D:\annaconda\python.exe C:\Users\竹石子\Desktop\电工杯\3.py Traceback (most recent call last): File "C:\Users\竹石子\Desktop\电工杯\3.py", line 232, in <module> best_stage1 = stage1_sa(df_transfer, df_garbage, distance) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\竹石子\Desktop\电工杯\3.py", line 174, in stage1_sa neighbor_cost = total_cost(neighbor_sol) ^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\竹石子\Desktop\电工杯\3.py", line 140, in total_cost transport = transport_cost(solution) ^^^^^^^^^^^^^^^^^^^^^^^^ File "C:\Users\竹石子\Desktop\电工杯\3.py", line 130, in transport_cost Ck = vehicle_data['cost_per_km'].values[0] ~~~~~~~~~~~~^^^^^^^^^^^^^^^ File "D:\annaconda\Lib\site-packages\pandas\core\frame.py", line 4078, in __getitem__ return self._get_item_cache(key) ^^^^^^^^^^^^^^^^^^^^^^^^^ File "D:\annaconda\Lib\site-packages\pandas\core\frame.py", line 4639, in _get_item_cache res = self._ixs(loc, axis=1) ^^^^^^^^^^^^^^^^^^^^^^ File "D:\annaconda\Lib\site-packages\pandas\core\frame.py", line 4014, in _ixs result._set_as_cached(label, self) File "D:\annaconda\Lib\site-packages\pandas\core\series.py", line 1478, in _set_as_cached self._cacher = (item, weakref.ref(cacher)) ^^^^^^^^^^^^^^^^^^^ KeyboardInterrupt 进程已结束，退出代码为 -1073741510 (0xC000013A: interrupted by Ctrl+C)

从错误信息来看，程序在运行过程中遇到了一个 `KeyboardInterrupt` 中断（用户通过 Ctrl+C 强制终止了程序）。此外，在中断之前，程序尝试访问 `vehicle_data['cost_per_km'].values[0]` 时可能出现了问题。以下是详细的分析和解决方案。 --- ### 1. **问题分析** #### (1) `KeyboardInterrupt` - 错误信息表明程序运行时间过长或陷入死循环，用户被迫中断。 - 模拟退火算法（SA）的迭代次数可能过高，导致程序运行缓慢。 #### (2) `vehicle_data['cost_per_km'].values[0]` 访问问题 - 如果 `vehicle_data` 是空的（即没有找到匹配的车辆类型），则会抛出 `IndexError`。 - 这可能是因为垃圾类型 `k` 超出了车辆参数表中的范围。 --- ### 2. **解决方案** #### (1) 防止空数据访问 在 `transport_cost` 函数中，添加对 `vehicle_data` 是否为空的检查，并在空时抛出明确的异常。 #### (2) 控制模拟退火算法的运行时间 - 减少每次温度下降的迭代次数 `iter_per_temp`。 - 设置最大迭代次数或时间限制以避免长时间运行。 #### (3) 优化代码逻辑 - 确保所有输入数据的格式正确。 - 添加日志输出以便调试。 --- ### 3. **修复后的代码** 以下是修复后的代码： ```python import numpy as np import pandas as pd import math import random import time import matplotlib.pyplot as plt # 设置随机种子以保证结果可重复 random.seed(42) # 数据加载 def load_data(): # 收集点坐标（附件1） df_points = pd.read_excel('附件1.xlsx', sheet_name='附件130个垃圾分类收集点坐标及总垃圾量', skiprows=1) df_points.columns = ['id', 'x', 'y', 'w_total', 'remark'] # 中转站候选参数（附件4） df_transfer = pd.read_excel('中转站候选位置及参数.xlsx', sheet_name='附件4中转站候选位置及参数', skiprows=1) df_transfer.columns = ['id', 'x', 'y', 'cost', 'time_window', 'capacity'] df_transfer['capacity'] = df_transfer['capacity'].apply(eval) # 车辆参数（附件2） df_vehicles = pd.read_excel('4种运输车辆参数.xlsx', sheet_name='附件24类运输车辆参数', skiprows=1) df_vehicles.columns = ['type', 'garbage_type', 'max_weight', 'volume', 'cost_per_km', 'alpha', 'beta'] # 收集点垃圾分布（附件3） df_garbage = pd.read_excel('附件3.xlsx', sheet_name='附件330个收集点的4类垃圾量分布', skiprows=1) df_garbage.columns = ['id', 'k1', 'k2', 'k3', 'k4'] return df_points, df_transfer, df_vehicles, df_garbage # 距离矩阵计算（处理非对称） def build_distance_matrix(df_points, df_transfer): nodes = [] # 处理厂（0号） nodes.append({'id': 0, 'x': 0, 'y': 0}) # 收集点1-30 for _, row in df_points.iterrows(): if row['id'] == 0: continue nodes.append({'id': row['id'], 'x': row['x'], 'y': row['y']}) # 中转站31-35 for _, row in df_transfer.iterrows(): nodes.append({'id': row['id'], 'x': row['x'], 'y': row['y']}) # 构建距离字典 distance = {} for i in nodes: for j in nodes: key = (i['id'], j['id']) # 处理特殊路径（根据附件5） if (i['id'] == 4 and