1 About the report

1.1 Introduction

This report brings together all the basic energy statistics across the 5 POC sites in an easy to read template. The charts and tables are organised in bulletted sections and subsections. We have used statistical techniques to handle missing data (which happens from time to time due to site wiring or equipment damages, uninformed changes and router hang situations). A short summary is added by us on each section as our take. The report is in HTML format and can be opened from any modern browser.

All the data used in this report is for the period Dec 1st to Mar 31st for a total of 5 sites. Except the DC Load section, which uses data for last 30 days from the date of this report. All data (fuel levels, events of DG, AC and DC electrical parameters) was automatically collected using IOT sensors and energy meters at a 1 minute frequency.

2 AC Mains (Grid power)

2.1 Monthly units

To start with let’s take a look at total AC kWh clocked for each site for each month.

Monthly GRID Energy used (kWh)
Dec'21 Jan'22 Feb'22 Mar'22
BagruUPG1 6,913 6,724 5,880 7,150
BagruUPG2 4,262 4,097 4,018 5,188
Dand01 5,188 4,943 4,911 5,699
Jobner 6,954 6,813 6,446 7,232
Sawarda 3,273 3,191 2,927 3,333
TOTAL 26,589 25,768 24,182 28,602

2.2 Computation of monthly units

Monthly GRID units Computation table1
Month Days
in Month		 Mean Daily
kWh Units		 Missed
Hours		 Percent
Missed Data		 Projected
Month(kWH)
BagruUPG1 Dec'21 31 223 72 9.7% 6,913
BagruUPG1 Jan'22 31 217 0 0.0% 6,724
BagruUPG1 Feb'22 28 210 0 0.0% 5,880
BagruUPG1 Mar'22 31 231 0 0.0% 7,150
BagruUPG2 Dec'21 31 137 0 0.0% 4,262
BagruUPG2 Jan'22 31 132 0 0.0% 4,097
BagruUPG2 Feb'22 28 144 0 0.0% 4,018
BagruUPG2 Mar'22 31 167 48 6.5% 5,188
Dand01 Dec'21 31 167 70 9.5% 5,188
Dand01 Jan'22 31 159 216 29.0% 4,943
Dand01 Feb'22 28 175 216 32.1% 4,911
Dand01 Mar'22 31 184 230 30.9% 5,699
Jobner Dec'21 31 224 51 6.9% 6,954
Jobner Jan'22 31 220 13 1.7% 6,813
Jobner Feb'22 28 230 30 4.4% 6,446
Jobner Mar'22 31 233 0 0.0% 7,232
Sawarda Dec'21 31 106 52 7.0% 3,273
Sawarda Jan'22 31 103 29 3.9% 3,191
Sawarda Feb'22 28 105 0 0.0% 2,927
Sawarda Mar'22 31 108 12 1.6% 3,333
1 The total projected kWh may vary to a small extent when compared with other reports, due to varying methods of handling missing data.

2.3 Normalised Grid Energy

Due to changing period each month (number of days in month and missing periods of data), the captured grid kWh may not give a representative trend month on month. Therefore we use normalized data of exactly 30 days (720 hours) by apportioning existing kWh for each month.

2.4 Trend & Variance

The trend & variance of kWh (per day per site) can be visualized much better in compact boxplot charts. The below box plot chart shows that data consistency is good for all sites.

Note on boxplot defaults. The taller the box the greater is the volatility of the data points (kWh each day). The median line is the solid horizontal line inside each box. The lower and upper hinges correspond to the first and third quartiles (the 25th and 75th percentile thresholds). The upper whisker extends from the hinge to the largest value no further than 1.5 times IQR from the hinge (where IQR is the inter-quartile range, or distance between the first and third quartiles). The lower whisker extends from the hinge to the smallest value at most 1.5 times of IQR of the hinge. Data beyond the end of the whiskers are called “outlying” points and are plotted individually.

2.5 Input Energy Components

Diesel Generator is the grid backup. Hence its consumption is tracked using AC source number two in our AC energy meters. Except Dand where we do not have two feeds for AC sources, on all other sites we are able to capture the energy supplied by DG set directly in the energy meter.

2.5.1 Monthly

Following is the kWh generated month wise by DG sets on each site.

