. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
Chapter 10
Numerical Solutions of Nonlinear Systems of Equations
Hung-Yuan Fan (范洪源)
Department of Mathematics, National Taiwan Normal University, Taiwan
Spring 2016
Section 10.1
Fixed Points for Functions of Several
Variables
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
Objective
To solve a system of nonlinear equations of the form
f1(x1, x2,·, xn) = 0, f2(x1, x2,·, xn) = 0, ...
fn(x1, x2,·, xn) = 0,
(1)
where each fi:Rn→ R is a (nonlinear) function for i = 1, 2, . . . , n.
The unknown vector x = [x1, x2,· · · , xn]T∈ Rn is called a solution to the nonlinear system (1).
Vector-Valued Functions
Reformulation of the Nonlinear System
Consider a vector-valued function F :Rn→ Rn defined by F(x) = [f1(x), f2(x),· · · , fn(x)]T∈ Rn ∀ x ∈ Rn.
The system of nonlinear equations (1) can be represented as F(x) = 0, x = [x1, x2,·, xn]T∈ Rn. (2) The functions f1, f2,· · · , fnare called the coordinate
functions (坐標函數) of F.
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
Two Vector Norms (向量的範數)
Definition (常用的向量範數) Let v = [v1, v2,· · · , vn]T∈ Rn.
T he l2
-norm (or Euclidean norm) of v is defined by
∥v∥2=√ vTv =
vu ut∑n
i=1
v2i.
T he l∞
-norm of v is defined by
∥v∥∞= max1≤i≤n|vi|.
MATLAB Command:
norm(v, 2) or norm(v) is used for∥v∥2, and norm(v, 'inf') is used for ∥v∥∞.
Example 1, p. 631
Place the following nonlinear system
3x1− cos(x2x3)−12 = 0
x21− 81(x2+ 0.1)2+ sin x3+ 1.06 = 0 e−x1x2 + 20x3+10π3−3 = 0
(3)
in the form (2).
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
Solution
Rewrite the nonlinear system (3) as
F(x1, x2, x3)≡ [f1(x1, x2, x3), f2(x1, x2, x3), f3(x1, x2, x3)]T
= [0, 0, 0]T= 0∈ R3, where the coordinate functions are defined by
f1(x1, x2, x3) = 3x1− cos(x2x3)−1 2,
f2(x1, x2, x3) = x21− 81(x2+ 0.1)2+ sin x3+ 1.06, f3(x1, x2, x3) = e−x1x2+ 20x3+ 10π− 3
3 .
Fixed-Point Forms
As in Chap. 2, we shall transform the root-finding problem (2) into a fixed-point problem
x = G(x), x∈ D,
where G : D⊆ Rn→ Rn is some vector-valued function with domain
D ={[x1, x2,· · · , xn]T| ai≤ xi≤ bi, i = 1, 2,· · · , n} (4) for some constants a1, a2,· · · , an and b1, b2,· · · , bn.
Fixed-Point Iteration (FPI) with an initial vector
x(0)∈ D: x(k) = G(x(k−1)), k = 1, 2, . . . ,provided that x(k) ∈ D ∀ k ≥ 1.
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
Fixed Points in R
nDef 10.5
The vector-valued function G : D⊆ Rn→ Rn has a fixed point at p∈ D ifG(p) = p.
Questions
Under what conditions does the sequence of vectors {x(k)}∞k=0 generated by FPI converge to the unique fixed point p∈ D?
What is the error bound for the absolute error ∥x(k)− p∥∞? What is the rate of convergence for FPI?
We may write
G(x) = [g1(x), g2(x),· · · , gn(x)]T,
where each gi is the ith component function of G for i = 1, 2, . . . , n.
Convergence Theorem of FPI
Thm 10.6
Let G be conti. on D withG(D)⊆ D, where the domain D is defined as in (4). Then
(1) G has at least one fixed point in D.
