蚁群算法python代码_Python蚁群算法代码范例+注释

以下是一个简单的蚁群算法Python代码范例，带有注释说明：

import random

# 定义城市数量
num_cities = 10

# 定义城市之间的距离矩阵
distance_matrix = []
for i in range(num_cities):
    row = []
    for j in range(num_cities):
        if i == j:
            row.append(0)
        else:
            row.append(random.randint(1, 100))
    distance_matrix.append(row)

# 定义参数
num_ants = 5
alpha = 1  # 信息素重要程度因子
beta = 2  # 启发式因子
evaporation = 0.5  # 信息素挥发速度
Q = 1  # 信息素增加强度系数
num_iterations = 10

# 初始化信息素矩阵
pheromone_matrix = [[1 / (num_cities * num_cities) for j in range(num_cities)] for i in range(num_cities)]

# 定义函数，用于选择下一个城市
def select_next_city(current_city, visited_cities, pheromone_matrix, distance_matrix):
    unvisited_cities = [i for i in range(num_cities) if i not in visited_cities]
    probabilities = [0 for i in range(num_cities)]
    total_probabilities = 0
    for city in unvisited_cities:
        probabilities[city] = (pheromone_matrix[current_city][city] ** alpha) * ((1 / distance_matrix[current_city][city]) ** beta)
        total_probabilities += probabilities[city]
    if total_probabilities == 0:
        return random.choice(unvisited_cities)
    probabilities = [probability / total_probabilities for probability in probabilities]
    r = random.random()
    current_probability = 0
    for i in range(len(probabilities)):
        current_probability += probabilities[i]
        if r <= current_probability:
            return unvisited_cities[i]

# 定义函数，用于更新信息素矩阵
def update_pheromone_matrix(pheromone_matrix, ant_paths, distance_matrix, evaporation, Q):
    for i in range(num_cities):
        for j in range(num_cities):
            pheromone_matrix[i][j] *= (1 - evaporation)
    for ant_path, ant_distance in ant_paths:
        for i in range(num_cities - 1):
            pheromone_matrix[ant_path[i]][ant_path[i+1]] += Q / ant_distance
        pheromone_matrix[ant_path[num_cities-1]][ant_path[0]] += Q / ant_distance

# 开始迭代
for iteration in range(num_iterations):
    # 每只蚂蚁都从一个随机城市出发
    ant_paths = []
    for ant in range(num_ants):
        current_city = random.randint(0, num_cities-1)
        visited_cities = [current_city]
        ant_path = [current_city]
        ant_distance = 0
        # 选择下一个城市，直到所有城市都被访问过
        while len(visited_cities) < num_cities:
            next_city = select_next_city(current_city, visited_cities, pheromone_matrix, distance_matrix)
            visited_cities.append(next_city)
            ant_path.append(next_city)
            ant_distance += distance_matrix[current_city][next_city]
            current_city = next_city
        ant_distance += distance_matrix[ant_path[num_cities-1]][ant_path[0]]
        ant_paths.append((ant_path, ant_distance))
    # 更新信息素矩阵
    update_pheromone_matrix(pheromone_matrix, ant_paths, distance_matrix, evaporation, Q)
    # 打印最佳路径
    ant_paths.sort(key=lambda x: x[1])
    print(f"Best path after iteration {iteration}: {ant_paths[0][0]}, distance: {ant_paths[0][1]}")

这个代码实现了一个简单的蚁群算法，用于求解旅行商问题。在代码中，首先定义了城市数量和城市之间的距离矩阵。然后定义了一些参数，包括蚂蚁数量、信息素重要程度因子、启发式因子、信息素挥发速度、信息素增加强度系数和迭代次数等。接着初始化了信息素矩阵。然后定义了两个函数，一个用于选择下一个城市，另一个用于更新信息素矩阵。最后在迭代中，每只蚂蚁都从一个随机城市出发，选择下一个城市，直到所有城市都被访问过，然后更新信息素矩阵。迭代结束后，打印最佳路径。