输出预测结果

浙江电压等级电量/400v_11市_xgb.py

@@ -0,0 +1,43 @@
+import pandas as pd
+import xgboost as xgb
+from sklearn.model_selection import train_test_split
+from sklearn.metrics import r2_score
+import numpy as np
+
+df = pd.read_excel(r'./400v入模数据.xlsx')
+df['stat_date'] = pd.to_datetime(df['stat_date'])
+
+print(df.corr()['0.4kv及以下'])
+
+X = df[(df['stat_date']>='2021-01-01')&(df['stat_date']<='2023-09-28')].drop(columns=['0.4kv及以下']).set_index('stat_date')
+
+y = df[(df['stat_date']>='2021-01-01')&(df['stat_date']<='2023-09-28')]['0.4kv及以下']
+x_eval = df[(df['stat_date']<='2023-09-30')&(df['stat_date']>='2023-09-01')].drop(columns=['0.4kv及以下']).set_index('stat_date')
+print(x_eval)
+y_eval = df[(df['stat_date']<='2023-09-30')&(df['stat_date']>='2023-09-01')][['0.4kv及以下','city']]
+
+
+x_train,x_test,y_train,y_test = train_test_split(X,y,test_size=0.2,random_state=42)
+model = xgb.XGBRegressor(max_depth=6,learning_rate=0.05,n_estimators=250)
+model.fit(x_train,y_train)
+y_pred = model.predict(x_test)
+print(r2_score(y_test,y_pred))
+
+predict = model.predict(x_eval)
+result = pd.DataFrame({'real':y_eval.drop(columns='city').values.reshape(-1),'pred':predict},index=x_eval.index)
+print(result.loc['2023-09-28':'2023-09-30'])
+
+
+dict2 = {'杭州':0,'湖州':1,'嘉兴':2,'金华':3,'丽水':4,'宁波':5,'衢州':6,'绍兴':7,'台州':8,'温州':9,'舟山':10}
+dict1 = {}
+for city in x_eval['city'].drop_duplicates():
+    eval_x = x_eval[x_eval['city']==city]
+    eval_y = y_eval[y_eval['city']==city]['0.4kv及以下']
+    pred = model.predict(eval_x)
+    loss_rate = (np.sum(pred[-3:])-np.sum(eval_y[-3:]))/np.sum(eval_y)
+    dict1[city] = loss_rate
+
+
+for key in dict2.keys():
+    dict2[key] = dict1[dict2[key]]
+print(dict2)

浙江电压等级电量/400v_杭州.py

@@ -38,10 +38,10 @@ df = df.loc[normal(df['0.4kv及以下']).index]
 
 print(df.head())
 
-x_train = df.loc['2022-7':'2023-7'].drop(columns='0.4kv及以下')
-y_train = df.loc['2022-7':'2023-7']['0.4kv及以下']
-x_eval = df.loc['2023-8'].drop(columns='0.4kv及以下')
-y_eval = df.loc['2023-8']['0.4kv及以下']
+x_train = df.loc['2022-7':'2023-8'].drop(columns='0.4kv及以下')
+y_train = df.loc['2022-7':'2023-8']['0.4kv及以下']
+x_eval = df.loc['2023-9'].drop(columns='0.4kv及以下')
+y_eval = df.loc['2023-9']['0.4kv及以下']
 
 x_train,x_test,y_train,y_test = train_test_split(x_train,y_train,test_size=0.2,random_state=42)
 model = xgb.XGBRegressor(max_depth=6,learning_rate=0.05,n_estimators=150)

