Research Article

A Knee Point-Driven Many-Objective Evolutionary Algorithm with Adaptive Switching Mechanism

Algorithm 3

Finding_knee_point (, , , and ).
Input:F (sorted population), T (rate of knee points in population), r, t (adaptive parameters)
1 K ← ∅ / knee points /
2 For allFiFdo
3   EFind_extreme_solution(Fi)
4   LCalculate_extreme_hyperplane(E)
5   Update r by Eq. (9)
6   fmax ← Maximum value of each objective in Fi
7   fmin ← Minimum value of each objective in Fi
8   Calculate V by Eq. (8)
9   Calculate the distance from each solution in Fi to L by Eq. (7)
10   Sort Fi in a descending order according to the distances
11   SizeFi
12   For allpFido
13      
14      KK ⋃ {p}
15      FiFiNB
16   End For
17   
18 End for
19 ReturnK, r and t