2D Lookup Table from a 6D Lookup Table via Interpolation

Hi everyone,

I have a question about lookup tables. I have a lookup Table for some aerodynamic data, that is depentend on 6 input parameters, so I have a 6D Lookup Table. If i know all the 6 parameters i can calculate the aerodynamic data via interpolation.

But i only have 4 input parameters, the other 2 are still unknown. So i want to create a 2D Lookup Table from the 6D one, that is dependent on the 2 unknown parameters.

What Matlab Function can you recomend me for this problem ?

Thank you so much :)

Answer 1

[Divyam]

Hi @Lennart V,

Assuming that you are looking to create the following 2D table from the 6D table by fixing the first 4 dimensions you can use the "interpn" or the "griddedInterpolant" functions for your use case.

Example: Using the "interpn" function

% Define fixed values for the known dimensions
x1_val = 0.2;
x2_val = 0.5;
x3_val = 0.1;
x4_val = 1.0;

% Create a meshgrid for the 2D unknowns. Must match 5th and 6th grid
[X5q, X6q] = meshgrid(X5, X6);

% Interpolate over the 6D table
G = interpn(X1, X2, X3, X4, X5, X6, F, x1_val, x2_val, x3_val, x4_val, X5q, X6q);

Here G is a 2D matrix of the same size as , which represents the aerodynamic data over the two unknowns , holding the rest fixed.

For repeated queries with different fixed values, the "griddedInterpolant" function is more efficient:

F_interp = griddedInterpolant({X1,X2,X3,X4,X5,X6}, F);

G = arrayfun(@(x5,x6) F_interp(x1_val,x2_val,x3_val,x4_val,x5,x6), X5q, X6q);

For more information regarding the "interpn" and the "griddedInterpolant" function, refer to the following documentation:

https://www.mathworks.com/help/matlab/ref/interpn.html

https://www.mathworks.com/help/matlab/ref/griddedinterpolant.html