浙江电压等级电量/400v_衢州.py

@@ -0,0 +1,75 @@
+import pandas as pd
+import matplotlib.pyplot as plt
+import xgboost as xgb
+from sklearn.model_selection import train_test_split
+from sklearn.metrics import r2_score
+def normal(x):
+    high = x.describe()['75%'] + 1.5*(x.describe()['75%']-x.describe()['25%'])
+    low = x.describe()['25%'] - 1.5*(x.describe()['75%']-x.describe()['25%'])
+    return x[(x<=high)&(x>=low)]
+
+def season(x):
+    if str(x)[5:7] in ('04', '05', '06', '11'):
+        return 0
+    elif str(x)[5:7] in ('01', '02', '03', '09', '10', '12'):
+        return 1
+    else:
+        return 2
+
+df = pd.read_excel('./浙江各地市分电压日电量数据/衢州 .xlsx')
+df = df[['stat_date','0.4kv及以下']]
+df['0.4kv及以下'] = df['0.4kv及以下']/10000
+df['stat_date'] = df['stat_date'].map(lambda x:x.strip())
+df['stat_date'] = pd.to_datetime(df['stat_date'])
+
+
+df_qw = pd.read_excel(r'C:\python-project\p1031\入模数据\衢州.xlsx')
+df_qw.columns = df_qw.columns.map(lambda x:x.strip())
+
+df_qw = df_qw[['dtdate','tem_max','tem_min','holiday','24ST']]
+df_qw['dtdate'] = pd.to_datetime(df_qw['dtdate'])
+
+
+df = pd.merge(df,df_qw,left_on='stat_date',right_on='dtdate',how='left')
+df.drop(columns='dtdate',inplace=True)
+df.set_index('stat_date',inplace=True)
+
+
+# list2 = []
+# list0 = []
+# list1 = []
+# for i in ('01','02','03','04','05','06','07','08','09','10','11','12'):
+#     month_index = df.index.strftime('%Y-%m-%d').str[5:7] == f'{i}'
+#     # print(df.loc[month_index]['0.4kv及以下'].max(),df['0.4kv及以下'].describe()['75%'])
+#     if df.loc[month_index]['0.4kv及以下'].mean() >= df['0.4kv及以下'].describe()['75%']:
+#         list2.append(i)
+#     elif df.loc[month_index]['0.4kv及以下'].mean() <= df['0.4kv及以下'].describe()['25%']:
+#         list0.append(i)
+#     else:
+#         list1.append(i)
+# print(list0,list1,list2)
+
+
+df['season'] = df.index.map(season)
+df = df.loc[normal(df['0.4kv及以下']).index]
+
+x_train = df.loc['2021-7':'2023-9'][:-3].drop(columns='0.4kv及以下')
+
+y_train = df.loc['2021-7':'2023-9'][:-3]['0.4kv及以下']
+x_eval = df.loc['2023-9'].drop(columns='0.4kv及以下')
+y_eval = df.loc['2023-9']['0.4kv及以下']
+
+
+x_train,x_test,y_train,y_test = train_test_split(x_train,y_train,test_size=0.2,random_state=42)
+model = xgb.XGBRegressor(max_depth=6,learning_rate=0.05,n_estimators=150)
+model.fit(x_train,y_train)
+y_pred = model.predict(x_test)
+print(r2_score(y_test,y_pred))
+
+predict = model.predict(x_eval)
+result = pd.DataFrame({'eval':y_eval,'pred':predict},index=y_eval.index)
+print(result)
+print((result['eval'][-3:].sum()-result['pred'][-3:].sum())/result['eval'].sum())
+
+
+

浙江电压等级电量/400v数据预处理.py

@@ -16,6 +16,7 @@ for excel,qw_excel in zip(os.listdir(fir_dir),os.listdir(qw_dir)):
 
     df_city = df_city[['stat_date','0.4kv及以下']]
     df_city['0.4kv及以下'] = df_city['0.4kv及以下']/10000
+    df_city = df_city.loc[normal(df_city['0.4kv及以下']).index]
     df_city['stat_date'] = df_city['stat_date'].map(lambda x:x.strip())
     df_city['stat_date'] = pd.to_datetime(df_city['stat_date'])
 
@@ -53,13 +54,14 @@ for excel,qw_excel in zip(os.listdir(fir_dir),os.listdir(qw_dir)):
         else:
             return 2
 
-
+    print(f'{excel[:2]}',list0)
     df['season'] = df.index.map(season)
 