DG Efficiency & EB Energy (kWh)
Month Fuel Consumption
(Litres)		 EER dgKWh ebKWh Total Input Energy
(kWh)
BagruUPG1 Dec'21 374 2,669 999 6,244 7,243
BagruUPG1 Jan'22 394 2,506 986 6,724 7,711
BagruUPG1 Feb'22 393 2,448 963 5,880 6,843
BagruUPG1 Mar'22 398 2,682 1,067 7,150 8,217
BagruUPG2 Dec'21 406 1,855 752 4,262 5,014
BagruUPG2 Jan'22 404 1,796 725 4,097 4,822
BagruUPG2 Feb'22 380 1,912 726 4,018 4,744
BagruUPG2 Mar'22 446 2,106 939 4,854 5,793
Dand01 Dec'21 261 2,660 695 4,853 5,548
Dand01 Jan'22 247 2,356 583 3,570 4,153
Dand01 Feb'22 113 2,122 239 4,273 4,513
Dand01 Mar'22 200 1,874 375 4,045 4,419
Jobner Dec'21 NA NA 647 6,506 7,153
Jobner Jan'22 299 2,707 809 6,813 7,621
Jobner Feb'22 213 2,627 559 6,216 6,775
Jobner Mar'22 206 2,686 553 7,232 7,785
Sawarda Dec'21 330 2,019 667 3,167 3,835
Sawarda Jan'22 393 2,070 814 3,088 3,902
Sawarda Feb'22 362 2,184 791 2,927 3,717
Sawarda Mar'22 386 2,246 866 3,333 4,199

2.5.2 For full Period

This is the split between Grid and DG energy aggregated over the entire 4 months time span.

Consolidated for 4 months
Months Dec'21 to Mar'22
Grid
kWh		 DG
kWh		 Fuel
Lits		 Percent
DG		 Average
EER
BagruUPG1 25,998 4,016 1,559 13% 2,575
BagruUPG2 17,231 3,142 1,635 15% 1,922
Dand01 16,741 1,892 821 10% 2,303
Jobner 26,766 2,568 NA 9% NA
Sawarda 12,514 3,138 1,472 20% 2,133

3 Run hours

These are tables with Total Run hours (or use hours) of various energy sources, on a per site per month basis. Since the total hours in a month are 720 or 744 (for 31 days month) and there could be few hours of downtime every month, the monthly totals may not add up to exact available hours of the month. Sometimes the total may exceed the total available hours by a small amount. This is due to the reason of overlapping of two energy sources e.g. Battery & DG or DG & Mains.

3.1 Bagru1

BagruUPG1
Run hours
Month Tot. Hrs
in Mnth		 HOURS LOGGED
EB DG BAT
Dec'21 744 641 95 6
Jan'22 744 612 102 4
Feb'22 672 568 99 1
Mar'22 744 640 102 0

3.2 Bagru2

BagruUPG2
Run hours
Month Tot. Hrs
in Mnth		 HOURS LOGGED
EB DG BAT
Dec'21 744 639 113 0
Jan'22 744 614 113 0
Feb'22 672 576 102 1
Mar'22 744 638 115 0

3.3 Dand

Dand01
Run hours
Month Tot. Hrs
in Mnth		 HOURS LOGGED
EB DG BAT
Dec'21 744 616 71 56
Jan'22 744 629 65 50
Feb'22 672 620 32 20
Mar'22 744 650 51 43

3.4 Jobner

Jobner
Run hours
Month Tot. Hrs
in Mnth		 HOURS LOGGED
EB DG BAT
Dec'21 744 680 62 5
Jan'22 744 664 75 3
Feb'22 672 608 50 12
Mar'22 744 690 49 3

3.5 Sawarda

Sawarda
Run hours
Month Tot. Hrs
in Mnth		 HOURS LOGGED
EB DG BAT
Dec'21 744 589 116 36
Jan'22 744 572 138 30
Feb'22 672 486 126 31
Mar'22 744 565 134 42

3.6 Charts - trends

Run hour trends can be seen in this plot for all 3 energy sources. As you see maximum Grid energy is used in Jobner, max DG energy in Sawarda and max Battery is deployed in Dand. Battery is hardly used in Bagru and Jobner sites.

4 Diesel Generator Metrics

These reports give a quick idea of the DG sets performance for each site. The number of sites is small hence classification of DG sets is not yet possible. In the series of DG metrics you can see total fuel consumption, total refuel (net of drain), total run hours, idling hours, CPH & EER. CPH has been calculated by two methods.