j['id'] == 31): distance[key] = 18 elif (i['id'] == 31 and j['id'] == 4): distance[key] = 15 else: dx = i['x'] - j['x'] dy = i['y'] - j['y'] distance[key] = round(math.hypot(dx, dy)) return distance # 第一阶段：中转站选址与分配 def stage1_sa(df_transfer, df_garbage, distance): # 模拟退火参数 T0 = 1000 alpha = 0.95 iter_per_temp = 50 # 减少每次温度下降的迭代次数 T_min = 1e-3 candidate_stations = df_transfer['id'].tolist() # 初始解 def initial_solution(): selected = random.sample(candidate_stations, 2) assignments = {} for i in range(1, 31): for k in range(1, 5): j = random.choice(selected) assignments[(i, k)] = j return {'selected': selected, 'assignments': assignments} # 计算容量违规 def capacity_penalty(solution): penalty = 0 for j in solution['selected']: cap = df_transfer[df_transfer['id'] == j]['capacity'].iloc[0] for k in range(4): total = sum(df_garbage[df_garbage['id'] == i][f'k{k+1}'].values[0] for (i, k_), j_ in solution['assignments'].items() if k_ == k+1 and j_ == j) if total > cap[k]: penalty += (total - cap[k]) * 1000 return penalty # 运输成本估计 def transport_cost(solution): cost = 0 for (i, k), j in solution['assignments'].items(): vehicle_data = df_vehicles[df_vehicles['type'] == k] if vehicle_data.empty: raise ValueError(f"未找到匹配的车辆类型: {k}") Ck = vehicle_data['cost_per_km'].values[0] d = distance[(0, j)] + distance[(j, i)] + distance[(i, 0)] w = df_garbage[df_garbage['id'] == i][f'k{k}'].values[0] cost += Ck * d * w return cost # 总成本 def total_cost(solution): construction = sum(df_transfer[df_transfer['id'].isin(solution['selected'])]['cost']) * 10000 transport = transport_cost(solution) penalty = capacity_penalty(solution) return construction + transport + penalty # 邻域生成 def neighbor(solution): new_sol = { 'selected': solution['selected'].copy(), 'assignments': solution['assignments'].copy() } op = random.choice(['add', 'remove', 'reassign']) if op == 'add': available = set(candidate_stations) - set(new_sol['selected']) if available: new_sol['selected'].append(random.choice(list(available))) elif op == 'remove': if len(new_sol['selected']) > 2: new_sol['selected'].remove(random.choice(new_sol['selected'])) elif op == 'reassign': i, k = random.choice(list(new_sol['assignments'].keys())) j = random.choice(new_sol['selected']) new_sol['assignments'][(i, k)] = j return new_sol # SA主循环 current = initial_solution() current_cost = total_cost(current) best = current.copy() best_cost = current_cost T = T0 start_time = time.time() # 记录开始时间 max_runtime = 60 # 最大运行时间（秒） while T > T_min and (time.time() - start_time) < max_runtime: for _ in range(iter_per_temp): neighbor_sol = neighbor(current) try: neighbor_cost = total_cost(neighbor_sol) except ValueError as e: print(e) # 打印错误信息并跳过当前邻居 continue delta = neighbor_cost - current_cost if delta < 0 or random.random() < math.exp(-delta / T): current = neighbor_sol current_cost = neighbor_cost if current_cost < best_cost: best = current.copy() best_cost = current_cost T *= alpha return best # 主程序 df_points, df_transfer, df_vehicles, df_garbage = load_data() distance = build_distance_matrix(df_points, df_transfer) best_stage1 = stage1_sa(df_transfer, df_garbage, distance) print("最佳方案:", best_stage1) ``` --- ### 4. **代码解释** 1. **防止空数据访问**： - 在 `transport_cost` 函数中，如果找不到匹配的车辆类型，则抛出明确的异常 `ValueError`。 2. **控制运行时间**： - 使用 `time.time()` 记录程序开始时间，并设置最大运行时间为 60 秒。 - 如果超过最大运行时间，则退出模拟退火主循环。 3. **优化逻辑**： - 减少了每次温度下降的迭代次数 `iter_per_temp`，以加快收敛速度。 --- ###