An implementation of multi objective particle swarm optimization technique for a minimization problem
In this project there are two ways that I have implemented nonlinear constraints:
- In MOPSO1 constraints are computed with objectives in one file and a zero or positive infeasabilty value is assinged to a particle where zero means it is feasable.
- In MOPSO2 constraints are in other function and if a particle is infeasable the onjective function will not be evaluated in order to save computation time, and objective values of that particle will be NaN.