2020数学建模E组校园供水系统智能管理

*附件是某校区水表层级关系以及所有水表四个季度的读数(以一定时间为间隔，如15分钟)与相应的用水数据。请利用这些信息和数据，建立数学模型，讨论以下问题：

1.统计、分析各个水表数据的变化规律，并给出校园内不同功能区（宿舍、教学楼、办公楼、食堂等）的用水特征。*

数据格式

数据总共三百多万条数据，要分析的是水表的用量特征

各个区域总用水量的分析结果及代码实现

# coding: utf-8

# In[1]:


import pandas as pd
data1 = pd.read_csv('season1.csv')

data2 = pd.read_csv('season2.csv')
data3 = pd.read_csv('season3.csv')
data4 = pd.read_csv('season4.csv')
data1.head(5)


# In[2]:


data1.describe()


# In[3]:


data1.info()


# In[4]:




data1.head(5)


# In[5]:


data1.head(5)


# In[6]:


data1['水表名'].unique()


# In[7]:


lever_table = pd.read_excel("附件_水表层级.xlsx")


# In[8]:


lever_table.to_csv('lever_table.csv',index=False)


# In[9]:


level = pd.read_csv('lever_table.csv')


# In[10]:


level.head(5)


# In[11]:


level['水表名'].unique()


# In[12]:


len(level['水表名'].unique()
   )


# In[13]:


len(data1['水表名'].unique())


# In[14]:


len(data2['水表名'].unique())


# In[15]:


len(data3['水表名'].unique())


# In[16]:


len(data4['水表名'].unique())


# In[17]:


data2.info()


# In[18]:


data3.info()


# In[19]:


data4.info()


# In[20]:


import numpy as np
len(data1.index)
data1['季度'] = pd.Series(["一季度" for i in range(len(data1.index))])

data1


# In[21]:


data2['季度'] = pd.Series(["二季度" for i in range(len(data2.index))])
data3['季度'] = pd.Series(["三季度" for i in range(len(data3.index))])
data4['季度'] = pd.Series(["四季度" for i in range(len(data4.index))])


# In[22]:


data4


# In[23]:


data = data1.append(data2, ignore_index=True)
data


# In[24]:


data = data.append([data3,data4],ignore_index=True)


# In[25]:


len(data['水表名'].unique())


# In[26]:


len(level['水表名'].unique())


# In[27]:


len(data1['水表名'].unique())


# In[28]:


len(data['水表号'].unique())


# In[29]:


data.head(4)


# In[30]:


level.head(4)


# In[31]:


data['水表名'].unique()


# In[32]:


len(level.index)


# In[33]:


use_water = data.groupby(by='水表名')['用量'].sum()


# In[34]:


use_water.sort_values(ascending=False)


# In[35]:


use_water.plot.hist()


# In[36]:


list_table_name = list(data['水表名'].unique())
list_table_name


# In[37]:


list_home = []
list_teaching_build = []
list_teacher_build = []
for name in list_table_name.copy():
    if name.find("学生宿舍") != -1 or name.find("留学生楼") != -1:
        list_home.append(name)
        list_table_name.remove(name)
list_home


# In[38]:


list_table_name
for name in list_table_name.copy():
    if name.find("XX") != -1 and name.find("楼") != -1:
        print(name)
        list_teaching_build.append(name)
        list_table_name.remove(name)
list_teaching_build


# In[39]:


list_table_name
for name in list_table_name.copy():
    if name.find("XX") != -1 and name.find("花") == -1:
        list_teacher_build.append(name)
        list_table_name.remove(name)
list_teacher_build


# In[40]:


list_agritural = []
for name in list_table_name.copy():
    if name.find("养殖") != -1 or name.find("养鱼") != -1 or name.find("大棚")!=-1 or name.find("花")!=-1:
        list_agritural.append(name)
        list_table_name.remove(name)
list_agritural


# In[41]:


for name in list_table_name.copy():
    if name.find("楼")!= -1:
        list_teaching_build.append(name)
        list_table_name.remove(name)
list_table_name


