Questions & Answers

The ATC extraction method

Q: What are the optimization variables in addition to the capacities Cn and Cn,o?
I was thinking that the net positions could be part of the optimization variables but you would then end up with one particular set of net positions in the optimal solution which is not what is desired. Basically, I have difficulty understanding how to capture “for all possible net positions” in an optimization setting.

A: No other variables besides the Cn and Cn,o are in the ATCE method. The ATCE method takes the DA FB domain as input to compute the DA NTCs. Under the ID allocation, the extracted NTCs are the constraints, as inputs, for the single intraday market coupling to determine the net positions, and all necessary grid constraints are already represented in the FB domain. If further constraints are needed on the net positions as inputs to the single intraday market coupling, those will be allocation constraints, to be considered directly by the intraday market coupling algorithm.

Q: How do capacities Cn and Cn,o relate to the constraints? Are they included in the RAM term of cross-border CNECs?

A: The Cn and Cn,o are the exogenous variables in the ATCE optimization. This optimization needs to respect the FB constraints, which are on the CNEC level. The extracted NTCs (i.e. Cn and Cn,o, being the results of the ATCE optimization) are on the bidding zone border level. The formula z2zPTDF * exchanges (i.e. exogenous variables) < RAM links the CNEC level z2zPTDF to the bidding zone border level exchanges. The Cn and Cn,o are not directly applied to the RAM of any CNECs.

The FB domain is modelled in the ATCE as z2zPTDF+ * C <= DA_RAM, where z2zPTDF+ refers to the z2zPTDF applying the positive PTDF filter and the threshold parameter. C refers to the variable vector of Cn and Cn,o.

In the mathematical terms, the inequality constraints are modelled as A*x<=b.

A is the z2zPTDF+
x is the vector C containing Cn and Cn,o variables
b is the DA_RAM

Corresponding to the matrix implementation, the z2zPTDF+ is a m-by-n matrix, m is the number of constraints, e.g. the number of presolved CNECs. n is the number of directional borders. Thus, the C is an n-by-1 vector consisting of [Cn and Cn,o], and the DA_RAM is also a n-by-1 vector.

Capacity calculation
(parameters/capacity)

Q: Remedial Actions (RA), how are they included in the capacity calculation process?
A: Non-costly RAs are always considered in the capacity calculation process. For the costly RAs it will be assessed whether they are available and can be utilized to increase the capacity on internal CNECs or internal combined dynamic constraints. If they are indeed available, they are applied to ensure the 70% rule despite the costs associated with them.

Q: Domain Validation – what is it?
A: It is a quality check performed by the TSO operators to validate the capacities calculated by the coordinated capacity calculator (Nordic RSC). Note: the Nordic RSC has changed to Nordic Regional Coordination Centre (Nordic RCC), which is now a separate legal entity.

Q: What would happen if you are forced to use intuitive FB?
A:Some part of the benefits of introducing FB are discarded. ACER decision 04/2020 removed the possibility of having an intuitive patch in the market coupling algorithm; as such, the application of FB intuitive is no longer an option.

Q: Will there be overloads in FB?
A:There are no overloads on monitored CNECs or combined dynamic constraints (i.e. the CNECs and combined dynamic constraints taken into account in the market coupling) in the DA market outcome, but due to changing conditions close to real-time and numerical discrepancies, there may be overloads in the operational hour that are mitigated by means of redispatch.

Q: Can you have negative capacity under FB?
A: Yes, however, when a negative RAM is calculated for the DA timeframe, the RAM value shall be set to zero and the potential overload shall be managed by redispatch. This is different for the ID timeframe, where a negative RAM will be applied in the allocation and/or via remedial actions.

Q: In regard to GSK strategies; what are the choices to be made by the TSOs?
A: As we have a zonal market system, the TSOs need, for each time frame, to make assumptions about how variation in a zonal net position is distributed amongst generating units and loads in the bidding zone. It should be noted though, that there is no difference between the NTC and FB world in this respect, and also – it is done already in today’s NTC calculations – though not as formalized nor coordinated. In FB capacity calculation, all uncertainties in the capacity calculation process add up in the RM, including the uncertainties linked to the GSK. As such, the different GSK options can be assessed by computing the corresponding RM values, reflecting the uncertainty that is linked to the use of the different GSK strategies, and to opt for the one that brings the lowest uncertainty.

