Python实现双轴组合图表柱状图和折线图的具体流程

import cx_Oracle
import xlrd
import xlwt
import matplotlib.pyplot as plt
import numpy as np
from matplotlib.ticker import FuncFormatter

plt.rcParams['font.sans-serif']=['SimHei']
plt.rcParams['axes.unicode_minus']=False
#设置坐标轴数值以百分比(%)显示函数
def to_percent(temp, position):
  return '%1.0f'%(1*temp) + '%'
#字体设置
font2 = {'family' : 'Times New Roman',
'weight' : 'normal',
'size'   : 25,
}

conn=cx_Oracle.connect('用户名/密码@IP:端口/数据库')
c=conn.cursor()
#sql查询语句，多行用()括起来
sql_detail=("select substr(date1,6,10)date1,round(avg(r_qty))r_qty,round(avg(e_qty))e_qty,""round(avg(r_qty)/avg(e_qty),2)*100 userate,round(avg(uptime),2)*100 uptime from 表tp "
"tp where 条件  "
"group by date1 order by date1 ")  
                              
x=c.execute(sql_detail)
#获取sql查询数据                         
data=x.fetchall()
#print(data)

#新建Excel保存数据
xl=xlwt.Workbook()
ws=xl.add_sheet("ROBOT 30 DAYS MOVE ")
#ws.write_merge(0,1,0,4,"ROBOT_30_DAYS_MOVE")
for i,item in enumerate(data):
    for j,val in enumerate(item):
        ws.write(i,j,val)
xl.save("E:\\ROBOT_30_DAYS_MOVE.xls")

#读取Excel数据
data1 = xlrd.open_workbook( "E:\\ROBOT_30_DAYS_MOVE.xls")
sheet1=data1.sheet_by_index(0)

date1=sheet1.col_values(0)
r_qty=sheet1.col_values(1)
e_qty=sheet1.col_values(2)
userate=sheet1.col_values(3)
uptime=sheet1.col_values(4)

#空值处理
for a in r_qty:
    if a=='':
        a=0
for a in e_qty:
    if a=='':
        a=0
for a in userate:
    if a=='':
        a=0
for a in uptime:
    if a=='':
        a=0
#将list元素str转int类型
r_qty = list(map(int, r_qty))
e_qty = list(map(int, e_qty))
userate = list(map(int, userate))
uptime = list(map(int, uptime))
#添加平均值mean求平均
r_qty.append(int(np.mean(r_qty))) 
e_qty.append(int(np.mean(e_qty))) 
userate.append(int(np.mean(userate))) 
uptime.append(int(np.mean(uptime))) 
date1.append('AVG')

#x轴坐标
x=np.arange(len(date1))
bar_width=0.35

plt.figure(1,figsize=(19,10))
#绘制主坐标轴-柱状图
plt.bar(np.arange(len(date1)),r_qty,label='RBT_MOVE',align='center',alpha=0.8,color='Blue',width=bar_width)
plt.bar(np.arange(len(date1))+bar_width,e_qty,label='EQP_MOVE',align='center',alpha=0.8,color='orange',width=bar_width)

#设置主坐标轴参数
plt.xlabel('')
plt.ylabel('Move',fontsize=18)
plt.legend(loc=1, bbox_to_anchor=(0,0.97),borderaxespad = 0.) 
#plt.legend(loc='upper left')
for x,y in enumerate(r_qty):
    plt.text(x,y+100,'%s' % y,ha='center',va='bottom')
for x,y in enumerate(e_qty):
    plt.text(x+bar_width,y+100,'%s' % y,ha='left',va='top') 
plt.ylim([0,8000])

#调用plt.twinx()后可绘制次坐标轴
plt.twinx()

#次坐标轴参考线
target1=[90]*len(date1)
target2=[80]*len(date1)

x=list(range(len(date1)))
plt.xticks(x,date1,rotation=45)

#绘制次坐标轴-折线图
plt.plot(np.arange(len(date1)),userate,label='USE_RATE',color='green',linewidth=1,linestyle='solid',marker='o',markersize=3)
plt.plot(np.arange(len(date1)),uptime,label='UPTIME',color='red',linewidth=1,linestyle='--',marker='o',markersize=3)

plt.plot(np.arange(len(date1)),target1,label='90%target',color='black',linewidth=1,linestyle='dashdot')
plt.plot(np.arange(len(date1)),target2,label='80%target',color='black',linewidth=1,linestyle='dashdot')

#次坐标轴刻度百分比显示
plt.gca().yaxis.set_major_formatter(FuncFormatter(to_percent))

plt.xlabel('')
plt.ylabel('Rate',fontsize=18)
#图列
plt.legend(loc=2, bbox_to_anchor=(1.01,0.97),borderaxespad = 0.) 
plt.ylim([0,100])
for x,y in enumerate(userate):
    plt.text(x,y-1,'%s' % y,ha='right',va='bottom',fontsize=14)
for x,y in enumerate(uptime):
    plt.text(x,y+1,'%s' % y,ha='left',va='top',fontsize=14) 

plt.title("ROBOT 30 DAYS MOVE")

#图表Table显示plt.table()
listdata=[r_qty]+[e_qty]+[userate]+[uptime]#数据
table_row=['RBT_MOVE','EQP_MOVE','USE_RATE(%)','UPTIME(%)']#行标签
table_col=date1#列标签
print(listdata)
print(table_row)
print(table_col)

the_table=plt.table(cellText=listdata,cellLoc='center',rowLabels=table_row,colLabels=table_col,rowLoc='center',colLoc='center')
#Table参数设置-字体大小太小，自己设置
the_table.auto_set_font_size(False)
the_table.set_fontsize(12)
#Table参数设置-改变表内字体显示比例，没有会溢出到表格线外面
the_table.scale(1,3)
#plt.show()

plt.savefig(r"E:\\ROBOT_30_DAYS_MOVE.png",bbox_inches='tight')
#关闭SQL连接
c.close()                                                      
conn.close()