-    dict1 = {'杭州':0,'湖州':1,'嘉兴':2,'金华':3,'丽水':4,'宁波':5,'衢州':6,'绍兴':7,'台州':8,'温州':9,'舟山':10}
-    df['city'] = dict1[excel[:2]]
-    df.reset_index(inplace=True)
-    result = pd.concat(result,df)
+    df.to_excel(f'./400v入模数据/{excel[:2]}.xlsx')
+    # dict1 = {'杭州':0,'湖州':1,'嘉兴':2,'金华':3,'丽水':4,'宁波':5,'衢州':6,'绍兴':7,'台州':8,'温州':9,'舟山':10}
+    # df['city'] = dict1[excel[:2]]
+    # df.reset_index(inplace=True)
+    # result = pd.concat([result,df])
+
 
-print(df)
 

浙江电压等级电量/test1.py

@@ -1,5 +1,4 @@
 import os
-
 import numpy as np
 import pandas as pd
 n1 = np.array([[1,1,1]])
@@ -9,17 +8,17 @@ n2 = np.array([]).reshape(3,-1)
 
 print(np.max([[1,2,3],[4,5,6]]))
 
-file_dir = r'C:\Users\user\Desktop\浙江各地市分电压日电量数据'
-df = pd.read_excel(r'C:\Users\user\Desktop\浙江省各地市日电量及分压数据21-23年.xlsx',sheet_name=1)
-df.columns = df.columns.map(lambda x:x.strip())
-for city in df['地市'].drop_duplicates():
-    df_city = df[df['地市']== city]
-    df_city['stat_date'] = df_city['stat_date'].map(lambda x:x.strip())
-    df_city['stat_date'] = pd.to_datetime(df_city['stat_date'],format='%Y-%m-%d')
-    df_city = df_city[df_city.columns[:-1]]
-    df_city.sort_values(by='stat_date',ascending=True,inplace=True)
-    df_city['stat_date'] = df_city['stat_date'].astype('str')
-    df_city.to_excel(fr'C:\Users\user\Desktop\浙江各地市分电压日电量数据\{city}.xlsx',index=False)
+# file_dir = r'C:\Users\user\Desktop\浙江各地市分电压日电量数据'
+# df = pd.read_excel(r'C:\Users\user\Desktop\浙江省各地市日电量及分压数据21-23年.xlsx',sheet_name=1)
+# df.columns = df.columns.map(lambda x:x.strip())
+# for city in df['地市'].drop_duplicates():
+#     df_city = df[df['地市']== city]
+#     df_city['stat_date'] = df_city['stat_date'].map(lambda x:x.strip())
+#     df_city['stat_date'] = pd.to_datetime(df_city['stat_date'],format='%Y-%m-%d')
+#     df_city = df_city[df_city.columns[:-1]]
+#     df_city.sort_values(by='stat_date',ascending=True,inplace=True)
+#     df_city['stat_date'] = df_city['stat_date'].astype('str')
+#     df_city.to_excel(fr'C:\Users\user\Desktop\浙江各地市分电压日电量数据\{city}.xlsx',index=False)
 # file_Dir = r'C:\Users\鸽子\Desktop\浙江各地市行业电量数据'
 # for excel in os.listdir(file_Dir):
 #     df1 = pd.read_excel(r'C:\Users\鸽子\Desktop\浙江各地市日电量数据-27-28).xlsx',sheet_name=1)
@@ -34,3 +33,8 @@ for city in df['地市'].drop_duplicates():
 #     df2 = pd.concat((df2,df1),ignore_index=True)
 #     df2.to_excel(fr'C:\Users\鸽子\Desktop\浙江各地市行业电量数据\{city}.xlsx')
 
+df = pd.read_csv(r'C:\Users\鸽子\Desktop\浙江各区县数据(2).csv')
+df.columns = df.columns.map(lambda x:x.strip())
+print(df.columns)
+
+print(dict(zip(df.columns,[(df[x]==0).sum()/len(df) for x in df.columns])))