PSO，粒子群算法

最新推荐文章于 2024-07-16 14:14:36 发布

.云哲.

最新推荐文章于 2024-07-16 14:14:36 发布

阅读量1k

点赞数

分类专栏：生物进化算法

本文链接：https://blog.csdn.net/luolinll1212/article/details/88529410

版权

生物进化算法专栏收录该内容

1 篇文章 0 订阅

订阅专栏

import numpy as np

def f1(x):
    return -x**2 + 3


EO = 1e-6
ps = 50
dim = 1
MaxNum = 50
x_bound = [0, 20]
v_bount = [-1, 1]
w = 0.8
c1 = c2 = 0.5
x = np.random.uniform(np.min(x_bound), np.max(x_bound), (ps, dim))
v = np.random.rand(ps, dim)
f = f1(x)
p_p = x
p_f = f
g_b_p = x[np.argmax(f)]
g_b_f = np.max(f)

for step in range(MaxNum):
    r1 = np.random.rand(ps, dim)
    r2 = np.random.rand(ps, dim)
    v = w * v + c1 * r1 * (p_p - x) + c2 * r2 * (g_b_p - x)
    v[v > np.max(v_bount)] = np.max(v_bount)
    v[v < np.min(v_bount)] = np.min(v_bount)

    x = x + v
    x[x > np.max(x_bound)] = np.max(x_bound)
    x[x < np.min(x_bound)] = np.min(x_bound)

    f = f1(x)
    p_g_i = np.greater(f, p_f)
    p_p[p_g_i] = x[p_g_i]
    p_f[p_g_i] = f[p_g_i]
    if np.max(p_f) > g_b_f:
        g_b_p = p_p[np.argmax(p_f)]
        g_b_f = np.max(p_f)