  • Ignition hours (Gross) : Time when DG was running (irrespective of load connected or not)
  • Loading hours (Net) : Time when DG was running and load connected to it.

The Second method will be closer to the actual CPH. EER, or Energy Exchange Ratio, as defined by us is Watt Hours per Litre and this is a more accurate measure of a DG’s performance.

4.1 Fuel & Run hours

Month Cons.
Actual(L)		 Drain
(L)		 Refuel
Gross(L)		 Refuel
Net(L)		 DG Run hrs
Ignition		 DG Run hrs
Connected		 DG Idle hrs
Detected1
BagruUPG1
Dec'21 374.3 0.0 352.4 352.4 94.6 86.3 5.7
Jan'22 393.6 0.0 405.3 405.3 101.7 96.9 4.8
Feb'22 393.4 85.2 452.1 366.9 99.5 93.2 6.2
Mar'22 398.0 0.0 402.3 402.3 102.0 95.1 6.9
BagruUPG2
Dec'21 405.5 0.0 358.1 358.1 113.0 97.1 15.9
Jan'22 403.6 0.0 401.3 401.3 113.1 100.1 12.9
Feb'22 379.6 19.6 419.8 400.2 101.5 87.3 14.2
Mar'22 446.0 0.0 440.5 440.5 114.9 96.1 17.0
Dand01
Dec'21 261.3 0.0 395.8 395.8 71.3 - -
Jan'22 247.4 0.0 167.4 167.4 64.9 - -
Feb'22 112.7 0.0 202.6 202.6 31.7 - -
Mar'22 200.0 0.0 82.1 82.1 50.7 - -
Jobner
Dec'21 - - - - - 49.8 10.2
Jan'22 298.7 0.0 240.5 240.5 75.0 66.2 6.8
Feb'22 212.8 68.2 278.9 210.7 49.5 45.5 3.6
Mar'22 205.9 0.0 195.8 195.8 49.0 43.6 5.4
Sawarda
Dec'21 330.5 0.0 341.2 341.2 115.8 105.4 1.4
Jan'22 393.2 0.0 380.9 380.9 137.6 136.1 1.2
Feb'22 362.0 0.0 351.2 351.2 126.5 125.7 0.8
Mar'22 385.8 0.0 381.3 381.3 133.6 131.9 1.3
1 DG idling is taken when Watt Hours supplied is zero while DG ignition is ON. Excludes missing watt hour periods

4.2 EER

EER
Month BagruUPG1 BagruUPG2 Dand01 Jobner Sawarda
Dec'21 2,683 1,855 2,639 2,883 2,031
Jan'22 2,505 1,796 2,546 2,707 2,071
Feb'22 2,448 1,912 2,279 2,629 2,181
Mar'22 2,682 2,096 2,445 2,686 2,246

4.3 CPH - Net & Gross

CPH
Month NET CPH1 GROSS CPH
Bagru1 Bagru2 Dand Jobner Sawarda Bagru1 Bagru2 Dand Jobner Sawarda
Dec'21 4.64 5.00 - 7.80 3.18 4.33 4.18 3.67 5.14 3.14
Jan'22 4.47 4.63 - 5.02 2.91 4.23 4.03 3.81 4.51 2.89
Feb'22 4.52 5.19 - 5.07 2.90 4.22 4.35 3.55 4.67 2.88
Mar'22 4.51 5.64 - 5.39 2.95 4.19 4.64 3.95 4.72 2.92
1 Net CPH = Fuel consumed per hour of DG running and connected to load

4.4 Run hours - Variance

Box plots reveal a lot more information about the distribution of DG run hours. As we see from the 5 box plot trends, Dand & Jobner have the lowest medians. Dand has a very close lower hinge in the box which means most data is close to the lowest data. We also see many outliers of high values in Dand. Since Dand has such a low median and 25 percentile line, it means the DG is getting fired several times for very short duration, which could be inefficient not just for DG battery life but for fuel economy also. Sawarda DG run hours are the most predictable with few outliers.

4.5 EER variations

As we see the EER-variations are almost the same across sites (all box plots have same height). This indicates similar data accuracy across sites. We also see that Bagru2 has very low EER compared to any of the 30 KVA sites indicating a problem in the DG set. Sawarda DG is 25KVA hence is expected to be lower than the other 4.

