求解带约束条件的遗传算法Python代码

以下是一个简单的带约束条件的遗传算法的Python代码实现，其中包含一个简单的约束条件函数：

import random

POPULATION_SIZE = 100
GENERATIONS = 50
CROSSOVER_PROBABILITY = 0.8
MUTATION_PROBABILITY = 0.2
ELITISM = True
ELITE_SIZE = 5
LOWER_BOUND = 0
UPPER_BOUND = 10

def fitness(chromosome):
    # 计算适应度值
    return sum(chromosome)

def constraint(chromosome):
    # 约束条件函数，返回 True 表示满足所有约束条件，返回 False 表示不满足
    return sum(chromosome) <= 15

def initialize_population():
    # 初始化种群
    return [[random.uniform(LOWER_BOUND, UPPER_BOUND) for _ in range(10)] for _ in range(POPULATION_SIZE)]

def selection(population):
    # 轮盘赌选择
    fitness_sum = sum(fitness(chromosome) for chromosome in population)
    selection_probabilities = [fitness(chromosome) / fitness_sum for chromosome in population]
    return random.choices(population, weights=selection_probabilities, k=2)

def crossover(parents):
    # 一点交叉
    if random.random() < CROSSOVER_PROBABILITY:
        crossover_point = random.randint(0, len(parents[0])-1)
        child1 = parents[0][:crossover_point] + parents[1][crossover_point:]
        child2 = parents[1][:crossover_point] + parents[0][crossover_point:]
        return child1, child2
    else:
        return parents

def mutation(chromosome):
    # 突变，随机将某个基因替换为随机值
    if random.random() < MUTATION_PROBABILITY:
        mutation_point = random.randint(0, len(chromosome)-1)
        chromosome[mutation_point] = random.uniform(LOWER_BOUND, UPPER_BOUND)
    return chromosome

def evolve(population):
    # 进化
    new_population = []
    if ELITISM:
        # 保留精英
        elite = sorted(population, key=fitness, reverse=True)[:ELITE_SIZE]
        new_population.extend(elite)
    while len(new_population) < POPULATION_SIZE:
        parents = selection(population)
        children = crossover(parents)
        mutated_children = [mutation(child) for child in children]
        if constraint(mutated_children[0]):
            new_population.append(mutated_children[0])
        if constraint(mutated_children[1]):
            new_population.append(mutated_children[1])
    return new_population

if __name__ == '__main__':
    population = initialize_population()
    for generation in range(GENERATIONS):
        population = evolve(population)
        best_chromosome = max(population, key=fitness)
        print(f'Generation: {generation+1}, Best fitness: {fitness(best_chromosome)}, Best chromosome: {best_chromosome}')

在这个例子中，我们定义了一个目标函数 fitness(chromosome)，它计算染色体的适应度值。约束函数 constraint(chromosome) 返回 True 表示染色体满足所有约束条件，返回 False 表示不满足。initialize_population() 函数初始化种群，selection() 函数使用轮盘赌选择进行选择，crossover() 函数进行一点交叉，mutation() 函数进行基因突变。evolve() 函数将这些操作组合在一起来生成新的种群。在每一代中，我们都记录最佳染色体的适应度值和染色体本身。