Meter reading 1
Project Title: Meter Readings from 3 Railway Stations
NAME: ASWATHY GUNADEEP
EMAIL: aswathygunadeep@gmail.com
COLLEGE / COMPANY: NATIONAL INSTITUTE OF TECHNOLOGY KARNATAKA
Based on the meter readings from the electricity board from a railway station, we intend to find out what affects the total ampere and power readings.
The data set includes information about:
- current and watt readings of the transformer in R(red), Y(yellow), and B(blue) phases. total active and apparent values.
- frequency and rpm of the windings of the electric grid.
- Average power factor and power factor of each phase.
- Harmonic distortions in each phase and interruptions.
setwd("C:/Users/user/Desktop/tarsha systems summer internship/datasets")
met1.df <- read.csv(paste("1.csv", sep=""))
View(met1.df)SUMMARY OF THE VARIABLES OF THE DATASET
str(met1.df)## 'data.frame': 10480 obs. of 40 variables:
## $ W.Total : num 5515 3891 4178 3911 5633 ...
## $ W.R : num 1607 1242 1341 1239 1689 ...
## $ W.Y : num 1941 1283 1388 1310 1991 ...
## $ W.B : num 1967 1366 1449 1362 1952 ...
## $ VAr.Total : num -4193 -4106 -4154 -3981 -3741 ...
## $ VAr.R : num -1253 -1287 -1311 -1270 -1174 ...
## $ Var.Y : num -1439 -1382 -1410 -1338 -1274 ...
## $ VAr.B : num -1500 -1437 -1433 -1373 -1293 ...
## $ P.F : num 0.796 0.688 0.709 0.701 0.833 ...
## $ P.F.R : num 0.789 0.694 0.715 0.698 0.821 ...
## $ P.F.Y : num 0.803 0.68 0.702 0.699 0.842 ...
## $ P.F.B : num 0.795 0.689 0.711 0.704 0.834 ...
## $ VA.Total : num 6928 5657 5892 5581 6762 ...
## $ VA.R : num 2038 1789 1875 1775 2057 ...
## $ VA.Y : num 2417 1885 1979 1872 2364 ...
## $ VA.B : num 2474 1983 2038 1934 2342 ...
## $ Volts.R : num 251 247 249 244 244 ...
## $ Volts.Y : num 247 246 247 242 240 ...
## $ Volts.B : num 252 250 251 245 244 ...
## $ Amps.Ave. : num 9.41 7.72 8 7.72 9.44 ...
## $ Amps.R : num 8.33 7.37 7.68 7.39 8.59 ...
## $ Amps.Y : num 9.9 7.86 8.11 7.8 9.89 ...
## $ Amps.B : num 10.01 7.93 8.2 7.98 9.82 ...
## $ Frequency : num 50.1 50 50.1 49.9 50 ...
## $ Wh.Rec. : num 26794 28127 29373 30585 32092 ...
## $ VAh.Rec. : num 32338 34018 35634 37218 39013 ...
## $ VArh.I.Rec. : num 0.0144 0.0144 0.0144 0.0144 0.0144 ...
## $ VArh.C.Rec. : num -17121 -18095 -19067 -20050 -20959 ...
## $ Neutral.Current : num 2.2 0 0 0 1.96 ...
## $ X.THD.Volt.R : num 3.22 3.21 2.98 2.57 2.27 ...
## $ X..THD.Volt.Y : num 2.36 2.25 2.52 2.08 1.81 ...
## $ X..THD.Volt.B : num 2.64 2.29 2.48 1.75 1.84 ...
## $ X.THD.Amps.R : num 20.9 19.6 17.9 14 11.8 ...
## $ X.THD.Amps.Y : num 16.5 16.3 14.9 21.1 11 ...
## $ X.THD.Amps.B : num 17.7 17.5 16.1 13.6 15.9 ...
## $ Rising.Demand : num 7255 5332 4985 4831 6036 ...
## $ Maximum.Demand : num 7269 7393 7393 7393 7393 ...
## $ RPM : num 1502 1499 1503 1498 1500 ...
## $ Load.Hours.Received: num 1.18e+09 1.24e+09 1.30e+09 1.35e+09 1.41e+09 1.47e+09 1.53e+09 1.59e+09 1.65e+09 1.71e+09 ...
