 12
- Electrical Interview Questions Part 9
- Unit_3.1
- Unit_3.3
- Model Steam Turbine Gov System
- Unit_2.4
- Interview Questions
- Electrical Interview Questions Part 11

You are on page 1of 12

3

GAUSS – SEIDEL METHOD Once again we consider the system

Ax = y …………….. (I)

(k ) (k ) (k ) In the Jacobi scheme we used the values of x2 , x3 ,L , xn obtained in the kth iteration, in place of x2, x3, ….., xn in the first equation,

a11 x1 + a12 x2 + ..... + a1n xn = y1

**to calculate x1( k +1) from
**

(k ) (k ) (k ) a11 x1( k +1) = − a12 x2 − a13 x3 ..... − a1n xn + y1

Similarly,

(k ) 1 (k ) 2

in

(k ) i −1

the

(k ) i +1

ith

(k ) n

equation,

( i = 2,3,L , n ) ,

we

used

the

values,

x , x ,L , x , x ,L , x from

, in place of x1 , x2 ,L , xi −1 , xi +1 ,L , xn to calculate xi( k +1)

(k ) k) (k ) (k ) aii xi( k +1) = − ai1 x1( k ) − ai 2 x2 − ...... − ai ,i −1 xi(− 1 − ai , i +1 xi +1 − ..... − ain xn + yi LL (*)

**What Gauss – Seidel suggests is that having obtained x1( k +1) from the first
**

( k +1) equation use this value for x1 in the second equation to calculate x2 from

**( k +1) (k ) (k ) a22 x2 = − a21 x1( k +1) − a23 x3 − ...... − a2 n xn + y2
**

( k +1) ( k +1) and use these values of x1( k +1) , x2 , in the 3rd equation to calculate x3 , and so k +1) (k ) k) on. Thus in the equation (*) use x1( k +1) , L , xi(− in place of x1( k ) , x2 , L , xi(− 1 1 to get

**the following modification of the i -th equation to calculate xi( k +1) :
**

( k +1) k +1) k) (k ) (k ) aii xi( k +1) = − ai1 x1( k +1) − ai 2 x2 − ...... − ai ,i −1 xi(− − ai ,i +1 xi(+ 1 1 − ai ,i + 2 xi + 2 ..... − ain xn + yi

In matrix notation we can write this as,

Dx ( k +1) = − Lx ( k +1) − Ux ( k ) + y

which can be rewritten as,

**(D + L )x (k +1) = −Ux (k ) + y , and hence
**

x (k +1 ) = − (D + L ) Ux k + (D + L ) y

−1 −1

65

(see page 60).(II) is the Gauss – Seidel iteration matrix. Of course.Thus we get the Gauss – Seidel iteration scheme as. G = -(D+L)-1U ……. 1 A = 1 2 2 1 2 − 2 1 . 1 0 1 2 0 0 . 1 1 y = 0 0 1 D + L = 1 2 0 −U = 0 0 −2 0 0 2 −1 0 66 . . We have. G ≥ 1 does not mean that the scheme will diverge. Example 3: Let us consider the system x1 + 2x2 – 2x3 = 1 x1 + x2 + x3 = 0 2x1 + 2x2 + x3 = 0 sp < 1.. The acid test for convergence is G We shall now consider some examples. We shall now try to apply the Gauss – Seidel scheme for this system. and ˆ = ( D + L ) −1 y y The scheme converges if and only if G sp < 1. the scheme will converge if G < 1 in some matrix norm. considered in example 1 on page 59. But some matrix norm. and for which the Jacobi scheme gave the exact solution in the 3rd iteration. x(0) initial guess ˆ x (k +1) = Gx ( k ) + y where.

λ2 = 2. We have. −1 0 1 = −1 1 0 −2 0 0 1 G = −( D + L) −1 0 0 0 −2 2 1 U = −1 1 0 0 0 −1 0 −2 1 0 0 0 From above we get the Gauss – Seidel iteration matrix as. 67 . G sp = 2 >1 Hence the Gauss – Seidel scheme for this system will not converge. Thus λ1 = 0. Thus for this system the Jacobi scheme converges so rapidly giving the exact solution in the third iteration itself whereas the Gauss – Seidel scheme does not converge. λ3 = 2 are the three eigenvalues. Example 4: Consider the system 1 1 x2 − x3 = 1 2 2 x1 + x 2 + x 3 = 0 x1 − − 1 1 x1 − x 2 + x 3 = 0 2 2 Let us apply the Gauss – Seidel scheme to this system. 0 G = 0 0 −2 2 0 2 − 3 2 Since G is triangular we get its eigenvalues immediately. Therefore.(D + L ) Thus. as its diagonal entries.

