Research Article
Decomposition-Based Multiobjective Evolutionary Optimization with Adaptive Multiple Gaussian Process Models
| (1) | Initialize N subproblems including weight vector , N individuals, and neighborhood size T | | (2) | Initialize the reference point , j = 1, 2, …, m | | (3) | Predefine | | (4) | while not terminate do | | (5) | for do | | | Model construction | | (6) | if then | | (7) | Define neighborhood individuals | | (8) | else | | (9) | Select individuals from to construct B | | (10) | end | | | Model updation | | (11) | if then | | (12) | Update using equations (5)–(8) | | (13) | end | | (14) | Select a Gaussian model according to | | (15) | Generate y based on the selected Gaussian model | | (16) | Add B and y into Algorithm 1 to generate a new offspring x | | | Population updating | | (17) | Set counter c = 0 | | (18) | for each do | | (19) | if and then | | (20) | is replaced by | | (21) | Set c = c+1 | | (22) | end | | (23) | end | | (24) | end | | (25) | G = G + 1 | | (26) | Update the reference point | | (27) | end | | (28) | Output the final population |
|