(2) If, in addition, ∃0 < K < 1s.t. each component function gi has conti. partial derivatives with
∂gi(x)
∂xj
≤ K
n, whenever x∈ D,
for i, j = 1, 2, . . . , n, then with any x(0)∈ D conv. to the
unique fixed point p
∈ D, and∥x(k)− p∥∞≤ Kk
1− K∥x(1)− x(0)∥∞ ∀ k.
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
How to check the continuity of G?
Thm 10.4 (分量函數的連續性)
Let g : D⊆ Rn→ R be a function and x0 ∈ D. If ∃ δ > 0 and M > 0 s.t. the partial derivatives of g exist on Nδ(x0)∩ D with
∂g(x)
∂xj
≤ M ∀ x ∈ Nδ(x0)∩ D,
for j = 1, 2, . . . , n, theng is conti. at x0. Continuity of G
G is conti. at x0∈ D ⇐⇒ each component function gi is conti. at x0 for i = 1, 2, . . . , n.
G is conti. on D⇐⇒ each gi is conti. on D for i = 1, 2, . . . , n.
Example 2, p. 633
(a) Place the nonlinear system in Example 1
3x1− cos(x2x3)−12 = 0
x21− 81(x2+ 0.1)2+ sin x3+ 1.06 = 0 e−x1x2 + 20x3+10π3−3 = 0
in a fixed-point form x = G(x), x∈ D, and show that there is a unique sol. on
D ={[x1, x2, x3]T| − 1 ≤ xi≤ 1, i = 1, 2, 3}.
(b) Perform the FPI withx(0)= [0.1, 0.1,−0.1]T and the stopping criterion ∥x(k)− x(k−1)∥∞< 10−5.
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
Solution of (a)
Solving the ith eq. of (3) for xi (i = 1, 2, 3)⇒
x
1= 13cos(x2x3) + 1
6 ≡g1(x1, x2, x3)
x
2= 19
√
x21+ sin x3+ 1.06− 0.1 ≡g2(x1, x2, x3) (5)
x
3= −120e−x1x2 −10π− 3
60 ≡g3(x1, x2, x3).
So, define a vectored-valued function G : D→ R3 by
G(x1, x2, x3) = [g1(x1, x2, x3),g2(x1, x2, x3),g3(x1, x2, x3)]T∈ R3 for any x = [x1, x2, x3]T∈ D. Now, consider the fixed-point form
x = G(x), x∈ D obtained from the original nonlinear system (3).
Solution of (a)–Conti’d
Fist, we shall claim thatG(D)⊆ D. It is easily seen from (5) that for any x∈ D, we have
|g1(x)| ≤ 1
3| cos(x2x3)| +1
6 ≤ 0.50,
|g2(x)| =1 9
√
x21+ sin x3+ 1.06− 0.1
≤ 1 9
√(1)2+ sin(1) + 1.06− 0.1 < 0.09
|g3(x)| = 1
20e−x1x2 +10π− 3 60
≤ 1
20e +10π− 3
60 < 0.61.
Hence, we know thatG(D)⊆ D.
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
Solution of (a)–Conti’d
Next, simple manipulation from Calculus gives that
∂g1
∂x2
= −x3
3 sin(x2x3), ∂g1
∂x3
= −x2
3 sin(x2x3), (6)
∂g2
∂x1 = x1
9√
x21+ sin x3+ 1.06, ∂g2
∂x3 = cos x3 18√
x21+ sin x3+ 1.06, (7)
∂g3
∂x1
= −x2
20 e−x1x2, ∂g3
∂x2
= −x1
20 e−x1x2, ∂g1
∂x1
= ∂g2
∂x2
= ∂g3
∂x3
= 0.
(8)
=⇒All first partial derivatives of g1, g2, g3 are conti. on D!
Solution of (a)–Conti’d
Now, from (6) =⇒ ∂g1
∂x2
≤ |x3|
3 · | sin(x2x3)| ≤ sin 1
3 < 0.281, ∂g1
∂x3
< 0.281.
