library(ggplot2) library(reshape2)
在上一次实验中,将”十一“节假日期间的交通流变化用柱状图显示,能够表现出”十一“节假日所带来的交通流突变效应。
本实验重点分析:
读取2016年10月1日交通流数据
tml1001 <- read.csv("D://data//thesis/201610/tmldata/tml1001all.csv",header = T) dim(tml1001)
数据有288行,似乎是全的。
names(tml1001)
首先整理一下数据,计算当量
tml1001$机动车当量 <- 1*tml1001$中小客车+1*tml1001$小货车+1.5*tml1001$大客车+1.5*tml1001$中货车+3*tml1001$大货车+ 4*tml1001$特大货车+4*tml1001$集装箱+1*tml1001$摩托车+4*tml1001$拖拉机
names(tml1001)
tml1001 <- tml1001[,-c(1,16)] names(tml1001)
绘制机动车交通流量图
plot(tml1001$时间序号,tml1001$机动车当量,type="b")
plot(tml1001$时间序号,tml1001$机动车当量,type="l")
qplot(x = 时间序号,y = 机动车当量,data=tml1001)+geom_line(color="steelblue")+ geom_smooth(method = "loess",span=0.1)+ scale_x_continuous(breaks = seq(0,288,24))+ scale_y_continuous(breaks = seq(0,120,20))
ggsave(filename = "10月1日交通流量点画线图.png",width = 6,height = 4,dpi = 600)
数据具有很多噪音
尝试使用LOESS技术平滑曲线,span取0.1。
loess1 <- loess(tml1001$机动车当量~tml1001$时间序号,span=0.1) plot(tml1001$时间序号,tml1001$机动车当量,type="l") lines(tml1001$时间序号,loess1$fit,col="red")
将LOESS平滑后的数据写入数据框,并存起来
tml1001$LOESS <- loess1$fitted write.csv(tml1001,file = "D:\\data\\thesis\\201610\\tmldata\\tml1001loess.csv")
tml0930 <- read.csv("D://data//thesis/201610/tmldata/tml0930all.csv",header = T) tml0930 <- tml0930[,-1] qplot(x = 时间序号,y = 机动车当量,data=tml0930)+geom_line(color="steelblue")+ geom_smooth(method = "loess",span=0.1)+ scale_x_continuous(breaks = seq(0,288,24))+ scale_y_continuous(breaks = seq(0,120,20))
tml1002 <- read.csv("D://data//thesis/201610/tmldata/tml1002all.csv",header = T) tml1002 <- tml1002[,-1] qplot(x = 时间序号,y = 机动车当量,data=tml1002)+geom_line(color="steelblue")+ geom_smooth(method = "loess",span=0.1)+ scale_y_continuous(breaks = seq(0,120,20))+ scale_x_continuous(breaks = seq(0,288,24))
tml1003 <- read.csv("D://data//thesis/201610/tmldata/tml1003all.csv",header = T) tml1003 <- tml1003[,-1] qplot(x = 时间序号,y = 机动车当量,data=tml1003)+geom_line(color="steelblue")+ geom_smooth(method = "loess",span=0.1)+ scale_y_continuous(breaks = seq(0,120,20))+ scale_x_continuous(breaks = seq(0,288,24))
tml1004 <- read.csv("D://data//thesis/201610/tmldata/tml1004all.csv",header = T) tml1004 <- tml1004[,-1] qplot(x = 时间序号,y = 机动车当量,data=tml1004)+geom_line(color="steelblue")+ geom_smooth(method = "loess",span=0.1)+ scale_y_continuous(breaks = seq(0,120,20))+ scale_x_continuous(breaks = seq(0,288,24))
tml1005 <- read.csv("D://data//thesis/201610/tmldata/tml1005all.csv",header = T) tml1005 <- tml1005[,-1] qplot(x = 时间序号,y = 机动车当量,data=tml1005)+geom_line(color="steelblue")+ geom_smooth(method = "loess",span=0.1)+ scale_y_continuous(breaks = seq(0,120,20))+ scale_x_continuous(breaks = seq(0,288,24))
tml1006 <- read.csv("D://data//thesis/201610/tmldata/tml1006all.csv",header = T) tml1006 <- tml1006[,-1] qplot(x = 时间序号,y = 机动车当量,data=tml1006)+geom_line(color="steelblue")+ geom_smooth(method = "loess",span=0.1)+ scale_y_continuous(breaks = seq(0,120,20))+ scale_x_continuous(breaks = seq(0,288,24))
tml1007 <- read.csv("D://data//thesis/201610/tmldata/tml1007all.csv",header = T) tml1007 <- tml1007[,-1] qplot(x = 时间序号,y = 机动车当量,data=tml1007)+geom_line(color="steelblue")+ geom_smooth(method = "loess",span=0.1)+ scale_y_continuous(breaks = seq(0,120,20))+ scale_x_continuous(breaks = seq(0,288,24))
tml0929 <- read.csv("D://data//thesis/201610/tmldata/tml0929all.csv",header = T) tml0929 <- tml0929[,-1] qplot(x = 时间序号,y = 机动车当量,data=tml0929)+geom_line(color="steelblue")+ geom_smooth(method = "loess",span=0.1)+ scale_x_continuous(breaks = seq(0,288,24))+ scale_y_continuous(breaks = seq(0,120,20))
