BISECTION METHOD
GROUP-01
2024-CIV-51
2024-CIV-52
2024-CIV-53
2024-CIV-54
2024-CIV-55
2024-CIV-56
2024-CIV-57
2024-CIV-58
2024-CIV-59
2024-CIV-60
2024R/2023-CIV-124
2024-CIV-240
2024-B
CIVIL ENGINEERING DEPARTMENT
UNIVERSITY OF ENGINEERING AND TECHNOLOGY,
LAHORE
Numerical Analysis
Numerical analysis is the branch of mathematics that focuses on
developing and using methods to find approximate solutions to
mathematical problems that are difficult or impossible to solve exactly.
It deals with:
•Finding numerical (approximate) answers instead of exact formulas.
•Designing algorithms to solve equations, integrals, differential
equations, optimization problems, etc.
•Controlling and minimizing error in the approximations.
•Using computers to perform step-by-step calculations.
Use of numerical methods in engineering:
Numerical analysis is used in engineering to
solve real-world problems that cannot be handled using exact
mathematical methods. Engineers apply numerical techniques to find
approximate solutions for complex equations in areas like fluid flow, heat
transfer, structural analysis, and vibration systems. It is essential in
simulations, optimization, and computer-based tools such as MATLAB,
ANSYS, and CAD software. By using algorithms like the bisection
method, finite element analysis, and numerical integration, engineers can
design safer structures, predict system behavior, and make efficient
decisions when analytical solutions are difficult or impossible to obtain.
Bisection Method
Bisection method is a numerical technique used to find the root
(solution) of an equation of the form f(x)=0, when the exact solution is
difficult to obtain.
Definition
The bisection method is a bracketing method that repeatedly divides an
interval into two halves and selects the sub-interval in which the function
changes sign. It works only if and have opposite signs (one positive, one
negative), indicating that a root lies between them.
Explanation:
1. Choose two values a and b such that f(a).f(b)<0
2. Find the midpoint m=(a+b)/2
3. Check the sign of f(m):
If f(a)*f(m)<0, then the root is [a,m]—>b=m
If f(m)*f(b)<0, then the root is in [m,b]—>a=m
4. Repeat the process with the new interval.
5. Stop when the difference between two midpoints (or a and b) becomes
very close to zero.
Conditions to Use the Bisection Method:
1. The function must be continuous on the chosen interval.
If the function has a break, jump, or is undefined in the interval, the
method won't work.
2. The function values at the endpoints must have opposite signs:
f(a)*f(b)<0
3. Only one root should ideally lie in the interval for correct convergence.
Limitations of the Bisection Method
• Slow convergence:
• It takes many iterations to reach high accuracy compared to faster
methods like Newton-Raphson.
• Requires a sign change:
• If both f(a) and f(b) are positive or both negative, the method cannot
start.
• Cannot detect multiple roots in the same interval.
• Not efficient for complex or large-scale problems.
• Doesn't work well if the root lies exactly at the endpoint unless
checked separately. small, or when f(m) is close to zero.
Manning’s Equation (for open channel flow):
Q = 1/n*(A)*(R⅔)*(S½)
Meaning of Each Term:
Q= Discharge or flow rate (Units: m³/s).
n= Manning’s roughness coefficient (Depends on channel material (e.g.,
concrete, earth, gravel)).
A= Cross-sectional area of flow (Units: m²).
R = Hydraulic radius = A/P (Units: m).
S= Channel bed slope.
P = Wetted perimeter (Length of the channel boundary in contact with the
water) (Units: m).
Problem statement:
Design a rectangular open channel for specific discharge and
following characteristics:
Channel width=b=4.0m
Bed slope=S=0.001
Manning’s roughness coefficient=n=0.015
Discharge=Q=20m³/s
Solution:
We know all the characteristics of rectangular open channel other
than the depth(y) so we formulate the equation as:
Q=1/n*(A)*(R⅔) *(S½)
We can also write it as a function of y:
f(y)=1/0.015*[(4×y) *(4y/4+2y)] ⅔*(0.001) ½]-20=0
The above equation is non-linear and can't be solved by ordinary methods
for solving equations. So, we will solve and find the root by using
bisection method.