## $ No.Of.Intrruptions : int 0 0 0 0 0 0 0 0 0 0 ...VISUALIZATION: CORRGRAM
library(corrgram)
par(mfrow=c(1,1))
sub.df <- subset(met1.df[,c(1,5,9,13,20,24,25,26,27,28,29,36,37,38,39,40)])
corrgram(sub.df, order=TRUE, lower.panel=panel.shade,
upper.panel=panel.pie, text.panel=panel.txt,
main="Corrgram of dataset")
Statistically, blue line shows that 2 continuous variables are positively correlated and negatively correlated if it is red. These are some of the possible conclusions that can be deduced from the above graph:
- From the corrgram, the variable average current strongly depends upon other variables such as apparent total power(VA Total) and rising demand.
- The total active power(W.Total) strongly depends upon apparent total power(VA Total).
- Frequency of the motor is greatly dependent on the RPM of the windings of the motor.
- Power Factor decreases Total reactive power.
- Neutral current and working load hours does not have much effect in the meter readings.
- Apparent energy received and active energy received are closely related, and both these variables are negatively correlated with Recative Capacitive Energgy Received and postively increases with Recative Inductive Energgy Received.
- Apparent energy received and active energy received also depends upon no. of interruptions.
CORRELATION MATRIIX
library(Hmisc)## Loading required package: lattice## Loading required package: survival## Loading required package: Formula## Loading required package: ggplot2##
