# Crank-Nicolson Method

### Description

The Crank-Nicolson method for solving ordinary differential equations is a combination of the generic steps of the forward and backward Euler methods.

### Usage

 1 cranknic(f, t0, t1, y0, ..., N = 100) 

### Arguments

 f function in the differential equation y' = f(x, y); defined as a function R \times R^m \rightarrow R^m, where m is the number of equations. t0, t1 start and end points of the interval. y0 starting values as row or column vector; for m equations y0 needs to be a vector of length m. N number of steps. ... Additional parameters to be passed to the function.

### Details

Adding together forward and backword Euler method in the cranknic method is by finding the root of the function merging these two formulas.

No attempt is made to catch any errors in the root finding functions.

### Value

List with components t for grid (or ‘time’) points between t0 and t1, and y an n-by-m matrix with solution variables in columns, i.e. each row contains one time stamp.

### Note

This is for demonstration purposes only; for real problems or applications please use ode23 or rkf54.

### References

Quarteroni, A., and F. Saleri (2006). Scientific Computing With MATLAB and Octave. Second Edition, Springer-Verlag, Berlin Heidelberg.

ode23, newmark
  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 ## Newton's example f <- function(x, y) 1 - 3*x + y + x^2 + x*y sol100 <- cranknic(f, 0, 1, 0, N = 100) sol1000 <- cranknic(f, 0, 1, 0, N = 1000) ## Not run: # Euler's forward approach feuler <- function(f, t0, t1, y0, n) { h <- (t1 - t0)/n; x <- seq(t0, t1, by = h) y <- numeric(n+1); y[1] <- y0 for (i in 1:n) y[i+1] <- y[i] + h * f(x[i], y[i]) return(list(x = x, y = y)) } solode <- ode23(f, 0, 1, 0) soleul <- feuler(f, 0, 1, 0, 100) plot(soleul$x, soleul$y, type = "l", col = "blue", xlab = "", ylab = "", main = "Newton's example") lines(solode$t, solode$y, col = "gray", lwd = 3) lines(sol100$t, sol100$y, col = "red") lines(sol1000$t, sol1000$y, col = "green") grid() ## System of differential equations # "Herr und Hund" fhh <- function(x, y) { y1 <- y[1]; y2 <- y[2] s <- sqrt(y1^2 + y2^2) dy1 <- 0.5 - 0.5*y1/s dy2 <- -0.5*y2/s return(c(dy1, dy2)) } sol <- cranknic(fhh, 0, 60, c(0, 10)) plot(sol$y[, 1], sol$y[, 2], type = "l", col = "blue", xlab = "", ylab = "", main = '"Herr und Hund"') grid() ## End(Not run)