Research Article
An Evolutionary Multiagent Framework for Multiobjective Optimization
| | Input: : multiobjective function; : initial solutions set; : number of agent team; : initial agent team; : number of iterations; | | | Output: optimal agent team ; solutions in system memory ; | | (1) | Initial T = 0 and i = 0; | | (2) | Decompose population into | | (3) | Copy into ; | | (4) | while do | | (5) | for each do | | (6) | Decode an agent into a multi-objective cooperative co-evolutionary algorithm A; | | (7) | Choose a sub-population in as initial population of algorithm A; | | (8) | Use algorithm A to optimize based on the collaboration mechanism in ; | | (9) | Update through ; | | (10) | Obtain score and age of agent ; | | (11) | end for | | (12) | Use elite genetic algorithm to evolve the agent team based on the score and age; | | (13) | Update through ; | | (14) | T = T + 1; | | (15) | end while |
|