From (7), we see that ∂g2
∂x1
≤ 1
9√
sin(−1) + 1.06 = 1 9√
0.218 < 0.238, ∂g2
∂x3
≤ 1
18√
sin(−1) + 1.06 = 1 18√
0.218 < 0.119, and furthermore, from (8), we also have
∂g3
∂x1
≤ e
20 < 0.14, ∂g3
∂x2
≤ e
20 < 0.14.
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
Solution of (a)–Conti’d
Thus, the partial derivatives of g1, g2, g3 are bounded on D. It follows from Thm 10.4 thatG must be conti. on Dand
∂gi
∂xj
≤ 0.281 = K n = K
3 ∀ x ∈ D
for i, j = 1.2.3. So, the sufficient conditions of Thm 10.6 are satisfied with the constantK = (0.281)(3) = 0.843 < 1.
Conclusions
G has a unique fixed point p∈ D by Thm 10.6.
This fixed point p is one of the solutions to the original nonlinear system (3).
Solution of (b)–Numerical Results
Finally, perform the FPI
x(k)= G(x(k−1)), k = 1, 2, . . .
with x(0) = [0.1, 0.1,−0.1]T∈ D and∥x(k)− x(k−1)∥∞< 10−5. Actual sol. p = [0.5, 0,−π6 ]T ≈ [0.5, 0, −0.5235987757]T.
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
A Test for the Error Bound
With the computed sol. x(5) and the actual fixed point p∈ D,
∥x(5)− p∥∞≤ 2 × 10−8.
WithK = 0.843, the theoretical error bound would become
∥x(5)− p∥∞≤ (0.843)5
1−0.843(0.423) < 1.15.
The error bound in Thm 10.6 might be much larger than
the actual absolute error!
Accelerating Convergence (加速收斂性)
Basic Ideas
Use the latest estimates generated by the FPI x(k)1 , x(k)2 ,· · · , x(k)i−1
instead of x(k−1)1 , x(k−1)2 ,· · · , x(k−1)i−1 to compute the ith componentx(k)i .
This is the same as the idea of Gauss-Seidel method for solving linear systems. (See Chapter 7)
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
Revisit Example 2
Reformulation as Gauss-Seidel Method
Consider the following Gauss-Seidel form for Example 2 x(k)1 = 1
3cos(x(k2 −1)x(k3−1)) +1 6, x(k)2 = 1
9
√
(x(k)1 )2+ sin x(k3−1)+ 1.06− 0.1, (9) x(k)3 = −1
20e−x(k)1 x(k)2 −10π− 3
60 , k = 1, 2, . . .
withx(0)= [0.1, 0.1,−0.1]T∈ R3 and the same stopping criterion
∥x(k)− x(k−1)∥∞< 10−5.
Applying the iteration (9) with given initial vectorx(0), the numerical results are shown in the following table. Note: In
general, this method does not always accelerate the
convergence! (不一定每次都能加速固定點迭代法!)
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
Section 10.2
Newton’s Method
In One-Dimensional Case
Review of Newton’s Method
Newton’s method for solving a nonlinear equation of one variable
f(x) = 0, x∈ R
can be regarded as a fixed-point iteration with g(x) = x− 1
f′(x) · f(x) ≡ x −ϕ(x)· f(x).
The quadratic convergence of Newton’s method is always expected if the initial guess is sufficiently close to a zero of f.
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
In Multidimensional Case
Objectives
For solving a nonlinear system
F(x) = [f1(x), f2(x),· · · , fn(x)]T= 0∈ Rn, x∈ Rn, try to develop a FPI with the vector-valued function
G(x) = x−A(x)−1F(x)
≡ [g1(x), g2(x),· · · , gn(x)]T∈ Rn, x∈ Rn, (10) assuming that A(x) = [aij(x)]∈ Rn×n is nonsingular at the
fixed point p of G.
Hopefully, the quadratic convergence can be achieved under reasonable conditions.