Q: Is the GSK system prone to gaming?
A: The GSK in itself does not affect how a producer is selected over another in the market coupling, as such it is not prone to gaming.

Q: Will the operational expertise and experience be part of the FB capacity calculation process?
A: Yes, but the FB capacity calculation is a more formalized methodology than the one applied today so that there is less dependency on operator experience. The TSOs are, however, responsible for providing the input data, as well as for validating the results from the capacity calculation – the FB capacity domain.

Q: There is a perceived disadvantage that under FB internal constraints can be taken into account, as this is not transparent. Would it not be preferable to countertrade or redispatch and have cross-border constraints only?
A: Both FB and NTC are equally able to take the same internal constraints into account. With NTC being a “black box” where scenarios and internal constraints are captured in one single value on each bidding-zone border, the level of detail under FB is higher, and the use of internal constraints is more transparent. Countertrade and redispatch are remedial actions (RAs) as defined in Article 9 of the CCM.

As described in Article 8 of the CCM, to avoid undue discrimination between internal and cross-zonal flows, the TSOs will take actions to meet the requirement from Article 16(8) of the Regulation 943/2019. For the costly (and non-costly) RAs it will be assessed whether they are available and relieve flows on internal CNECs or internal combined dynamic constraints (Article 10 of the CCM), and whether it is economically efficient to apply RAs. If RAs are available, they are applied to ensure the 70% rule despite the costs associated with them to meet requirement set in Article 16(8) of the Regulation 943/2009.

Q: How does the ACER recommendation on internal constraints affect the Nordic FB methodology?
A: The Nordic CCM follows Article 16(8) of the Regulation 943/2019 and the related ACER recommendation 01/2019. This is elaborated more on rules for avoiding undue discrimination between internal and cross-zonal exchanges in Article 8 of the CCM and will be monitored as recommended by ACER in the recommendation 01/2019.

Q: Are stability/dynamic issues simulated in offline grid models?
A: Yes, the stability/dynamic issues are simulated in offline grid models that are enhanced for the dynamic analysis.

Q: What will happen with unforeseen outages of e.g. power plants?
A: If the information is available before the DA / ID capacity calculation starts, it will be taken into account. If the information is available during the DA capacity calculation or after the DA firmness deadline, it will trigger a recalculation of the ID capacity.

Q: Will the PTDF matrix be stable for all 24 hours?
A: No, they will vary from hour to hour, as grid topology and grid usage changes over time.

Q: To what extent are the exchanges with Germany and the Baltics influenced by a Nordic FB?
A: A so-called advanced hybrid coupling is applied in the Nordic CCR to model the exchanges from/towards the neighbouring CCRs on the HVDC links. A DC link (being a fully controllable active power flow) is an NTC by nature. Combining these NTCs with the FB methodology applied for the AC grid is done by means of the advanced hybrid coupling: the converter stations of the DC interconnectors are modelled as ‘virtual’ bidding zones in the FB system (a bidding zone, without order books though), having their own PTDFs reflecting how the exchange on the DC link impacts the AC grid elements.

Q: There are currently no phase shifters in the Nordics. If such a device is installed, does it fit in the FB methodology?
A: Yes, it is foreseen to be modelled as a DC link where the allocation mechanism sets the power flow on the controllable device using ‘advanced hybrid coupling’. However, if the phase shifter is ‘small’, e.g. Charlottenberg-Eidskog (NO1-SE3) being connected to the lower voltage grid, operated jointly by TSO and local DSO, it can be managed directly by the involved TSO/DSO as inputs to the capacity calculation, instead of explicitly modelling it as virtual bidding zones.

Q: Under FB, will the capacity on the DC cables deviate from the nominal capacity (e.g. due to the west-coast cut)?
A: No, the capacity on DC connectors will – in principle - be the nominal value; it is the market clearing algorithm that determines the optimal flow on the DC cable. Bymodelling the HVDC links as virtual hubs, the DC links compete with all other exchanges to make use of the scarce capacity in the AC grid. A restriction on the west-coast cut may in this way have an impact on the resulting flow on the DC cable.

Q: Are losses in the AC grid taken into account in the FB model?
A: No, as this would require a non-linear market model.

