I tried to use MATLAB to optimize lookup tables in simulations. I need help with memory overflows..
1 view (last 30 days)
Show older comments
What I use is Rody Oldenhuis's GODLIKE optimization function (https://www.mathworks.com/matlabcentral/fileexchange/24838-godlike-a-robust-single---multi-objective-optimizer). It is pretty robust and works really well for most simple functions that I have tried. What I do now is to apply this to a simulink simulation with lookup tables. As a simple example, let us assume that the simulation has one input which goes through one lookup table and then goes to one output. A matlab function accepts the values for this lookup table as input, runs the simulation, reads the output and calculates the difference from a desired constant value (or error) and returns this error. This function becomes the "objective function" for the optimizer input. The output to be minimized is the error and the variable to be optimized are the values in the lookup table.
This logic runs pretty well in theory and also works ok with simple simulations like the one I mentioned. However a higher complexity function provides a memory overflow error after about 6000 - 7000 iterations. This is now what I want to rectify. My skills in memory overflow handling are not the best, so I would like to ask the community.
0 Comments
Answers (0)
See Also
Categories
Find more on Genetic Algorithm in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!