Thm 10.7 (FPI 二次收斂的充分條件) Let G(p) = p. Suppose that∃ δ > 0 with
(i) ∂g∂xi
j is conti. on Nδ(p) for i, j = 1, 2, . . . , n;
(ii) ∂x∂2gi
j∂xk is conti. on Nδ(p), and∃ M > 0 s.t.
∂2gi(x)
∂xj∂xk
≤ M ∀ x ∈ Nδ(p),
for i, j, k = 1, 2, . . . , n;
(iii) ∂g∂xi(p)
j = 0 for i, j = 1, 2, . . . , n.
Then∃δˆ≤ δ s.t. the seq. {x(k)}∞k=0 generated by FPI converges
quadratically to p for any x
(0)∈ Nδˆ(p). Moreover,∥x(k)− p∥∞≤ n2M
2 ∥x(k−1)− p∥2∞ ∀ k ≥ 1.
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
Derivation of the Matrix A(x)
WriteA(x)−1= [bij(x)]∈ Rn×n. From (10)⇒
gi(x) = xi−
∑n k=1
bik(x)fk(x), i = 1, 2, . . . , n.
For each i, j = 1, 2, . . . , n, the first partial derivatives of gi are
∂gi(x)
∂xj
=
1− ∑n
k=1
(∂bik(x)
∂xj fk(x) +bik(x)∂f∂xk(x)j )
, i = j,
−∑n
k=1
(∂bik(x)
∂xj fk(x) +bik(x)∂f∂xk(x)
j
)
, i̸= j.
(11)
Derivation of the Matrix A(x)–Conti’d
From condition (iii) of Thm 10.7 and (11), we immediately obtain
0 = ∂gi(p)
∂xj =
1− ∑n
k=1
bik(p)∂f∂xk(p)j , i = j,
−∑n
k=1
bik(p)∂f∂xk(p)
j , i̸= j. (12) Define the Jacobian matrix J(x) = [∂f∂xi(x)
j ]∈ Rn×n by
J(x) =
∂f1
∂x1(x) ∂x∂f1
2(x) · · · ∂x∂f1n(x)
∂f2
∂x1(x) ∂x∂f2
2(x) · · · ∂x∂f2n(x)
... ... ...
∂fn
∂x1(x) ∂x∂fn
2(x) · · · ∂x∂fnn(x)
, x∈ Nδ(p).
It follows from (12) that A(p)−1J(p) = I or A(p) = J(p).
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
Newton’s Method
So, it is appropriate to choose A(x) = J(x)for x∈ Nδ(p).
Basic form of Newton’s method for nonlinear systems:
x(k) = G(x(k−1)) = x(k−1)−A(x(k−1))−1F(x(k−1))
= x(k−1)−J(x(k−1))−1F(x(k−1)), k = 1, 2, . . . , (13) where x(0) ∈ Nˆδ(p) and J(x) is nonsingular on Nˆδ(p) with 0 <δˆ≤ δ.
Quadratic convergence of Newton’s method is guaranteed
from Thm 10.7 if the initial guess is sufficiently close to p!Some Comments on Newton’s Method (13)
We DO NOT computeJ(x(k−1))−1
explicitly in practical
computation.In order to save the operation counts, we first solve the linear system
J(x(k−1))y=−F(x(k−1))
for the correction vectoryusing Gaussian Elimination
with Partial Pivoting, and then compute the next iterate via
x(k)= x(k−1)+y.
Floating-point operation counts ≈ O(23
n
3) per iteration.. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
Pseudocode of Newton’s Method
To approx. the sol. of the nonlinear system F(x) = 0, x∈ Rn. Algorithm 10.1: Newton’s Method for Systems
INPUT dim. n; initial x∈ Rn; tol. TOL; max. no. of iter. N0. OUTPUT an approx. sol. x to the nonlinear system.
Step 1 Set k = 1.
Step 2 While (k≤ N0) do Steps 3–7
Step 3 Compute F(x) and the Jacobian matrixJ(x).
Step 4 Solve the n× n linear systemJ(x)y=−F(x).
