Implementation of Directed Search Domain (DSD) Multiobjective optimisation algorithm
DSD is a library written in the MATLAB programming language to support the searching for the Pareto solution in n-dimensional multiobjective optimisation context.
To start, there are two inputs which should be provided for the algorithm, (i) x0, which is the starting point with the same dimension of the deisgn variable space and (ii) deg, which is the degree of the generating cone; the enough smaller degree (e.g. 5 ) will provide a reasonable set of evenly distributed Pareto points. Therefore, the designer should start with:
[ParetoX, ParetoF] = Pareto_Generation(x0, deg)
To set the constraints and the objective functions, designer should refer to: const.m and Objective_Evaluation.m respectively. In addition, to set the bounds for each deisgn variables, there is a % Bounds% session provided inside the Pareto_Generation.m file.