pso代码

以下是一个简单的PSO算法的Python实现，可以用于解决优化问题：

import numpy as np

class Particle:
    def __init__(self, bounds):
        self.position = np.random.uniform(bounds[0], bounds[1])
        self.velocity = np.zeros_like(self.position)
        self.best_position = self.position
        self.best_fitness = np.inf

    def update(self, omega, phi_p, phi_g, swarm_best_position):
        r_p = np.random.uniform(size=self.position.shape)
        r_g = np.random.uniform(size=self.position.shape)
        self.velocity = omega * self.velocity \
                        + phi_p * r_p * (self.best_position - self.position) \
                        + phi_g * r_g * (swarm_best_position - self.position)
        self.position += self.velocity

    def evaluate(self, fitness_func):
        fitness = fitness_func(self.position)
        if fitness < self.best_fitness:
            self.best_position = self.position
            self.best_fitness = fitness

class PSO:
    def __init__(self, bounds, num_particles, max_iter, fitness_func):
        self.bounds = bounds
        self.num_particles = num_particles
        self.max_iter = max_iter
        self.fitness_func = fitness_func
        self.swarm = [Particle(bounds) for _ in range(num_particles)]
        self.swarm_best_position = np.inf * np.ones_like(bounds[0])
        self.swarm_best_fitness = np.inf

    def optimize(self, omega=0.5, phi_p=0.5, phi_g=0.5):
        for i in range(self.max_iter):
            for particle in self.swarm:
                particle.evaluate(self.fitness_func)
                if particle.best_fitness < self.swarm_best_fitness:
                    self.swarm_best_position = particle.best_position
                    self.swarm_best_fitness = particle.best_fitness
            for particle in self.swarm:
                particle.update(omega, phi_p, phi_g, self.swarm_best_position)
        return self.swarm_best_position, self.swarm_best_fitness

使用示例：

# 定义要优化的函数
def sphere(x):
    return np.sum(x**2)

# 实例化PSO类并运行PSO算法
bounds = (-5.12, 5.12)
num_particles = 50
max_iter = 100
pso = PSO(bounds, num_particles, max_iter, sphere)
best_position, best_fitness = pso.optimize()
print('Optimal solution:', best_position)
print('Optimal value:', best_fitness)