Q: Regarding DC links - how will ramping constraints be handled in FB?
A: They are treated as separate allocation constraints and are provided to the NEMOs to be taken into account in the market coupling. They are not part of the, or expressed as, FB parameters.

Q: Can you confirm that the FB domain will always be larger than the NTC domain?
A: Yes, from theory, and yes from a practical point of view, when both the FB and NTC are subject to the same level of operational security.

Capacity allocation
(market results)

Q: How are you able to perform market simulations today?
A: We are using the simulation facility, which mimics theEuphemia algorithm used for the SDAC in the production environment. More details can be found in the Euphemia Public Description

Q: What setup has been used in the market simulations?
A: Historical order books and capacities in the SDAC, reflecting the current market coupling. (. For the Nordic FB market simulations, the same setup is used but the NTC capacity constraints in the Nordic CCR are replaced by FB constraints.

Q: If you change the maximum price at the PXs will the outcome of the optimization change?
A: No, in essence not (unless the market clears at the maximum price).

Q: What is the optimization that will be used in the FB allocation mechanism?
A: It is the same as today: it is the social welfare (consisting of consumer surplus, producer surplus, and congestion rent) that is optimized. The change is introduced through the constraints to the optimization. NTC applies constraints to the flows on each bidding zone border (NTCs), and FB introduces constraints on the maximum allowed flows on CNEs (PTDF matrices).

Q: Why is the Congestion Revenue (CR) part of the social welfare?
A:Congestion revenue is a natural part of the social welfare when transmission capacity turns out to be a scarce good, expressed by the price difference. Note: The objective function of the NTC and FB market coupling is the same: it is the social welfare that is maximized. Congestion revenue as such is not maximized.

Q: How will the system price be calculated in FB?
A: System price is a reference price computed by Nordpool. This work is not related to the Nordic TSOs or the Nordic RCC.

Q: Will the auction structure on the DA be changed?
A: No, it is at least not within the scope of this project.

Long Term Capacity Calculation

Q: What does the LT CCM cover?
A: At least the year-ahead (YA) and month-ahead (MA) capacity calculation. Please refer to the approved LT CCM and the presentation regarding the LT during the June 29, 2022 stakeholder event

Q: MA forecasts are needed, what will be provided to the market participants?
A: The FB domain, and the extracted NTCs based on the FB domain, as interim solution, will be provided to the market participants. The target solution is to provide the FB parameters.

Intraday market

Q: FB is the better capacity calculation methodology close to real time (ID, balancing). Why start with DA instead of ID?
A: The majority of the trades occur on the DA. The ID platform XBID cannot support FB yet. As the DA platform Euphemia already supports FB it has been decided to harvest the benefit of FB.

Q: Is there a risk to end up in a corner of the FB domain after the DA stage, and that the ID trade is stuck?
A:It is possible but not likely, thanks to the design of the ATCE method. The ATCE method uses the DA leftover capacities to compute a transitional solution ID ATC. The Nordic TSOs then apply this transitional solution at the ID gate-opening before the XBID can handle the FB constraints. Given that the DA market clearing point is already at the ‘corner’ of the (original) DA FB domain, the ATCE method is designed to adjust the original FB domain by introducing the ‘maxz2zPTDF threshold’, negative PTDF filter and additional RAM on CNECs where necessary, accepting certain manageable operational risks. In fact, the ATCE method is applied within this ‘adjusted’ FB domain. Consequently, the DA market clearing point is highly unlikely not being stuck at the corner of the ‘adjusted’ FB domain.

When there is a need to perform the ID capacity calculation reassessment, triggered by the availability of the newly created CGMs, the ATCE method is further enhanced by adopting the ‘shortest distance approach’. This approach eliminates the risk of infeasible solution of the ATCE method, where the latest market point, considering the latest continuous trading information, violates the FB domain computed by the latest CGMs. Please refer to the ATCE descriptive document for more information

Q: If you use non-intuitive FB in the DA market, how will this work out on the ID market?
A: DA and ID market flows can be different, like today. The ID capacity will result from a dedicated ID capacity calculation. As such, there may be ID capacity available or not. ACER decision 04/2020 removed the possibility of having an intuitive patch in the market coupling algorithm; as such, the application of FB intuitive is no longer an option.

