人工鱼群算法的python代码

人工鱼群算法（Artificial Fish Swarm Optimization, AFSO）是一种模拟生物社会群体行为的优化算法，常用于解决复杂的全局优化问题。Python语言因其简洁性和丰富的库支持，非常适合实现此类算法。以下是一个简单的二维空间的人工鱼群搜索过程的Python代码示例：

import numpy as np

class Particle:
    def __init__(self, position, velocity, best_position):
        self.position = position
        self.velocity = velocity
        self.best_position = best_position

class PSO:
    def __init__(self, n_particles, search_space, max_iter, omega=0.7, c1=2, c2=2):
        self.particles = [Particle(np.random.uniform(search_space[0], search_space[1]), np.zeros(2), np.inf*np.ones(2)) for _ in range(n_particles)]
        self.search_space = search_space
        self.max_iter = max_iter
        self.c1, self.c2 = c1, c2
        self.w = omega

    def update(self):
        r1, r2 = np.random.rand(), np.random.rand() # 随机数
        for particle in self.particles:
            vel_new = self.w * particle.velocity + self.c1 * r1 * (particle.best_position - particle.position) + self.c2 * r2 * (self.global_best_position - particle.position)
            pos_new = particle.position + vel_new
            if pos_new[0] < self.search_space[0] or pos_new[0] > self.search_space[1]:
                vel_new[0] = -vel_new[0]
            if pos_new[1] < self.search_space[0] or pos_new[1] > self.search_space[1]:
                vel_new[1] = -vel_new[1]
            
            particle.position, particle.velocity = pos_new, vel_new
            if fitness(pos_new) < fitness(particle.best_position): # 更新个体最优解
                particle.best_position = pos_new
                
            if fitness(pos_new) < fitness(self.global_best_position): # 更新全局最优解
                self.global_best_position = pos_new

def fitness(position):
    # 这里应替换为你要优化的实际函数
    return position[0]**2 + position[1]**2  # 示例：计算位置到原点的距离平方

# 主程序
pso = PSO(50, (0, 100), 100)
for i in range(pso.max_iter):
    pso.update()
    print(f"Iteration {i+1}: Best Position: {pso.global_best_position}")