tml0927 <- read.csv("D://data//thesis/201610/tmldata/tml0927all.csv",header = T) tml0927 <- tml0927[,-1] qplot(x = 时间序号,y = 机动车当量,data=tml0927)+geom_line(color="steelblue")+ geom_smooth(method = "loess",span=0.1)+ scale_y_continuous(breaks = seq(0,120,20))+ scale_x_continuous(breaks = seq(0,288,24))
tml0926 <- read.csv("D://data//thesis/201610/tmldata/tml0926all.csv",header = T) tml0926 <- tml0926[,-1] qplot(x = 时间序号,y = 机动车当量,data=tml0926)+geom_line(color="steelblue")+ geom_smooth(method = "loess",span=0.1)+ scale_y_continuous(breaks = seq(0,120,20))+ scale_x_continuous(breaks = seq(0,288,24))
tml0925 <- read.csv("D://data//thesis/201610/tmldata/tml0925all.csv",header = T) tml0925 <- tml0925[,-1] qplot(x = 时间序号,y = 机动车当量,data=tml0925)+geom_line(color="steelblue")+ geom_smooth(method = "loess",span=0.1)+ scale_y_continuous(breaks = seq(0,120,20))+ scale_x_continuous(breaks = seq(0,288,24))
tml0924 <- read.csv("D://data//thesis/201610/tmldata/tml0924all.csv",header = T) tml0924 <- tml0924[,-1] qplot(x = 时间序号,y = 机动车当量,data=tml0924)+geom_line(color="steelblue")+ geom_smooth(method = "loess",span=0.1)+ scale_y_continuous(breaks = seq(0,120,20))+ scale_x_continuous(breaks = seq(0,288,24))
tml0923 <- read.csv("D://data//thesis/201610/tmldata/tml0923all.csv",header = T) tml0923 <- tml0923[,-1] qplot(x = 时间序号,y = 机动车当量,data=tml0923)+geom_line(color="steelblue")+ geom_smooth(method = "loess",span=0.1)+ scale_y_continuous(breaks = seq(0,120,20))+ scale_x_continuous(breaks = seq(0,288,24))
tml0922 <- read.csv("D://data//thesis/201610/tmldata/tml0922all.csv",header = T) tml0922 <- tml0922[,-1] qplot(x = 时间序号,y = 机动车当量,data=tml0922)+geom_line(color="steelblue")+ geom_smooth(method = "loess",span=0.1)+ scale_y_continuous(breaks = seq(0,120,20))+ scale_x_continuous(breaks = seq(0,288,24))
tml0921 <- read.csv("D://data//thesis/201610/tmldata/tml0921all.csv",header = T) tml0921 <- tml0921[,-1] qplot(x = 时间序号,y = 机动车当量,data=tml0921)+geom_line(color="steelblue")+ geom_smooth(method = "loess",span=0.1)+ scale_y_continuous(breaks = seq(0,120,20))+ scale_x_continuous(breaks = seq(0,288,24))
tml0920 <- read.csv("D://data//thesis/201610/tmldata/tml0920all.csv",header = T) tml0920 <- tml0920[,-1] qplot(x = 时间序号,y = 机动车当量,data=tml0920)+geom_line(color="steelblue")+ geom_smooth(method = "loess",span=0.1)+ scale_y_continuous(breaks = seq(0,120,20))+ scale_x_continuous(breaks = seq(0,288,24))
tml0919 <- read.csv("D://data//thesis/201610/tmldata/tml0919all.csv",header = T) tml0919 <- tml0919[,-1] qplot(x = 时间序号,y = 机动车当量,data=tml0919)+geom_line(color="steelblue")+ geom_smooth(method = "loess",span=0.1)+ scale_y_continuous(breaks = seq(0,120,20))+ scale_x_continuous(breaks = seq(0,288,24))
names(tml1001)
tml1001try <- tml1001[,c(4,5,15)] names(tml1001try)
tmlmelt <- melt(tml1001try,id="时间序号") qplot(tmlmelt$时间序号,tmlmelt$value,data=tmlmelt,color=variable)+ facet_grid(variable~.,scales="free_y")
tmlall <- read.csv("D:\\data\\thesis\\201610\\tmldata\\tml.csv",header = T) dim(tmlall)
unique(tmlall$日期)
qplot(tmlall$时间序号,tmlall$中小客车,data=tmlall,color=tmlall$日期,geom="line")+ facet_grid(tmlall$日期~.,scales="free_y")
图太密了
ggsave("多日图对比2.png",width=20,height=10)
jjr <- unique(tmlall$日期)[12:18] tmljjr <- tmlall[tmlall$日期==jjr,] ggplot3 <- qplot(tmljjr$时间序号,tmljjr$中小客车,data=tmljjr,color=tmljjr$日期,geom="line")+ facet_grid(tmljjr$日期~.,scales="free_y")+geom_smooth(method = "loess",span=0.1)
ggsave("多日图对比.png",width=20,height=10)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.