PROJECT TITLE :

Balancing Convergence and Diversity in Decomposition-Based Many-Objective Optimizers

ABSTRACT:

The decomposition-primarily based multiobjective evolutionary algorithms (MOEAs) typically build use of aggregation functions to decompose a multiobjective optimization downside into multiple single-objective optimization problems. But, thanks to the character of contour lines for the adopted aggregation functions, they usually fail to preserve the diversity in high-dimensional objective house even by using numerous weight vectors. To address this downside, we tend to propose to take care of the specified diversity of solutions in their evolutionary method explicitly by exploiting the perpendicular distance from the solution to the weight vector in the objective space, that achieves better balance between convergence and variety in many-objective optimization. The idea is implemented to reinforce two well-performing decomposition-primarily based algorithms, i.e., MOEA, based on decomposition and ensemble fitness ranking. The two enhanced algorithms are compared to several state-of-the-art algorithms and a series of comparative experiments are conducted on a number of test issues from two well-known test suites. The experimental results show that the two proposed algorithms are usually a lot of effective than their predecessors in balancing convergence and diversity, and they're conjointly terribly competitive against alternative existing algorithms for solving many-objective optimization problems.