| 查看: 269 | 回复: 0 | ||
[求助]
有关鱼群算法流程的文献，有几句不是很懂，求虫友指导
|
|
1 To generate initialization groups: set the initial bulletin board iterations which record the best state of artificial fish not changed or changed little: Beststep←0, the initial number of iterations Num. ← 0, N which represents the number of artificial fish is randomly generated under controlling. 2 To calculate initial value on bulletin board: calculate the initial the artificial f  ish value in the initial fish swarm,，and compare the value, take Y as the minimum to enter the bulletin board, this fish is assigned to the bulletin board.3 To choices behavior: the artificial fish respectively Simulates rear-end behavior and clusters behavior, comparing the value, the smaller Y was selected, the acts accordingly are implemented, the default behavior is foraging behavior. 4 Bulletin board: after each operation, testing their value among them, if superior, replace it, and set Beststep←0. 5 Mutation of conditional: if the number of the value Beststep do not change reached the maximum threshold maxbest consecutively, the implementation goes to step (6); otherwise go to step (7). 6 Mutation operation: In addition to the fish in the best individual outside bulletin board, all other artificial fish follows these steps. 6.1The dimension of all the various artificial fish generates random numbers r∈(0,1)，is mutation probability. The dimension of the individual is randomly initialized; otherwise the dimensions of the artificial fish remain the same. 6.2 To calculate the value Y of the formation of new artificial fish, and to compare with the optimal value on the bulletin board, if better than the bulletin board, the new artificial replaces it. 6.3 To set Beststep←0. 7 Termination of conditional: if Num has reached the preset maximum number of iterations maxnumber or the optimal solution is satisfactory within the error bounds|y_n-y_(n+1) |<ε,ε>0 is a very small positive number, if not met, then Num← Num+1,Beststep←Beststep+ 1, go to step (3)，the next fish swarm optimization process begins; otherwise goes to step (8). 8 To terminate the algorithm, output the optimal solution. |
回复此楼» 猜你喜欢
现代”学阀”该如何界定
已经有14人回复
-
求合成方法
已经有8人回复
-
师弟论文见刊大半年才想起来申请专利，还能抢救一下吗？
已经有4人回复
-
上海工程技术大学 激光智能制造课题组 2027级博士研究生招生
已经有5人回复
-
课题组招2027级博士 上海工程技术大学 激光智能制造方向
已经有5人回复
-
有机合成以后会不会被AI改变？做科研的虫友怎么看
已经有8人回复
-
我的奶奶
已经有3人回复