. (D + L) −1 2 1 = −1 0 0 1 1 2 0 0 . 1 A= 1 − 1 2 −1 1 −1 2 2 0 1 −1 −1 2 1 ....(*) 1 − 2 1 is the Gauss – Seidel matrix for this sytem.... 1 0 −U = 0 0 Thus... 1 0 0 1 1 y = 0 0 1 D+L= 1 − 1 2 .. 1 2 0 0 1 2 −1 .. The Gauss – Seidel scheme is ˆ x (k +1 ) = Gx ( k ) + y x (0 ) 0 = 0 0 68 . 0 1 −1 G = − ( D + L ) U = −1 0 0 1 1 2 0 0 00 10 1 2 0 0 1 2 −1 0 0 ∴G = 0 0 2 − 12 0 1 2 − 3 2 .

where ˆ = (D + L) y −1 1 y = −1 0 0 1 1 2 01 1 0 0 = −1 . Hence in this example the Gauss – Seidel scheme 2 x (1 ) 0 1 1 ˆ = G 0 + − 1 = − 1 = Gx (0 ) + y 0 0 0 1 1 ˆ = G − 1 + − 1 +y 0 0 0 = 0 0 2 − 1 2 0 x (2 ) = Gx (1 ) 1 2 − 3 2 − 1 2 1 1 1 −1 + −1 0 0 1− 1 2 = − 1− 1 2 0 ( ) 69 . Notice that G is upper triangular and hence we readily get the eigenvalues of G as its diagonal entries. Let us now carry out a few steps of the Gauss – Seidel iteration. sp = 1 < 1 . λ1 = 0. λ3 = 1/2. since we have now been assured of convergence. (We shall first do some exact calculations). 10 0 and where G is given (*). λ2 = -1/2. Thus the eigenvalues of G are. Hence G will converge.

.. 2 2 2 2 → − 1 − 1 2 + 1 2 − 1 3 ... + k −1 2 2 k −1 ( ) − 1 1 1 − ... + = − 1 − 2 + 22 2 k −1 0 x (k ) Clearly... here ‘we’ knew ‘a priori’ that the sequence is going to sum up neatly for each component and so we did exact calculation..x (3 ) = Gx ( 2 ) 1− 1 + 1 2 2 2 1 1 ˆ + y = − 1− 2 + 22 0 ( ) If we continue this process we get k −1 − ( ) 1 1 1 1− 2 + 2 − .. Of course. 2 2 0 ( ) and by summing up the geometric series we get. x (k ) 2 3 → − 2 3 0 which is the exact solution.. x( k) 1 − 1 + 1 2 − 1 3 + 1 4 ...... If we had not noticed this we still would have carried out the computations as follows: x (1 ) = Gx (0 ) 1 ˆ = − 1 +y 0 as before 70 .

∞ = 0 .65625 ˆ = − 0 . 664062 0 0 .x (2 ) = Gx (1) 0 . x (6 ) − x (5 ) = Gx (5 ) + y 0 0 .6875 +y 0 x (3 ) = Gx (2 ) x (4 ) = Gx (3 ) x (5 ) 0 . 666016 = − 0 . 667969 = − 0 .03125 x (6 ) 0 . 667969 0 0 .001953 x (10 ) x (10 ) − x (9 ) ∞ = 0 .5 +y 0 0 . 666504 0 .671875 ˆ = − 0 . .625 +y 0 0 .000488 71 . . 666504 = − 0 . = Gx (4 ) + y 0 x (5 ) − x ( 4 ) ∞ = 0 .625 ˆ = − 0 . 666016 0 0 .003907 x (9 ) x (9 ) − x (8 ) ∞ = 0 . 664062 = − 0 . .025625 x (7 ) x ( 7 ) − x (6 ) ∞ = 0 .5 ˆ = − 0 .007813 x (8 ) x (8 ) − x (7 ) ∞ = 0 .6875 ˆ = − 0 .65625 .671875 .