# In[42]:


list_backup = list_table_name
name_list = data['水表名'].tolist()  
name_list
type_list = []
for name in name_list:
    if name in list_backup:
        type_list.append("后勤")
    elif name in list_agritural:
        type_list.append("农业")
    elif name in list_home:
        type_list.append('宿舍')
    elif name in list_teacher_build:
        type_list.append("活动地方")
    else:
        type_list.append('教学楼')
type_list


# In[43]:


data['类型'] = pd.Series(type_list)
data


# In[44]:


type_sum = data.groupby(by='类型')['用量'].sum()


# In[45]:


import matplotlib  as mpl
mpl.rcParams['font.sans-serif'] = ['SimHei']
type_sum.plot.barh(alpha=0.7)


# In[48]:


df_group_two = data.groupby(by=['类型'])
from matplotlib import pyplot as plt
i = 0
plt.figure(figsize=(25, 35), dpi=100)
for types ,group in df_group_two:
    i += 1
    plt.subplot(3,2,i)
    group.groupby(by='水表名').sum()["用量"].plot.barh(title=types,color='blue',alpha=.5)
plt.show()


# In[50]:


pd.read_csv('data.csv')


# In[49]:


# 判断用电层次关系

总用水量随着季度月份天时间的变化结果及代码实现

不同区域用水量随着季度月份天和时间的变化

# coding: utf-8

# In[1]:


import pandas as pd
from matplotlib import pyplot as plt
data = pd.read_csv('data.csv')
data


# 
# data['timeStamp'] = pd.to_datetime(data['采集时间'])
# data.set_index("timeStamp", inplace=True)
# data_month = data.resample("M",closed="right",label='right'，loffset='-1')
# data_month
# for name,group in data_month:
#     print("*"*30)
#     print(name)
#     print(group)
#     print("*"*30)

# In[2]:


data['timeStamp'] = pd.to_datetime(data['采集时间']) 
data.set_index("timeStamp", inplace=True) 
data_quarter = data.resample("Q") 
data_quarter 
for name,group in data_quarter: 
    print("*"*30) 
    print(name) 
    print(group) 
    print(""*30)


# In[3]:


import matplotlib  as mpl
mpl.rcParams['font.sans-serif'] = ['SimHei']
data_quarter.sum()["用量"].plot.line(style="r*-.",alpha=0.5,title="用水量随着季度变化趋势")


# In[4]:


data_month = data.resample('M').sum()["用量"].plot.line(style="r*-.",alpha=0.5,title="用水量随着月份变化趋势")


# In[5]:


for datatime in data.index:
    print(datatime.month)
    datatime.hour
    break;


# In[6]:


data['hour'] = pd.to_datetime(data['采集时间']).dt.hour
data['day'] = pd.to_datetime(data['采集时间']).dt.day


# In[7]:


data.groupby(by='hour').sum()['用量'].plot.line(style="r*-.",alpha=0.5,title="用水量随着时间变化趋势")
plt.grid()


# In[8]:


data.groupby(by='day').sum()['用量'].plot.line(style="r*-.",alpha=0.5,title="用水量随着天变化趋势")
plt.grid()


# In[9]:


# 不同区域用水特征
# 不同区域随着季度用水变化
i = 0
plt.figure(figsize=(20, 20), dpi=100)
for index,group in data.groupby(by='类型'):
    i += 1
    plt.subplot(3,2,i)
    group.resample("Q").sum()['用量'].plot.line(style="r*-.",alpha=0.5,title="{}用水量随着季度变化趋势".format(index))
plt.show()


# In[10]:


# 不同区域随着月份用水变化
i = 0
plt.figure(figsize=(20, 20), dpi=100)
for index,group in data.groupby(by='类型'):
    i += 1
    plt.subplot(3,2,i)
    group.resample("M").sum()['用量'].plot.line(style="r*-.",alpha=0.5,title="{}用水量随着月份变化趋势".format(index))
plt.show()


# In[11]:


# 不同区域随着每月的天数用水变化
i = 0

plt.figure(figsize=(20, 20), dpi=100)
for index,group in data.groupby(by='类型'):
    i += 1
    plt.subplot(3,2,i)
    group.groupby(by='day').sum()['用量'].plot.line(style="r*-.",alpha=0.5,title="{}用水量随着天数变化趋势".format(index))
    plt.grid()
plt.show()


