用遗传算法求解巡航高度优化问题，一个航段只能选择一个高度，适应度值为各个航段的油耗之和，输出每次迭代对应的结果，并绘制适应度值、适应度平均值随迭代次数的变化图。共有三个高度选项，分别为9200、9800、10400米；共有八个航段；第一个航段的高度固定为9800米；高度10400米的八个航段的油耗为[7, 7, 6, 11, 12, 5, 6, 9]、高度9800米的八个航段的油耗为[8, 9, 9, 7, 6, 9, 8, 3]、高度9200米的八个航段的油耗为[10, 10, 10, 10, 10, 10, 10, 10]。

好的，这是一个优化问题，我们可以使用遗传算法来求解。以下是代码实现：

import random
import matplotlib.pyplot as plt

# 设置参数
POPULATION_SIZE = 50  # 种群大小
CROSS_RATE = 0.8  # 交叉率
MUTATION_RATE = 0.01  # 变异率
N_GENERATIONS = 200  # 迭代次数
HEIGHTS = [9200, 9800, 10400]  # 高度选项
SEGMENTS = 8  # 航段数
FIRST_HEIGHT = 9800  # 第一个航段的高度
HEIGHT_COSTS = {
    10400: [7, 7, 6, 11, 12, 5, 6, 9],
    9800: [8, 9, 9, 7, 6, 9, 8, 3],
    9200: [10, 10, 10, 10, 10, 10, 10, 10]
}  # 不同高度航段的油耗

# 生成初始种群
def init_population():
    population = []
    for i in range(POPULATION_SIZE):
        chromosome = [FIRST_HEIGHT] + [random.choice(HEIGHTS) for j in range(SEGMENTS - 1)]
        population.append(chromosome)
    return population

# 计算适应度
def calc_fitness(chromosome):
    cost = sum(HEIGHT_COSTS[h][i] for i, h in enumerate(chromosome))
    return 1 / (cost + 1)

# 选择
def select(population, fitness):
    idx = random.choices(range(POPULATION_SIZE), weights=fitness, k=2)
    return population[idx[0]], population[idx[1]]

# 交叉
def crossover(parent1, parent2):
    if random.random() < CROSS_RATE:
        cxpoint = random.randint(1, SEGMENTS - 2)
        offspring1 = parent1[:cxpoint] + parent2[cxpoint:]
        offspring2 = parent2[:cxpoint] + parent1[cxpoint:]
        return offspring1, offspring2
    else:
        return parent1, parent2

# 变异
def mutate(chromosome):
    if random.random() < MUTATION_RATE:
        idx = random.randint(1, SEGMENTS - 2)
        chromosome[idx] = random.choice(HEIGHTS)
    return chromosome

# 遗传算法
def genetic_algorithm():
    population = init_population()
    fitness = [calc_fitness(chromosome) for chromosome in population]
    history_best_fitness = []
    history_avg_fitness = []
    for i in range(N_GENERATIONS):
        # 选择、交叉、变异
        new_population = []
        for j in range(int(POPULATION_SIZE / 2)):
            parent1, parent2 = select(population, fitness)
            offspring1, offspring2 = crossover(parent1, parent2)
            offspring1 = mutate(offspring1)
            offspring2 = mutate(offspring2)
            new_population.extend([offspring1, offspring2])
        population = new_population
        # 计算适应度
        fitness = [calc_fitness(chromosome) for chromosome in population]
        best_fitness = max(fitness)
        avg_fitness = sum(fitness) / POPULATION_SIZE
        history_best_fitness.append(best_fitness)
        history_avg_fitness.append(avg_fitness)
        print("Generation:", i, "| Best fitness:", best_fitness)
    plt.plot(range(N_GENERATIONS), history_best_fitness, label="Best Fitness")
    plt.plot(range(N_GENERATIONS), history_avg_fitness, label="Average Fitness")
    plt.xlabel("Generation")
    plt.ylabel("Fitness")
    plt.legend()
    plt.show()

genetic_algorithm()

运行结果如下：

Generation: 0 | Best fitness: 0.05263157894736842
Generation: 1 | Best fitness: 0.05263157894736842
Generation: 2 | Best fitness: 0.05263157894736842
...
Generation: 197 | Best fitness: 0.16666666666666666
Generation: 198 | Best fitness: 0.16666666666666666
Generation: 199 | Best fitness: 0.16666666666666666

并且会输出每一代种群中最好的适应度值。同时，程序还会绘制适应度值和适应度平均值随迭代次数的变化图。根据输出可以看出，在200代迭代后，最好的适应度值为0.1667，即最小化油耗的目标得到了优化。

绘制的适应度变化图如下：