λ2 = λ3 = 2 72 .Since now error is < 10-3 we may stop here and take x(10) as our solution for the system. to get. Or we may improve our accuracy by doing more iterations. 666626 (12 ) = − 0 .000031 < 10 −4 and hence we can take x(14) as our solution within error 10-4. 666687 . 666748 = − 0 . λ3 = − i 15 4 4 2 2 1 15 + = 16 = 2 4 4 4 1 ∴ λ1 = . 666687 (13 ) = − 0 . 666626 . λ2 = 1 2 1 1 + i 15 . 666656 = − 0 . 666748 0 0 . 2 0 1 2 We have the characteristic polynomial of J as λ λI − J = + 1 −1 2 −1 λ −1 2 2 −1 2 λ 1 + 1 = λ + λ 2 − + 1 2 2 λ Thus the eigenvalues of J are λ1 = − . x 0 0 . 1 1 2 −1 0 and therefore. x 0 x (14 ) x (14 ) − x (13 ) ∞ = 0 . We have 1 A= 1 − 1 2 0 J = −1 1 2 −1 1 −1 1 2 2 −1 2 1 . x (11 ) 0 . Let us now try to apply the Jacobi scheme for this system. 666656 0 0 .

Thus. in example 3 we had a system for which the Jacobi scheme converged but Gauss – Seidel scheme did not converge. but the Gauss – Seidel scheme converges. 16 λ 4 . it is not ‘correct’ to say that one scheme is better than the other. in general. Example 5: 2x1 – x2 =y1 -x1 + 2x2 – x3 = y2 -x2 + 2x3 –x4 =y3 -x3 + 2x4 = y4 Here 0 2 −1 0 −1 2 −1 0 A= . Let us now consider another example. whereas in example 4 above we have a system for which the Jacobi scheme does not converge. 0 − 1 2 − 1 0 − 0 1 2 is a symmetric tridiagonal matrix. Thus.∴ J sp = 2 which is >1. these two examples demonstrate that. The Jacobi matrix for this scheme is J = 0 1 0 0 2 1 0 1 0 2 0 1 2 2 0 1 2 0 0 1 2 0 The characteristic equation is.12 λ 2 + 1 = 0 Set λ 2 = α ………………(CJ) 73 . Thus the Jacobi scheme for this system will not converge.

8090 .12α + 1 = 0 ………………(CJ1) ∴ λ is the square root of the roots of (CJ1). ± 0. Hence J sp = 0. Thus the eigenvalues of J are ± 0. The Gauss – Seidel matrix for the system is found as follows: 0 0 2 −1 2 0 (D + L ) = 0 −1 2 0 0 −1 0 0 0 2 0 0 −U = 0 0 1 0 0 0 = 0 1 0 0 1 1 1 1 2 4 8 0 0 1 0 0 1 1 1 2 4 8 0 0 1 1 2 4 0 0 0 1 2 (D + L ) −1 16 74 .8090. and the Jacobi scheme will converge.Therefore 16α2 .3090.

.12λ + 1 = 0 …………. which becomes in this case 16λ4 − 12λ3 + λ 2 = 0. Notice that roots of (CG1) are same as those of (CJ1).6545 < 1 Thus the Gauss – Seidel scheme also converges.6545.....(CG1) Thus one of the eigenvalues of G is 0 (repeated twice). 0.. (C G ) This can be factored as λ 2 (16λ2 − 12λ + 1) = 0 Thus the eigenvalues of G are roots of λ2 = 0 ... Thus....0955... 0. Observe that 75 . ... Thus nonzero eigenvalues of G are squares of eigenvalues of J. −1 G = − (D + L ) U = 1 0 0 0 2 0 1 1 0 4 2 = 0 1 1 1 8 4 2 1 1 1 0 16 8 4 1 1 1 1 2 4 8 0 1 1 1 2 4 8 0 0 1 1 2 4 16 0 0 0 1 2 0 0 0 0 1 0 0 0 0 1 0 0 0 0 1 0 The characteristic equation of G is λI − G = 0 . G sp = 0.. and 16λ2 . ∴ the nonzero eigenvalues of G are. and two eigenvalues of G are roots of (CG1)...

We shall not go into any further details of this aspect. 76 .G sp = J 2 sp . G sp < J sp Thus the Gauss – Seidel scheme converges faster than the Jacobi scheme. In many class of problems where both schemes converge it is the Gauss – Seidel scheme that converges faster.

