In principle, ‘Off-Target’ as a measure is a gauge of how far away from expectations each unit’s output is for each discrete Dispatch Interval.
Readers should understand that (broadly speaking) the greater the Off-Target number, the more difficult it is for the AEMO to balance supply and demand. Hence this is a metric that is worth monitoring in some detail:
1) This is the case in general terms; but
2) More specifically, the AEMO has defined logic for determining ‘Conformance Status’ which references the Off-Target metric, so is another (more specific) reason to watch this number.
(A) Calculating ‘Off-Target’?
Readers should understand that:
1) This is a Metric that applies for every DUID for every Dispatch Interval.
2) The metric is calculated on an individual unit (i.e. DUID) basis … and works for both sides of the supply-demand balance:
(a) It works for most Supply-Side units
i. … specifically for any Scheduled or Semi-Scheduled unit … specifically:
> generators;
> battery discharge; and
> can only be implied for Negawatts (i.e. Raw Off-Target is 0MW by definition, because compliance is assumed to be perfect within NEMDE itself … i.e. you can’t measure an invisible negawatt!).
ii. It can’t be calculated for Non-Scheduled units because of the absence of a Dispatch Target from the AEMO.
(b) Demand-Side consumption units (Scheduled Loads);
Because it’s possible for a unit to be ‘too high’ or ‘too low’ at different points in time, it’s necessary that there are both negative Off-Target outcomes, and positive ones. In some of the AEMO’s Operations Procedures, however, there is some ambiguity about what is positive and what is negative.
For this reason (amongst others) the way in which we have calculated Off-Target has evolved between the prior GRC2018 and the GSD2019.
These changes reflect our evolving understanding of the ways in which this metric can be useful – including with respect to AEMO’s logic used in determining Conformance Status.
(B) Why might units be ‘Off-Target’?
Before we get into the specifics of how this has been calculated (and how the specific formula has evolved over time) it’s important to reinforce that (as noted in this article), units can be Off-Target for a number of different reasons, including the following:
Reason 1 ) units sometimes trip
Reason 2) units sometimes fail to start
Reason 3) sometimes there’s a discrepancy in the forecast output of Semi-Scheduled plant for whatever reason
Reason 4) sometimes there are intentional (but unexpected) deviations from target, such as in response to unexpected price outcomes.
Reason 5) in that article from January 2019 we noted that a unit might be Off-Target because it was providing FCAS services. However this was based on the simplified formulation of Off-Target used in the GRC2018, and readers will see (below) that our formulation has matured such that:
5a) Whilst provision of Regulation FCAS is a reason why a unit might have a ‘Dispatch Error’;
5b) Provision of Regulation FCAS (when Enabled) should not be a reason for a non-zero Off-Target (though, on rare occasions, perhaps provision of Contingency FCAS might be).
Hence readers will understand that Off-Target is of interest not just because of the quantum (and number sign) of the deviation, but also because of the reasons behind why it has happened.
(C) How ‘Off-Target’ has been calculated in our products
We have progressively grown more sophisticated in terms of how we treat Off-Target in our products:
(C1) Method #1 (Simplified Off-Target in the GRC2018)
Our 1st attempt to calculate Off-Target was in the Generator Report Card 2018. In this case we used a simplified definition of Off-Target, being the following:
Simplified Off-Target (for DI) = Interval Output – Interval Target
Hence that meant with the number sign:
Positive Simplified Off-Target indicates that Output was above the AEMO dispatch target (over-production); and
Negative Simplified Off-Target indicates that Output was lower than the AEMO dispatch target (under-production).
– hence (using this logic) a large negative Off-Target would be a possible unit trip.
… but note that the number sign is now reversed as noted below.
This article was written in January 2019 (in the development of the GRC2018) using this context.
(C2a) Method #2a (Off-Target in ez2view)
Building on the lessons learnt through the GRC2018 process we introduced the Off-Target concept into ez2view (the ‘Unit Dashboard Widget’) through progressive developments during 2019.
In doing this we upgraded the calculation so it would more directly reflect the AEMO’s automated processes (as described in their Operations Procedures) with respect to Conformance Status. We chose to:
1) Reverse the direction of the calculation, to match what’s (more clearly) prescribed for Dispatch Error in the Operations Procedures, and then
2) Also add in the allowances for Enablement for Regulation FCAS, as prescribed in the procedures.
(C2b) Method #2a (improved ‘Raw Off-Target’ in the GSD2019)
In the Generator Statistical Digest 2019 we calculate Off-Target in the same way as we calculate this for the ez2view software.
We label it ‘Raw Off-Target’ (when we present statistical summaries on for each month in CAL 2019) to reinforce that other considerations need to be taken into account before determining ‘Conformance Status’, as is done in ez2view:
• For Semi-Scheduled units, the status of the Semi-Dispatch Cap in each dispatch interval needs to be taken into account; and
• For all Scheduled and Semi-Scheduled units, the ‘Small Trigger’ and ‘Large Trigger’ are also taken into account.
(C) Progressing to ‘Aggregate Raw Off-Target’
In several of our analytical products (such as in Appendix 17 within GenInsights21) we have extended the concept of a unit-level Aggregate Raw Off-Target to aggregate in several collections of units … which is why we coined the term ‘Aggregate Raw Off-Target‘.
More recently we’ve prepared this linked Glossary Page to clarify how this is calculated, and where it is discussed.
(D) Articles about ‘Off-Target’
Articles published on WattClarity at various points in time that have been tagged with Off-Target are here.