# In[12]:


# 不同区域随着每天的时辰用水变化
i = 0

plt.figure(figsize=(20, 20), dpi=100)
for index,group in data.groupby(by='类型'):
    i += 1
    plt.subplot(3,2,i)
    group.groupby(by='hour').sum()['用量'].plot.line(style="r*-.",alpha=0.5,title="{}用水量随着时辰变化趋势".format(index))
    plt.grid()
plt.show()


# In[15]:


# 判断周末和非周末用水量变化
weekday = []
for datatime in pd.to_datetime(data['采集时间']):
    if datatime.weekday()!=5 and datatime.weekday()!=6:
        weekday.append(0)
    else :
        weekday.append(1)
weekday


# In[16]:


len(weekday)


# In[17]:


data['weekday'] = pd.Series(weekday)
data


# In[18]:


data.info()


# In[19]:


data.head(5)


# In[20]:


pd.to_csv('data1.csv',index=False)


# In[21]:


data.to_csv('data1.csv',index=False)

分析周末与工作日不同区域用水变化结果及代码实现

# coding: utf-8

# In[1]:


import pandas as pd
import matplotlib.pyplot as plt
data1  = pd.read_csv('data1.csv')
data1.head(5)


# In[2]:


coll_time = pd.to_datetime(data1['采集时间'])


# In[3]:


weekday = []
for time in coll_time:
    if time.weekday() != 5 and time.weekday()!=6:
        weekday.append(0)
    else:
        weekday.append(1)
weekday


# In[4]:


data1['weekday'] = weekday


# In[5]:


data1.head(5)


# In[6]:


set(weekday)
data1['timeStamp'] = pd.to_datetime(data1['采集时间']) 
data1.set_index("timeStamp", inplace=True) 
data1


# In[7]:


# 根据不同weekday画图
from matplotlib import pyplot as plt
i = 0
plt.figure(figsize=(20, 20), dpi=100)
plt.rcParams['font.sans-serif'] = ['SimHei']
for index1, group1 in data1.groupby(by='类型'):
    i = i + 1
    plt.subplot(3,2,i)
    for index2, group2 in group1.groupby(by='weekday'):
        if index2==1:
            group2.resample('M').mean()["用量"].plot.line(style="*-.",alpha=0.8,title="{}用水量随着月份变化趋势".format(index1),label='周末')
        else:
            group2.resample('M').mean()["用量"].plot.line(style="*-.",alpha=0.8,title="{}用水量随着月份变化趋势".format(index1),label='工作日')
    plt.legend()
#     break
    
        


# In[8]:


i = 0
plt.figure(figsize=(20, 20), dpi=100)
for index1, group1 in data1.groupby(by='类型'):
    i = i + 1
    plt.subplot(3,2,i)
    for index2, group2 in group1.groupby(by='weekday'):
        if index2==1:
            group2.groupby(by='hour').mean()["用量"].plot.line(style="*-.",alpha=0.8,title="{}用水量随着时辰变化趋势".format(index1),label='周末')
        else:
            group2.groupby(by='hour').mean()["用量"].plot.line(style="*-.",alpha=0.8,title="{}用水量随着时辰变化趋势".format(index1),label='工作日')
    plt.grid()
    plt.legend()


# In[9]:


data1['time'] = pd.to_datetime(data1['采集时间']).dt.time
i = 0
plt.figure(figsize=(20, 20), dpi=100)
for index1, group1 in data1.groupby(by='类型'):
    i = i + 1
    plt.subplot(3,2,i)
    for index2, group2 in group1.groupby(by='weekday'):
        if index2==1:
            group2.groupby(by='time').mean()["用量"].plot.line(style="*-.",alpha=0.8,title="{}用水量随着时间变化趋势".format(index1),label='周末')
        else:
            group2.groupby(by='time').mean()["用量"].plot.line(style="*-.",alpha=0.8,title="{}用水量随着时间变化趋势".format(index1),label='工作日')
    plt.grid()
    plt.legend()


# In[13]:


data1.to_csv('data2.csv',index=False)