## Attaching package: 'Hmisc'## The following objects are masked from 'package:base':
##
## format.pval, unitsdata <- cor(sub.df,use = "complete.obs")
round(data,2)## W.Total VAr.Total P.F VA.Total Amps.Ave. Frequency
## W.Total 1.00 0.28 0.19 0.96 0.97 0.01
## VAr.Total 0.28 1.00 -0.82 0.14 0.17 -0.03
## P.F 0.19 -0.82 1.00 0.26 0.25 0.01
## VA.Total 0.96 0.14 0.26 1.00 1.00 0.03
## Amps.Ave. 0.97 0.17 0.25 1.00 1.00 0.03
## Frequency 0.01 -0.03 0.01 0.03 0.03 1.00
## Wh.Rec. 0.03 0.32 -0.29 -0.01 0.00 0.01
## VAh.Rec. 0.03 0.32 -0.29 0.00 0.00 0.02
## VArh.I.Rec. 0.00 0.46 -0.47 -0.05 -0.04 0.01
## VArh.C.Rec. -0.04 -0.30 0.27 0.00 -0.01 -0.02
## Neutral.Current 0.32 0.04 0.02 0.34 0.33 0.02
## Rising.Demand 0.62 0.17 0.18 0.49 0.48 0.02
## Maximum.Demand 0.05 0.16 -0.12 0.03 0.04 0.16
## RPM 0.01 -0.03 0.01 0.03 0.03 1.00
## Load.Hours.Received -0.03 -0.02 0.02 -0.02 -0.02 0.01
## No.Of.Intrruptions 0.03 0.38 -0.35 -0.02 -0.01 0.02
## Wh.Rec. VAh.Rec. VArh.I.Rec. VArh.C.Rec.
## W.Total 0.03 0.03 0.00 -0.04
## VAr.Total 0.32 0.32 0.46 -0.30
## P.F -0.29 -0.29 -0.47 0.27
## VA.Total -0.01 0.00 -0.05 0.00
## Amps.Ave. 0.00 0.00 -0.04 -0.01
## Frequency 0.01 0.02 0.01 -0.02
## Wh.Rec. 1.00 1.00 0.60 -1.00
## VAh.Rec. 1.00 1.00 0.60 -1.00
## VArh.I.Rec. 0.60 0.60 1.00 -0.56
## VArh.C.Rec. -1.00 -1.00 -0.56 1.00
## Neutral.Current -0.10 -0.10 -0.02 0.11
## Rising.Demand 0.06 0.06 0.00 -0.06
## Maximum.Demand 0.85 0.85 0.34 -0.86
## RPM 0.01 0.02 0.01 -0.02
## Load.Hours.Received 0.00 0.00 0.00 0.00
## No.Of.Intrruptions 0.98 0.98 0.65 -0.98
## Neutral.Current Rising.Demand Maximum.Demand RPM
## W.Total 0.32 0.62 0.05 0.01
## VAr.Total 0.04 0.17 0.16 -0.03
## P.F 0.02 0.18 -0.12 0.01
## VA.Total 0.34 0.49 0.03 0.03
## Amps.Ave. 0.33 0.48 0.04 0.03
## Frequency 0.02 0.02 0.16 1.00
## Wh.Rec. -0.10 0.06 0.85 0.01
## VAh.Rec. -0.10 0.06 0.85 0.02
## VArh.I.Rec. -0.02 0.00 0.34 0.01
## VArh.C.Rec. 0.11 -0.06 -0.86 -0.02
## Neutral.Current 1.00 0.19 -0.09 0.02
## Rising.Demand 0.19 1.00 0.09 0.02
## Maximum.Demand -0.09 0.09 1.00 0.16
## RPM 0.02 0.02 0.16 1.00
## Load.Hours.Received 0.00 -0.01 0.00 0.01
## No.Of.Intrruptions -0.10 0.05 0.80 0.02
## Load.Hours.Received No.Of.Intrruptions
## W.Total -0.03 0.03
## VAr.Total -0.02 0.38
## P.F 0.02 -0.35
## VA.Total -0.02 -0.02
## Amps.Ave. -0.02 -0.01
## Frequency 0.01 0.02
## Wh.Rec. 0.00 0.98
## VAh.Rec. 0.00 0.98
## VArh.I.Rec. 0.00 0.65
## VArh.C.Rec. 0.00 -0.98
## Neutral.Current 0.00 -0.10
## Rising.Demand -0.01 0.05
## Maximum.Demand 0.00 0.80
## RPM 0.01 0.02
## Load.Hours.Received 1.00 0.00
## No.Of.Intrruptions 0.00 1.00- Relation between average current, VA Total and rising demand:
library(car)## Loading required package: carDatascatterplotMatrix(formula = ~ Amps.Ave. + VA.Total
+ Rising.Demand, cex=0.6, data=met1.df, diagonal="histogram", col="red")## Warning in applyDefaults(diagonal, defaults = list(method =
## "adaptiveDensity"), : unnamed diag arguments, will be ignored
PEARSON’S CORRELATION TEST:
par(mfrow=c(3,1))
