heySourabh / WENO

This is a Java library for 1D Weighted Essentially Non-Oscillatory (WENO) reconstruction.

Weighetd Essentially Non-Oscillatory (WENO)

This is a Java library for one dimensional Weighted Essentially Non Oscillatory (WENO) reconstruction of an arbitrary order of accuracy. Download Library

You can use the library by passing the average-values and order-of-ENO-stencils as arguments.

ReconstructedValues values = new WENO().getWenoReconstructedValues(averageValues, k);

Example for 5th order WENO method (k=3):

int k = 3; // Order of ENO = 3 = 5th order WENO
int numNeigh = k - 1; // number of neighbours on each side
// populate the average values from neighbourhood
double[] averageValues = new double[2 * k - 1];
for (int j = -numNeigh; j <= numNeigh; j++) {
    averageValues[j + numNeigh] = u[i + j];
}
// Get the reconstructed values:
ReconstructedValues values = new WENO().getWenoReconstructedValues(averageValues, k);
// Read the values from ReconstructedValues object
u_imh = values.value_imh; // u at i-1/2
u_iph = values.value_iph; // u at i+1/2

Example for 9th order WENO method (Same code as above with k=5):

int k = 5; // Order of ENO = 5 = 9th order WENO
int numNeigh = k - 1; // number of neighbours on each side
// populate the average values from neighbourhood
double[] averageValues = new double[2 * k - 1];
for (int j = -numNeigh; j <= numNeigh; j++) {
    averageValues[j + numNeigh] = u[i + j];
}
// Get the reconstructed values:
ReconstructedValues values = new WENO().getWenoReconstructedValues(averageValues, k);
// Read the values from ReconstructedValues object
u_imh = values.value_imh; // u at i-1/2
u_iph = values.value_iph; // u at i+1/2