Firstly, we will find the interval of values of y as: [a, b] such that
f(a)*f(b)<0.
f(1) = –13.5646
f(2) = –3.1345
f(3) = 8.5678
If we take a=2 and b=3 then the condition satisfies:
f(a)*f(b)<0 , Now we start bisecting the interval and check the new value
of our function f(y) at midpoint(m) of the interval. Then we decide
whether to take interval [a,m] or interval [m,b], depends which satisfies
the condition, i,e, f(p1)*f(p2) <0
Iterations
1.
a=2.0000, b=3.0000, m = (2.0000+3.0000)/2 =2.5000
f(m)= f(2.5000) =2.6159 ,so we take interval [2.0000,2.6159]
2.
a=2.0000, b=2.6159, m = (2.0000+2.6159)/2 =2.3080
f(m)= f(2.3080) =0.3797 ,so we take interval [2.0000,2.3080]
3.
a=2.0000, b=2.3080, m= (2.0000+2.3080)/2 =2.1540
f(m) =f(2.1540) =-1.3897 ,so we take interval [2.1540,2.3080]
4.
a=2.1540, b=2.3080, m = (2.1540+2.3080)/2 =2.2310
f(m) =f(2.2310) =-0.5079 ,so we take interval [2.2310,2.3080]
5.
a=2.2310, b=2.3080, m = (2.2310+2.3080)/2 =2.2695
f(m) =f(2.2695) =-0.0648 ,so we take interval [2.2695,2.3080]
6.
a=2.2695, b=2.3080, m = (2.2695+2.3080)/2 =2.2888
f(m) =f(2.2888) =0.1579 ,so we take interval [2.2695,2.2888]
7.
a=2.2695, b=2.2888, m = (2.2695+2.2888)/2=2.2792
f(m) =f(2.2792) =0.0470 ,so we take interval [2.2695,2.2792]
8.
a=2.2695, b=2.2792, m = (2.2695+2.2792)/2 =2.2744
f(m) =f(2.2744) =-0.0083 ,so we take interval [2.2744,2.2792]
9.
a=2.2744, b=2.2792, m = (2.2744+2.2792)/2 =2.2768
f(m) =f(2.2768) =0.0192 ,so we take interval [2.2744,2.2768]
10.
a=2.2744, b=2.2768, m = (2.2744+2.2768)/2 =2.2756
f(m) =f(2.2756) =0.0056 ,so we take interval [2.2744,2.2756]
11.
a=2.2744, b=2.2756, m = (2.2744+2.2756)/2 =2.2750
f(m) =f(2.2750) =-0.0012 ,so we take interval [2.2750,2.2756]
12.
a=2.2750, b=2.2756, m = (2.2750+2.2756)/2 =2.2753
f(m) =f(2.2753) =0.0021 ,so we take interval [2.2750,2.2753]
13.
a=2.2750, b=2.2753, m = (2.2750+2.2753)/2=2.2752
f(m)=f(2.2752) =0.0009 ,so we take interval [2.2750,2.2752]
14.
a=2.2750, b=2.2752, m = (2.2750+2.2752)/2 =2.2751
f(m) =f(2.2751) =-0.0002 ,so we take interval [2.2751,2.2752]
15.
a=2.2751, b=2.2752, m = (2.2751+2.2752)/2=2.2752
f(m) =f(2.2752) =0.0009 ,so we take interval [2.2751,2.2752]
Conclusion
As two Iterations repeated, we can say y, depth of the rectangular water
channel, should be approximately 2.2752 m to satisfy the equation.
f(y)= (1/0.015) *(4×y) *(4y/4+2y) ⅔*(0.001) ½–20=0
Means if we want to construct a rectangular water channel having a
width of 4m, slope of 0.001, roughness coefficient of 0.015 and a 20m³/s
discharge should be through it then it's depth must be approximately
2.2752m.
CONCLUDED
