【python项目】中国五大城市PM2.5 数据分析（含数据集及完整代码）

import pandas as pd import numpy as np import matplotlib.pyplot as plt beijing_data = pd.read_csv(r"D:\PM2.5data\BeijingPM20100101_20151231.csv") chengdu_data = pd.read_csv(r"D:\PM2.5data\ChengduPM20100101_20151231.csv") guangzhou_data = pd.read_csv(r"D:\PM2.5data\GuangzhouPM20100101_20151231.csv") shanghai_data = pd.read_csv(r"D:\PM2.5data\ShanghaiPM20100101_20151231.csv") shenyang_data = pd.read_csv(r"D:\PM2.5data\ShenyangPM20100101_20151231.csv") # 设置Matplotlib的字体为"SimHei"以支持中文显示 plt.rcParams['font.sans-serif'] = ['SimHei'] plt.rcParams['axes.unicode_minus'] = False city_data_list = [beijing_data, shanghai_data, guangzhou_data, chengdu_data, shenyang_data] # 假设您的五个城市数据集已经被加载 beijing_data['PM2.5'] =beijing_data[['PM_Dongsi', 'PM_Dongsihuan', 'PM_Nongzhanguan']].mean(axis=1) shanghai_data['PM2.5'] =shanghai_data[['PM_Jingan', 'PM_Xuhui']].mean(axis=1) guangzhou_data['PM2.5'] =guangzhou_data[['PM_City Station','PM_5th Middle School']].mean(axis=1) chengdu_data['PM2.5'] =chengdu_data[['PM_Caotangsi','PM_Shahepu']].mean(axis=1) shenyang_data['PM2.5'] =shenyang_data[['PM_Taiyuanjie','PM_Xiaoheyan']].mean(axis=1) data_city1=beijing_data.groupby(['year', 'month', 'day'])['PM2.5'].mean().reset_index() data_city2=shanghai_data.groupby(['year', 'month', 'day'])['PM2.5'].mean().reset_index() data_city3=guangzhou_data.groupby(['year', 'month', 'day'])['PM2.5'].mean().reset_index() data_city4=chengdu_data.groupby(['year', 'month', 'day'])['PM2.5'].mean().reset_index() data_city5=shenyang_data.groupby(['year', 'month', 'day'])['PM2.5'].mean().reset_index() data_city1_us=beijing_data.groupby(['year', 'month', 'day'])['PM_US Post'].mean().reset_index() data_city2_us=shanghai_data.groupby(['year', 'month', 'day'])['PM_US Post'].mean().reset_index() data_city3_us=guangzhou_data.groupby(['year', 'month', 'day'])['PM_US Post'].mean().reset_index() data_city4_us=chengdu_data.groupby(['year', 'month', 'day'])['PM_US Post'].mean().reset_index() data_city5_us=shenyang_data.groupby(['year', 'month', 'day'])['PM_US Post'].mean().reset_index() data_cities_us=[data_city1_us,data_city2_us,data_city3_us,data_city4_us,data_city5_us] avg_us = [] for data in city_data_list: daily_avg = data.groupby(['year', 'month', 'day'])['PM_US Post'].mean() overall_avg = daily_avg.mean() avg_us.append(overall_avg) cities_data =[data_city1, data_city2, data_city3, data_city4, data_city5] city_names = ["北京","上海","广州","成都","沈阳"] avg_pm25 = [data['PM2.5'].mean() for data in cities_data] exceed_days = [data[data['PM2.5'] > 75].shape[0] for data in cities_data] exceed_days_us = [data[data['PM_US Post'] > 75].shape[0] for data in data_cities_us] # 设定颜色方案 bar_colors = ['#08306b', '#08519c', '#2171b5', '#4292c6', '#6baed6'] # 从深蓝到浅蓝的渐变色 background_color = '#f7fbff' # 轻柔的背景颜色 # 创建2x1的布局，调整figsize的参数 fig, axes = plt.subplots(nrows=2, ncols=1, figsize=(10, 8)) fig.patch.set_facecolor(background_color) # 设置背景颜色 for ax in axes: ax.set_facecolor(background_color) # 设置子图的背景颜色 ax.spines['top'].set_color('gray') ax.spines['right'].set_color('gray') ax.spines['bottom'].set_color('gray') ax.spines['left'].set_color('gray') # 在第一个子图中绘制五个城市的PM2.5平均数值 bars1 = axes[0].bar(city_names, avg_pm25, color=bar_colors, width=0.6) # 使用渐变色 for bar, value in zip(bars1, avg_pm25): height = bar.get_height() axes[0].text(bar.get_x() + bar.get_width() / 2, height / 2, f'{value:.2f}', ha='center', color='black', fontsize=10) # 调整字体大小 for bar, days in zip(bars1, exceed_days): height = bar.get_height() axes[0].text(bar.get_x() + bar.get_width() / 2, height + 2, f'超标天数: {int(days)}', ha='center', fontsize=10) # 调整字体大小 axes[0].axhline(y=75, color='r', linestyle='--') axes[0].set_xlabel('城市', fontsize=12) axes[0].set_ylabel('PM2.5 平均数值', fontsize=12) axes[0].set_title('五个城市的日PM2.5平均数值及超标天数', fontsize=14) # 在第二个子图中绘制美国观测的各城市平均数值 bars2 = axes[1].bar(city_names, avg_us, color=bar_colors, width=0.6) # 使用渐变色 for bar, value in zip(bars2, avg_us): height = bar.get_height() axes[1].text(bar.get_x() + bar.get_width() / 2, height / 2, f'{value:.2f}', ha='center', color='black', fontsize=10) # 调整字体大小 for bar, days in zip(bars2, exceed_days_us): height = bar.get_height() axes[1].text(bar.get_x() + bar.get_width() / 2, height + 2, f'超标天数: {int(days)}', ha='center', fontsize=10) # 调整字体大小 axes[1].axhline(y=75, color='r', linestyle='--') axes[1].set_xlabel('城市', fontsize=12) axes[1].set_ylabel('PM2.5 平均数值 (美国观测)', fontsize=12) axes[1].set_title('美国观测的各城市日PM2.5平均数值及超标天数', fontsize=14) plt.tight_layout() plt.show()