Bottom Science- Bisection Method
- Secant Method
- Regular Falsi (False Position) Method
- Newton Raphson Method
- Gauss Elimination Method
- Gauss Jordan Method
- Gauss-Seidel Method
- Lagrange Interpolation Method
- Newton Divided Difference Interpolation
- Newton Forward Difference Interpolation
- Newton Backward Difference Interpolation
- Trapezoidal Rule
- Simpson 1/3rd Rule
- Simpson 3/8 Rule
- Euler’s Method
- Euler’s Modified Method
- Runge-Kutta 2nd Order Method
- Runge-Kutta 4th Order Method
- Cubic Spline Method
- Bilinear Interpolation Method
- Milne’s Method
- More topics coming soon…
Milne’s method is a numerical method used to solve ordinary differential equations (ODEs) of the form y'(x) = f(x, y(x)), where y(x) is the unknown function and f(x, y(x)) is a given function.
It is a predictor-corrector method, which means that it uses an initial approximation (the “predictor”) to estimate the value of the unknown function at a new point, and then uses this estimate to obtain a more accurate value (the “corrector”).
The method is based on Taylor series expansions of the unknown function, and consists of the following steps:
- Given an initial value y(x_0) = y_0 and a step size h, use a one-step method (such as Euler’s method) to estimate y(x_0 + h) = y_1.
- Use the estimate y_1 to compute a Taylor series expansion of the unknown function around x_0, up to and including the second derivative: y(x) = y_0 + y'(x_0)(x – x_0) + (1/2)y”(x_0)(x – x_0)^2 + O((x – x_0)^3)
- Use the estimate y_1 to compute an improved estimate of the value of the unknown function at x_0 + h, denoted y_2.
- Repeat the process to estimate y(x_0 + 2h), y(x_0 + 3h), and so on.
The method is a 4th order method, which means that the global truncation error of this method is of the order O(h^5) which means it is more accurate than Euler’s method which is of order O(h^2)
It is worth mentioning that Milne’s method is less common to use in practice than other methods like Runge-Kutta method and Adams-Bashforth method.
Here is a basic example of how to use Milne’s method to solve the simple ODE y'(x) = -y(x) with the initial value y(0) = 1.
Milne’s Method in Python
import numpy as np
def milnes_method(f, y0, x0, x_end, h):
# Initialize the solution array
x = np.arange(x0, x_end+h, h)
y = np.zeros(len(x))
y[0] = y0
#Milne's Method [By Bottom Science]
# Iterate over the steps
for i in range(1, len(x)):
y_pred = y[i-1] + h*f(x[i-1], y[i-1])
y_corrected = y[i-1] + (h/4)*(3*f(x[i], y_pred) - f(x[i-1], y[i-1]))
y[i] = y_corrected
return x, y
# Define the ODE function
def f(x, y):
return -y
# Set the initial condition and the step size
y0 = 1
x0 = 0
x_end = 10
h = 0.1
# Solve the ODE
x, y = milnes_method(f, y0, x0, x_end, h)
for i in range(len(x)):
print("x = ",x[i]," y = ",y[i])