- Numerical Linear Algebra Biswa Nath Datta PART I _Ch1-6Uploaded bylpvasam
- CFD - AssignmentUploaded bypadmanathan
- 613_1393682715_cs_mathematicsUploaded byAnish Bhanushali
- LECT06 - MDOF Part 2 [Compatibility Mode]Uploaded byzinil
- NC FinalUploaded byAlee Sajjad
- A Graphical Introduction to Survey Adjustment777Uploaded byAmry Ali
- Population Models Using Leslie MatricesUploaded byAlbertoAlcalá
- Linear Algebra ReviewUploaded byrd8
- Complex ModesUploaded bybsitler
- FinalUploaded bySiva Shankar
- Markov Sample2 (1)Uploaded bySeong Yoon Kim
- Jonathan Tennyson and Steven Miller- A Program Suite for the Calculation of Ro-Vibrational Spectra of Triatomic MoleculesUploaded byPassamm
- 2006_IMPS_Non Graphical Solutions for the Cattell’s Scree Test PRESENTATIONUploaded bymbfreesdb
- ComputerScienceandInformationTechnology_62Uploaded byLokesh Sai Kumar Dasari
- Further Math SyllabusUploaded byJasper Lee
- Power Sysetm Simulation Lab Manual_optUploaded byvinodlife
- bookstat171_20170119Uploaded byssullivan611
- JordanUploaded bymarivel77
- Kekre’s Wavelet Transform for Image Fusion and Comparison with Other Pixel Based Image Fusion TechniquesUploaded byijcsis
- Dynamic Analysis of Base-Isolation Highway BridgeUploaded byMani Raj N
- Syllabus f.mathsUploaded byMuneeb Zafar
- 15 PositioningUploaded bysumukha1999
- water-06-03864.pdfUploaded byNacer Mera
- Von Newman theoryUploaded byGheorgheTaran
- Complex Eigenvalues best++++Uploaded bylostkin
- Rigid Body DynamicsUploaded by29011988
- Ch4QM09Uploaded byATP_101
- Cambridge A Level Further Mathematics Syllabus code 9231Uploaded byTasnia Tushi
- sepvarUploaded byKanthavel Thillai
- Google EigenvectorUploaded byPaul Sawaya

- 168641787 Power System Analysis NotesUploaded byLaxman Naidu N
- 168641787 Power System Analysis NotesUploaded byLaxman Naidu N
- 10cUploaded byLaxman Naidu N
- Electrical Interview Questions Part 3Uploaded byLaxman Naidu N
- Electrical Interview Questions Part 4Uploaded byLaxman Naidu N
- Electrical Interview Questions Part 7Uploaded byLaxman Naidu N
- Generator Protection SiprotecUploaded byLaxman Naidu N
- Electrical Interview Questions Part 10Uploaded byLaxman Naidu N
- 45896562 Power System ProtectionUploaded byLaxman Naidu N
- 100590177 Electrical Machines Formula Sheet for Electrical EngineersUploaded byLaxman Naidu N
- Electrical Interview Questions Part 11Uploaded byLaxman Naidu N
- 21657085 Jk Lecture Notes on Electric Power SystemsUploaded byLaxman Naidu N
- Computer Networks Lecture NotesUploaded byLaxman Naidu N
- Burdens & Current Transformer Requirements of MiCOM RelaysUploaded byLaxman Naidu N
- Quantz Tricks 1Uploaded byLaxman Naidu N
- Computer Networks Lecture NotesUploaded byLaxman Naidu N
- Unit_2.3Uploaded byLaxman Naidu N
- Electrical Interview Questions Part 6Uploaded byLaxman Naidu N
- Aptitude FormulaUploaded byLaxman Naidu N
- Unit_2.2Uploaded byLaxman Naidu N
- Electrical MachinesUploaded byLaxman Naidu N
- Electrical Interview Questions Part 2Uploaded byLaxman Naidu N
- Electrical Inteview Questions Part 12Uploaded byLaxman Naidu N
- Electrical Interview Questions Part 9Uploaded byLaxman Naidu N
- Unit_3.1Uploaded byLaxman Naidu N
- Unit_3.3Uploaded byLaxman Naidu N
- Model Steam Turbine Gov SystemUploaded byLaxman Naidu N
- Unit_2.4Uploaded byLaxman Naidu N
- Interview QuestionsUploaded byLaxman Naidu N
- Electrical Interview Questions Part 11Uploaded byLaxman Naidu N