4.6 Top 10 Idling

TOP 10 DG IDLING CASES
Site Date Ignition Time Connected Time Idling Time Fuel Consumed
(Litres)		 Fuel CPH
jobner001 2021-12-05 17838s (~4.96 hours) 2520s (~42 minutes) 15318s (~4.26 hours) 10.8 2.2
jobner001 2021-12-16 13122s (~3.64 hours) 6779.99999999999s (~1.88 hours) 6342.00000000001s (~1.76 hours) 10.4 2.9
jobner001 2021-12-06 7722s (~2.14 hours) 2580s (~43 minutes) 5141.99999999999s (~1.43 hours) 5.5 2.6
bagru2 2022-01-14 15236s (~4.23 hours) 10260s (~2.85 hours) 4975.99999999999s (~1.38 hours) 13.1 3.1
bagru2 2021-12-26 18321s (~5.09 hours) 13440s (~3.73 hours) 4880.99999999999s (~1.36 hours) 17.2 3.4
bagru2 2022-03-14 17382s (~4.83 hours) 13140s (~3.65 hours) 4241.99999999999s (~1.18 hours) 17.4 3.6
bagru2 2022-03-27 32800s (~9.11 hours) 29220s (~8.12 hours) 3580s (~59.67 minutes) 34.1 3.7
bagru2 2022-01-08 14489s (~4.02 hours) 10920s (~3.03 hours) 3569s (~59.48 minutes) 15.1 3.8
bagru2 2022-03-12 16246s (~4.51 hours) 12900s (~3.58 hours) 3346s (~55.77 minutes) 16.3 3.6
bagru2 2022-03-25 11066s (~3.07 hours) 7740s (~2.15 hours) 3326s (~55.43 minutes) 9.9 3.2

4.7 Balance Tank

4.7.1 Balance Tank at Refuelling time

Sometimes a culture of late refueling is seen amongst contractors, perhaps to conserve their cash flow, or just plain laziness. This may result in tanks going down to dangerously low levels risking a site outage in case of a long power outage. It may be a good idea to see the tank balance over a large number of refuelings. As of now we have a total of 170 refuelling observations across 5` sites. As we see

We see that there are a total of 2 sites that have done refuellings at a dangerously low less than 5% tank capacity. These are Jobner, Sawarda

Refuel Instances with Dangerously low levels
For a period Apr 20, 2021 to Apr 12, 2022
Site Total
Refuel count		 Below 5%
Tank Lev
Bagru1 31 -
Bagru2 32 -
Dand 38 -
Jobner 28 1
Sawarda 28 12

5 Daily kWh trend

The underlying data used for all the above reports can be seen in the following daily energy plots. The light blue solid line is the statistically interpolated EB kWh data, red line is the DG kWh and the dark blue smoothed line at the top is the total AC kWh (sum of EB and DG kWh).

Note: the light grey dashed lines inside EB and DG line charts are from the raw data if no interpolation was performed.

5.1 Bagru1

5.2 Bagru2

5.3 Dand

5.4 Jobner

5.5 Sawarda

Note: Underlying raw data in csv format can be supplied for all the above charts.

6 DC Load

DC load contributes the biggest energy consumption for any BTS site. The DC load pattern for the full day is different for each site depending on the radio equipment used as well as the mix of load e.g. for Transmission / Back Haul Optical equipment the DC load will be mostly stationary throughout the day.

6.1 DC load as component of AC load

6.2 Variance - Within a Day

All 5 sites have a unique DC load distribution over the day. The higher the Inter Quartile Range (IQR) the better the off hour load reduction.

7 Missing Data

7.1 Energy Meter

Missing data is an important metric for the overall quality of the IOT project. We should strive to get this value to zero for all sites. The following plot shows missed or incorrect data measured in downtime hours per month for each site. The Heavy Outage in Dand seen in Jan and Feb was due to the uninformed changes done by Airtel SPS/AC source wiring/configuration resulting in reduced AC wattage measured for the period. Omnicomm technician had to do a rewiring at site. This caused a dip in total AC wattage measure for almost 20 days starting Jan 20th.

7.2 Fuel level

When we examined missing data for fuel, the top fuel sensor outage occured for the site : Jobner in the month Dec’21 for a total period of 312 hours. There was a net fuel level difference across this outage period starting on Dec 18 of 158.1 litres.