Questions
1–2 questions per paper
Difficulty
Medium
Importance
Moderate yield for HPCL/NTPC/BHEL
Overview
Numerical Methods provide algorithms for solving complex engineering problems where analytical solutions are difficult or impossible to obtain. These methods are essential for PSU exams as they test your ability to perform rapid, iterative calculations using standardized formulas. Mastery involves memorizing convergence rates and error formulas for iterative procedures.
Root Finding Methods
Root finding algorithms are used to solve non-linear equations f(x) = 0 through iterative approximations. Bisection is reliable but slow, while Newton-Raphson is faster but requires the derivative and a good initial guess.
- Bisection method convergence: Linear
- Newton-Raphson formula: x(n+1) = x(n) - f(x(n))/f'(x(n))
- Newton-Raphson convergence: Quadratic
- Secant method formula: x(n+1) = x(n) - f(x(n)) * [(x(n) - x(n-1)) / (f(x(n)) - f(x(n-1)))]
- Secant method convergence: Superlinear (1.618)
Numerical Integration
These methods approximate the definite integral of a function by using samples of the function at various points. They are frequently tested through rule-based identification for specific polynomial degrees.
- Trapezoidal rule error: O(h^2)
- Simpson's 1/3 rule requires an even number of intervals
- Simpson's 3/8 rule requires intervals to be a multiple of three
- Trapezoidal formula: (h/2) * [y0 + yn + 2(y1 + ... + yn-1)]
- Simpson's 1/3 rule: (h/3) * [y0 + yn + 4(sum of odd) + 2(sum of even)]
Solution of Ordinary Differential Equations (ODEs)
Numerical ODE solvers allow for the step-by-step calculation of a function given its slope and an initial condition. Fourth-order Runge-Kutta is the industry standard for precision versus effort.
- Euler's Method: y(n+1) = y(n) + h * f(x(n), y(n))
- RK2 Method (Heun's method) error: O(h^2)
- RK4 Method error: O(h^4)
- RK4 is widely preferred for its high stability and accuracy
- Predictor-Corrector methods: Milne's and Adams-Bashforth
Formula Sheet
Newton-Raphson: x(n+1) = x(n) - f(x(n))/f'(x(n))
Trapezoidal Rule: I = (h/2) * [y0 + yn + 2(y1 + y2 + ... + yn-1)]
Simpson's 1/3 Rule: I = (h/3) * [y0 + yn + 4(y1 + y3 + ...) + 2(y2 + y4 + ...)]
Simpson's 3/8 Rule: I = (3h/8) * [y0 + yn + 3(y1 + y2 + y4 + y5 + ...) + 2(y3 + y6 + ...)]
Euler's: y(n+1) = y(n) + h*f(xn, yn)
RK4: y(n+1) = y(n) + (1/6)*(k1 + 2k2 + 2k3 + k4)
Exam Tip
Always memorize the convergence rate order for each method, as PSU exams frequently ask which method converges fastest.
Common Mistakes
- Forgetting to switch the calculator to Radian mode when trigonometric functions appear in Newton-Raphson iterations.
- Applying Simpson's 1/3 rule to a set of data points that is not divisible by 2.
- Miscounting intervals versus data points (n intervals involve n+1 data points).
More Revision Notes
Ready to test yourself?
Play topic-wise Numerical Methods questions in Aspirant Arcade — gamified MCQ practice.
Download Free