Time limit in a programatically programmed parameter estimation
显示 更早的评论
I have defined an estimation object to identify the best parameter set for a simulink model to fit a desired transient behavior. However, since the optimization problem is highly non-linear, it is far from being convex and many local minima appears. As a result of this, when a pattern search algorithm is selected the simulink model to be identified may enter in regions of the parameter space where some model assumptions are invalid, the model behavior become erratic and it stalls. When this situation arises, the only solution is to restart Matlab and therefore no solution can be found to the problem.
In this connection, is there a way to fix a time limit (as it exist for Genetic Algorithms) for the optimization algorithm in order to avoid Matlab crashes?
回答（0 个）
类别
在 帮助中心 和 File Exchange 中查找有关 Genetic Algorithm 的更多信息
2012-1-27
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
Translated by 