cor.test(met1.df$Amps.Ave.,met1.df$VA.Total)##
## Pearson's product-moment correlation
##
## data: met1.df$Amps.Ave. and met1.df$VA.Total
## t = 1098.4, df = 10478, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.9955180 0.9958478
## sample estimates:
## cor
## 0.995686cor.test(met1.df$Amps.Ave.,met1.df$Rising.Demand)##
## Pearson's product-moment correlation
##
## data: met1.df$Amps.Ave. and met1.df$Rising.Demand
## t = 55.873, df = 10478, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4642231 0.4937277
## sample estimates:
## cor
## 0.4791107cor.test(met1.df$VA.Total,met1.df$Rising.Demand)##
## Pearson's product-moment correlation
##
## data: met1.df$VA.Total and met1.df$Rising.Demand
## t = 58.172, df = 10478, p-value < 2.2e-16
## alternative hypothesis: true correlation is not equal to 0
## 95 percent confidence interval:
## 0.4794728 0.5084195
## sample estimates:
## cor
## 0.4940831From the analysis, it is evident that apparent power is more if the average current produced is more.
- Factors driving total active power(W Total)
par(mfrow=c(1,1))
xyplot(W.Total ~ VA.Total, data=met1.df, main="total active power v/s total apparent power", type = c("p", "g"),xlab="apparent power", ylab="active power", col="green")
par(mfrow=c(1,2))
hist(met1.df$W.Total,
main="Total Active power",
col=c("red","blue","green","yellow"),
xlab="Total Active power", xlim=c(0,20000), ylim=c(0,3000))
hist(met1.df$VA.Total,
main="Total Apparent power",
col=c("red","blue","green","yellow"),
xlab="Total Apparent power",xlim=c(0,20000),ylim=c(0,4000))
par(mfrow=c(1,1))
sub1.df <- subset(met1.df[,c(1,13,36)])
data <- cor(sub1.df,use = "complete.obs")
round(data,2)## W.Total VA.Total Rising.Demand
## W.Total 1.00 0.96 0.62
## VA.Total 0.96 1.00 0.49
## Rising.Demand 0.62 0.49 1.00Total active power increases with total active power and the variable rising demand also slightly affects its value.
- Relation between frequecy and RPM
Dependent t-Test: NULL Hypothesis: There is no significent difference between the 2 variables.
par(mfrow=c(1,2))
log.transformed.fr = log(met1.df$Frequency)
boxplot(log.transformed.fr,col="turquoise")## Warning in bplt(at[i], wid = width[i], stats = z$stats[, i], out = z$out[z
## $group == : Outlier (-Inf) in boxplot 1 is not drawnlog.transformed.rpm = log(met1.df$RPM)
boxplot(log.transformed.rpm,col="orange")## Warning in bplt(at[i], wid = width[i], stats = z$stats[, i], out = z$out[z
## $group == : Outlier (-Inf) in boxplot 1 is not drawn
t.test(met1.df$Frequency,met1.df$RPM,paired=TRUE)##
## Paired t-test
##
## data: met1.df$Frequency and met1.df$RPM
## t = -10390, df = 10479, p-value < 2.2e-16
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
## -1449.586 -1449.039
## sample estimates:
## mean of the differences
## -1449.313Since the p-value is very less(<0.01), we reject our null hypothesis.The box plots revealed several outliers.
- Relation between active energy received and apparent energy received
par(mfrow=c(1,1))
xyplot(Wh.Rec. ~ VAh.Rec., data=met1.df, main="active energy received v/s apparent energy received", type = c("p", "g"),xlab="apparent energy received", ylab="active energy received", col="violet")
There is almost a linear increase.
- Power factor and total reactive power
par(mfrow=c(1,1))
xyplot(VAr.Total ~ P.F, data=met1.df, main="Power factor and total reactive power ", type = c("p", "g"),xlab="Power factor", ylab="total reactive power", col="darkolivegreen")
The graph is a skewed one and also reactive power decreases with power factor.
VARIATION OF EACH VARIABLE IN EACH PHASE
R PHASE
library(corrplot)## corrplot 0.84 loadedlibrary(gplots)##
## Attaching package: 'gplots'## The following object is masked from 'package:stats':
##
## lowesspar(mfrow=c(1,1))
sub2.df <- subset(met1.df[,c(2,6,10,14,17,21,30,33)])corrplot.mixed(corr=cor(sub2.df, use="complete.obs"),
upper="ellipse", tl.pos="lt", main="R Phase")
Y PHASE
par(mfrow=c(1,1))
sub3.df <- subset(met1.df[,c(3,7,11,15,18,22,31,34)])corrplot.mixed(corr=cor(sub3.df, use="complete.obs"),
upper="ellipse", tl.pos="lt", main="Y Phase")
B PHASE
par(mfrow=c(1,1))
sub4.df <- subset(met1.df[,c(4,8,12,16,19,23,32,35)])corrplot.mixed(corr=cor(sub4.df, use="complete.obs"),
upper="ellipse", tl.pos="lt", main="B Phase")
For all the 3 phases, the same trends are shown as we have discussed above.