Step 5 Set x = x +y.
Step 6 If∥y∥ < TOL then OUTPUT(x); STOP.
Step 7 Set k = k + 1.
Step 8 OUTPUT(‘Maximum number of iterations exceeded’); STOP.
Example 1, p. 641 (See also Example 2 of Sec. 10.1) Apply Newton’s Method to solve the nonlinear system
3x1− cos(x2x3)−12 = 0
x21− 81(x2+ 0.1)2+ sin x3+ 1.06 = 0 e−x1x2+ 20x3+ 10π3−3 = 0
withx(0)= [0.1, 0.1,−0.1]T and∥x(k)− x(k−1)∥∞< 10−5.
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
Numerical Results of Example 1
The Jacobian matrix J(x) is easily obtain from Calculus as
J(x1, x2, x3) =
3 x3 ∼ (x2x3) x3sin(x2x3) 2x1 −162(x2+ 0.1) cos x3
−x2e−x1x2 −x1e−x1x2 20
.
Actual sol. p = [0.5, 0,−π6 ]T≈ [0.5, 0, −0.5235987757]T.
Section 10.3
Quasi-Newton Methods
(擬牛頓法)
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
Newton’s Method v.s. Broyden’s Method (1/2)
For Each Iterate of Newton’s Method
At least n2 scalar functional evaluations for the Jacobian matrix J(x(k)) andn scalar functional evaluations for F(x(k)).
Solving a linear system involving the Jacobian requires O(n3) operation counts.
Self-Correcting: it will generally correct for roundoff error
with the successive iterations.Quadratic convergenceoccurs if a good initial guess is given.
Newton’s Method v.s. Broyden’s Method (2/2)
For Each Iterate of Broyden’s Method
Only n scalar functional evaluations are required!
The amount of operation counts for solving the linear system is reduced to O(n2).
It is Not Self-Correcting with the successive iterations.
Only superlinear convergenceoccurs if a good initial guess is given, i.e., we have
klim→∞
∥x(k+1)− p∥
∥x(k)− p∥ = 0,
where p∈ Rn is a solution of the nonlinear system F(x) = 0.
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
About Broyden’s Method . . .
lt belongs to a class of least-change secant update
methods that produce algorithms called quasi-Newton.
The quasi-Newton methods replace the Jacobian matrix in Newton’s method with an approximate matrix that is
easily updated at each iteration.
References (參考文獻)
[Broy] C G. Broyden, A class of methods for solving nonlinear simultaneous equations, Math. Comp., 19(92), 577-593, 1965.
[DM] J. E. Dennis, Jr. and J. J. Moré, Quasi-Newton methods, motivation and theory, SIAM Rev., 19(1), 46–89, 1977.
Derivation of Broyden’s Method (1/2)
For an initial approx. x(0)∈ Rn, compute the Jacobian matrix A0 = J(x(0))∈ Rn×n and the first iterate
x(1)= x(0)−A0−1F(x(0)) as Newton’s method.
If we let
s1= x(1)− x(0) and y1= F(x(1))− F(x(0)), want to determine a matrixA1 ≈ J(x(1))∈ Rn×n satisfying the quasi-Newton condition or secant condition
A1(x(1)− x(0)) = F(x(1))− F(x(0)) or A1s1 = y1. (14)
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
Derivation of Broyden’s Method (2/2)
To determineA1 uniquely, Broyden[Broy] imposed
A1z = A0z ∀ z ∈ Rn with sT1z = 0 (15) on the secant codition (14).So, it follows from (14) and (15) that [DM]
A1= A0+(y1− A0s1)
∥s1∥22 · sT1
and hence x(2) = x(1)−A1−1F(x(1)).
In general, for k≥ 2, we have
Ak = Ak−1+(yk− Ak−1sk)
∥sk∥22 ·skT, (16) x(k+1) = x(k)−Ak−1F(x(k)),
wheresk= x(k)− x(k−1)=−A−1k−1F(x(k−1)) and yk= F(x(k))− F(x(k−1)).
