6.5. DC source DGN additions
Fri, 2024-04-05 13:56 — RV-C Admin
Justification:
The following additions are proposed to the DC Source status and configuration DGNs
| Attachment | Size |
|---|---|
| 6.5. DC Source DGN Additions.docx | 24.4 KB |
Would the following edits
Would the following edits sufficiently clarify this to resolve the concern?
Bits 3.0-3.3 Current Smoothing Algorithm
0 = no smoothing
1 = exponential weighted moving average:
st = (1/N)xt + (1 – 1/N)st-1
where st is the new average
and st-1 is the previous average
and xt is the new data
and N is determined by smoothing period
2 = boxcar (unweighted) moving average
st =1/N Σxi for i = (t-N+1) to t
15 = no change/not supported
Bits 3.4-3.7 Voltage Smoothing Algorithm . . . .
Byte 4 Current Smoothing Period
This parameter can be used to change the smoothing period used for the reported Current measurements in DC Source Status 1.
Resolution is 0.1s, with a range of 0.1 to 25.0 seconds.
For both smoothing methods, N = tsmooth/tsamp
Where tsmooth is this parameter and tsamp is the internal Current sample rate, both in seconds.
Note: If unused, assumed to match the DC Source Status 1 Normal Broadcast Gap.
Byte 5 Voltage Smoothing Period
This parameter can be used to change the smoothing period used for the reported Voltage measurements in DC Source Status 1.
Resolution is 0.1s, with a range of 0.1 to 25.0 seconds.
For both smoothing methods:
N = tsmooth/tsamp
Where tsmooth is this parameter and tsamp is the internal Voltage sample rate, both in seconds.
Note: If unused, assumed to match the DC Source Status 1 Normal Broadcast Gap.
OBJECTION On
OBJECTION
On DC_SOURCE_CONFIGURATION_COMMAND_4
Current Time Constant / Voltage Time Constant.
There are several algorithms that can be used to smooth and sample this sort of data. It isn't clear which is intended, and for no algorithm is there enough information here to fully describe the parameters. From the Name and Description, I would guess that Exponential Smoothing is what is intended. From the Unit, I would assume a Moving Average. In the first case, there are two parameters of interest - the sampling rate and the smoothing factor. In the latter case, only the sampling rate is required.
These are just two possibilities - there are lots of other algorithms available. Intuitively, a Moving Average would be more appropriate for Current, Exponential Smoothing for Voltage, but there are a number of factors to consider.
I suggest something like:
Bits 3.0-3.3 Current Smoothing Algorithm 0 = Single Sample. 1 = Moving Average. 2 = Exponential Smoothing.
Bits 3.4-3.7 Voltage Smoothing Algorithm . . . .
Byte 4 Current Sample Rate (0.1 sec)
Byte 5 Current Smoothing Constant
Byte 6 Voltage Sample Rate (0.1 sec)
Byte 7 Voltage Smoothing Constant