Questions
1-2 questions in major PSU papers
Difficulty
Medium
Importance
Essential for scoring in the General Engineering/Math section
Overview
Numerical Methods provide algorithmic techniques to find approximate solutions to mathematical problems that are difficult to solve analytically. In PSU exams, this topic is vital for solving non-linear, algebraic, and differential equations efficiently using iteration-based logic. Aspirants must master the convergence properties and standard iterative formulas to solve numerical problems rapidly.
Root Finding Methods
These methods locate the roots of non-linear equations f(x)=0. Understanding the rate of convergence and the specific iterative step for each method is essential for solving objective-type questions quickly.
- Bisection Method: Linear convergence, guaranteed to converge, slow.
- Newton-Raphson Method: Quadratic convergence, formula x(n+1) = x(n) - f(x(n))/f'(x(n)).
- Regula-Falsi Method: Linear convergence, based on the principle of false position.
- Secant Method: Faster than Bisection, superlinear convergence rate approx 1.618.
- Newton-Raphson fails if f'(x) = 0.
Numerical Integration
Numerical integration estimates the definite integral of a function based on discrete data points. These formulas are frequently tested in engineering aptitude sections to evaluate areas under a curve or work-done problems.
- Trapezoidal Rule: Uses linear interpolation, error order O(h^2).
- Simpson's 1/3 Rule: Requires an even number of intervals, error order O(h^4).
- Simpson's 3/8 Rule: Requires number of intervals to be a multiple of 3.
- Trapezoidal Rule is exact for first-degree polynomials.
- Simpson's rules are exact for polynomials up to degree 3.
Linear and Differential Equations
These methods solve systems of linear equations and initial value problems for differential equations. Focus on Gaussian elimination and Runge-Kutta methods as they are standard curriculum for competitive exams.
- Gauss-Seidel Method: Iterative approach for linear systems, converges if the matrix is diagonally dominant.
- Euler's Method: Simplest approach for first-order ODEs, y(n+1) = y(n) + h*f(x(n), y(n)).
- Runge-Kutta 4th Order: Highly accurate standard method for solving ODEs.
- Jacobi iteration: Stationary iterative method for linear systems.
Formula Sheet
Newton-Raphson: x(n+1) = x(n) - [f(x(n))/f'(x(n))]
Trapezoidal Rule: Integral = (h/2) * [y(0) + 2(y(1)+...+y(n-1)) + y(n)]
Simpson's 1/3 Rule: Integral = (h/3) * [y(0) + 4(y(odd)) + 2(y(even)) + y(n)]
Simpson's 3/8 Rule: Integral = (3h/8) * [y(0) + 3(y(1)+y(2)) + 2y(3) + ... + y(n)]
Euler's Formula: y(n+1) = y(n) + h*f(x(n), y(n))
Regula-Falsi: x = [a*f(b) - b*f(a)] / [f(b) - f(a)]
Exam Tip
Memorize the order of convergence for each method, as PSU exams frequently feature theoretical MCQs comparing the efficiency of Newton-Raphson versus Bisection.
Common Mistakes
- Confusing the interval requirements for Simpson's 1/3 rule (even) versus 3/8 rule (multiple of 3).
- Applying Newton-Raphson method without checking if the initial guess is close to the root, leading to divergence.
- Forgetting to switch the calculator to 'Radian' mode when evaluating trigonometric functions in numerical methods.
More Revision Notes
Ready to test yourself?
Play topic-wise Numerical Methods questions in Aspirant Arcade — gamified MCQ practice.
Download Free