Test Multiple Starter Cycles with Ni-Cad Batteries Following IEEE 1115-2014 Recommendation

Python test procedure of multiple discharge starter cycles on Nickel-Cadmium batteries with the IEEE 1115-2014 algorithm. IEEE 1115-2014 is Recommended Practice for Sizing Nickel-Cadmium Batteries for Stationary Applications

Aim of this algorithm is to take an elementary battery block profile, specifically its capacity, its Kt capacity rating factor (defined as battery capacity divided by current for a given duration), temperature derating factor, aging factor, maximum depth of discharge, etc (see details in Usage section) and find how many battery blocks should be installed to stand a given number of starting cycles

Starting cycles are defined as periods. Discharge current is considered constant during each period. Calculations are done to assess the required capacity for different so called sections. Section n is defined as sequence of period from period 1 to period n

For more details about this algorithm see documentation folder where a couple of pdf files details the process and the rationale. One document refers to IEEE 485 which is a very similar algorithm targeting lead acid batteries instead of Ni-Cad batteries

What it does

Considering a battery block with a given capacity and a certain number of starter cycles discharging batteries, the evaluation calculates, given a number of sections, i.e. steps of starter cycles, how many battery blocks should be needed

This program displays calculation results and optionally intermediate calculations. Those intermediate calculation details are comma separated which eases load those results as a CSV file

Usage

python process.py def.json

with def.json being a json file with the following example structure:

 {
  "title": "Test...\n details...",
  "nominalCapacity": 100,
  "numberOfSections": 12,
  "verbose": true,
  "deratingFactorOnTemp": 1.0,
  "randomSize": 0.0,
  "designMargin": 1.2,
  "agingFactor": 1.0,
  "finalTolerance": 0.99,
  "csvFileNames":{
   "startingCycles": "starting.csv",
   "ampsByDurationFileName": "amps-per-duration.csv"
   }
 }

Where:

• title is a header, possibly multiline (with \n line breaks) displayed at the beginning of the result report
• nominalCapacity is the elementary battery black capacity in Ah
• numberOfSections is the number of starter sections to consider to compute the evaluation
• verbose if true will produce a report with full calculation details, if false only result will be displayed
• deratingFactorOnTemp, randomSize, designMargin, agingFactor are values used by the IEEE 1115-2014 algorithm
• finalTolerance will relax the final size to estimate the number of batteries required (meaning if we are close to a required battery number threshold, threshold will not be crossed). This parameter which range should reasonably be some percent (e.g. between 0.99 and 0.95 for 1 to 5% tolerance) is not defined in the standard, so if no such tolerance is suited, set it at 1.0
• csvFileNames is a place to define file names for Starting Profile and battery block Current per Duration. See next section for details

Starting Profile File

This is a CSV (Comma Separated Value) file which defines a starter current profile, composed of cycles. It is composed of three columns with a compulsory header line: Duration*, Current*, Cycle. Example:

Duration,Current,Cycle
00:15:00,5,1
00:10:00,35,1
01:15:00,15,1
00:10:00,60,1
00:10:00,5,1
00:15:00,5,2
00:10:00,35,2
01:15:00,15,2
00:10:00,60,2
00:10:00,5,2

In this example, 2 cycles are defined with the same profile of 5 Amps, 35 Amps, 15 Amps, 60 Amps, 5 Amps discharge current for respectively 15, 10, 75, 10, and 10 minutes (given as hh:mm:ss, hh and mm being integers and ss a float)

Current per Duration File

This is a CSV (Comma Separated Value) file which defines battery current for different durations. It is composed of two columns with a compulsory header line: Duration, Current. Example:

Duration,Current
00:01:00,189
00:15:00,133
00:30:00,94
01:00:00,58
02:00:00,37
03:00:00,27
04:00:00,22
05:00:00,18
06:00:00,16
07:00:00,14
08:00:00,13
10:00:00,11
12:00:00,9.3

In this example, elementary battery block current is defined versus duration, given as hh:mm:ss, hh and mm being integers and ss a float

Result Sample

Test with a 104Ah battery defined in "sample/test-with-104Ah-battery.json" IEEE 1125-2014 simulation started on 2020/11/09 Starting data file: "./sample/starting-data.csv" Battery current data file: "./sample/amps-per-duration-to-1.05V.csv"

Section 1 skipped

Section 2

Period	Load	Change	Duration	Remaining	Kt	Temp Derating	Size
1	5.00	5.00	00:15:00.0	00:25:00.0	0.9912	1.19	5.90
2	35.00	30.00	00:10:00.0	00:10:00.0	0.6907	1.19	24.66
Total: 30.56 Ah

Section 3 skipped

Section 4

Period	Load	Change	Duration	Remaining	Kt	Temp Derating	Size
1	5.00	5.00	00:15:00.0	01:50:00.0	2.6725	1.19	15.90
2	35.00	30.00	00:10:00.0	01:35:00.0	2.4542	1.19	87.62
3	15.00	-20.00	01:15:00.0	01:25:00.0	2.2870	1.19	-54.43
4	60.00	45.00	00:10:00.0	00:10:00.0	0.6907	1.19	36.99
Total: 86.07 Ah

Result testing 1 cycle Maximum section size: 86.07 + Random size: 0.00 = Uncorrected size: 86.07 Ah Uncorrected size: 86.07 x Design margin: 1.15 x Aging factor: 1.25 = Size: 123.73 Ah Number of batteries required: 2 (with a 1.0% tolerance)