Remarks
From (16), we see that Ak is obtained from the previous Ak−1 plus an rank-1 updating matrix.
This technique is called the least-change secant updates.
In single-variable Newton’s method, may write f′(xk)≈ f(xk)− f(xk−1)
xk− xk−1 or f′(xk)(xk−xk−1)≈ f(xk)−f(xk−1);
while we try to determine uniquelyAk ≈ J(x(k))s.t.
Ak(x(k)− x(k−1)) = F(x(k))− F(x(k−1)) in the multidimensional case.
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
A Question
With the special structure of Ak, how to reduce the number of arithmetic calculations toO(n2) for solving the n× n linear system Ak−1F(X(k))?
Thm 10.8 (Sherman-Morrison Formula)
If A∈ Rn×n is nonsingular and x, y∈ Rn are nonzero vectors with yTA−1x̸= −1, then A+xyT is nonsingular and
(A+xyT)−1= A−1−A−1xyTA−1 1 + yTA−1x.
Reformulation of A
−1kFor each k≥ 1, from (16) and Sherman-Morrison formula =⇒
Ak−1 =
(Ak−1+(yk− Ak−1sk)
∥sk∥22
· sTk
)−1
= A−1k−1−A−1k−1(y
k−Ak−1sk
∥sk∥22
)sTkA−1k−1
1 + sTkA−1k−1(
yk−Ak−1sk
∥sk∥22
)
= A−1k−1− (A−1k−1yk− sk)(sTkA−1k−1)
∥sk∥22+ sTkA−1k−1yk− ∥sk∥22
= A−1k−1+(sk− A−1k−1yk)(sTkA−1k−1) sTkA−1k−1yk
= A−1k−1+(sk−A−1k−1yk)(sTkA−1k−1)
−sTk · (−A−1k−1yk) . (17)
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
Algorithm 10.2: Broyden’s Method
INPUT dim. n; initial x∈ Rn; tol. TOL; max. no. of iter. N0. OUTPUT an approx. sol. x of nonlinear system F(x) = 0.
Step 1 Set A0 = J(x): the Jacobian matrix evaluated at x.
v = F(x). (Note: v = F(x(0)).) Step 2 Set A = A−10 . (Use Gaussian elimination.)
Step 3 Set s =−Av; x = x + s; k = 1. (Note: s = s1, x = x(1).) Step 4 While (k≤ N0) do Steps 5–11.
Step 5 Set w = v; v = F(x); y = v− w. (Note: y = yk.) Step 6 Set z =−Ay. (Note: z = −A−1k−1yk.)
Step 7 Set p =−sTz. (Note: p = sTkA−1k−1yk.)
Step 8 Set uT= sTA; A = A +1p(s + z)uT. (Note: A = A−1k .) Step 9 Set s =−Av; x = x + s. (Note: s = −A−1k F(x(k)) and
x = x(k+1).)
Step 10 If∥s∥ < TOL then OUTPUT(x); STOP.
Step 11 Set k = k + 1.
Step 12 OUTPUT(‘Maximum number of iterations exceeded’); STOP.
Example 1, p. 651 (See also Example 2 of Sec. 10.1) Use Broyden’s Method to solve the nonlinear system
3x1− cos(x2x3)−12 = 0
x21− 81(x2+ 0.1)2+ sin x3+ 1.06 = 0 e−x1x2+ 20x3+ 10π3−3 = 0
withx(0)= [0.1, 0.1,−0.1]T and∥x(k)− x(k−1)∥2 < 10−5.
. . . . . .
. . . . . . . .
. . . . . . . .
. . . . . . . .
. . .
. .
. . . . .
Numerical Results for Example 1
The superlinear convergence of Broyden’s method for Example 1 is demonstrated in the following table, and the computed solutions are less accurate than those computed by Newton’s method.
Actual sol. p = [0.5, 0,−π6 ]T ≈ [0.5, 0, −0.5235987757]T.