Q: Is there less capacity for the ID after FB is introduced?
A: Conceptually for the transitional solution (i.e. by applying the ATCE method), the ID capacity of a border or a bidding zone, at the ID gateopening, can be larger or smaller than the current NTC capacity for the same MTU. The actual observations of the ID capacity comparison between the extracted ID ATC and the currently operational ID initial capacities can be found in the NRA evaluation report and the ATCE data publication

Q: Is it true that the optimal solution (in terms of welfare), referred to, holds only for the spot market?
A: Yes, the welfare assessed is ‘day-ahead welfare’ only. Welfare considerations of subsequent market timeframes, ID and balancing, are not considered.

Q: How to split capacity between DA and ID?
A: Capacity calculation is a continuous process: by using the latest available information, the most capacity is provided for the upcoming timeframe(s). Or in other words: for the ID timeframe dedicated grid models will be created and dedicated capacity calculation will be performed to serve the ID market as good as possible, considering the already allocated DA capacity. Note, in this respect, that an integral part of the capacity calculation is the assessment of the uncertainty that the TSOs are facing in their capacity calculation. It is expected that the uncertainty for the DA stage is larger than that for the ID stage, as better forecasts are available for the ID and less assumptions need to be made. The Reliability Margin (RM) reserved at the DA stage can thus partly be released on the ID stage.

Q: Shouldn’t there be the same capacity calculation methodology (CCM) for DA and ID?
A:This is indeed the target: FB for both DA and ID. The CACM requires that a CCM is proposed for both the DA and the ID; they can be different. As the current release of the XBID solution does not support a FB model, FB ID is not feasible initially, and an ATC extraction will be applied as an intermediate solution for ID.

Q: The SH information tool shows the border flows based on the Nordic NPs. If you plug in the DA NPs that are determined by the PXs, can you assess the expected ID capacity?
A: Yes, please refer to the published ATCE method description for the ID gateopening to better understand the relationship between the DA MC results, the extracted NTC and corresponding ID ATC, amongst others.

In short, the ATCE method produces the extracted NTC values based on the DA FB domain and the DA market outcome. Together with the DA AAF information, one can assess/compute the ID ATC, using ID ATC = extracted NTC – DA AAF, for the ID gateopening. As a reminder, the TSOs publish the extracted NTC and ID ATCs for the ID gateopening. After that, e.g. for later ID auctions (22.00 D-1, or 10.00 D) a new capacity calculation may be performed, based on an updated / dedicated (D-1 or ID) CGM.

Q: FB ID: when will this be applied?
A: The FB ID go-live is dependent on many factors. One of the most important aspects is the ability of the SIDC platform handling the FB constraints.

Q: How often will you recompute the ID capacity?
A: The frequency of the reassessment of intraday cross-zonal capacity shall be dependent on the availability of input data relevant for capacity calculation and execution of intraday auctions, as well as any events impacting the cross-zonal capacity. Please refer to the approved Nordic DA/ID CCM article 21.

Data provision and tools provided

Q: Are you able to do an ATC extraction from the FB domain (for information purposes)?
A: Yes, but you have to be aware that the ATC domain extracted from the FB domain is arbitrary – there is actually an infinite solution space. The current DA procedures at the Nordic RCC and TSOs do not include such an ATCE extraction before the DA market coupling.

Q: Will the shadow prices of the internal CNECs and internal combined dynamic constraints be published?
A: The shadow prices of all market relevant CNECs will be publicly available on the Nordic RCC website and the JAO platform during the EPR.

Q: What is the shadow price?
A: The shadow price is a by-product of a constrained optimization; there is a shadow price for each constraint in a constrained optimization. Within the FBMC context, where the FB constraints act as constraints in the market optimization, the shadow price of a FB constraint indicates the social welfare increase when we would increase the margin of the FB constraint with 1 MW. It is a kind of ‘price-tag’ that is labelled to the FB constraint on a specific hour and day.

Q: During the external daily parallel operation, will the FB capacities and simulated market results be published on a daily basis?
A: Yes, it is of utmost important to have this in place at that time. Yes, they are publicly available during the EPR. The FB CC results are available on the JAO website (updated daily including the FB CC results).

The the FB MC and other associated information are published on the Nordic RCC website (updated weekly including the market coupling results with daily resolution. The publication is delayed due to grace period of the simulation facility.)

Q: Is input expected from the generators to build the D-2 CGM?
A: The generator information is embedded in the individual grid models from the TSOs, as inputs to build the D-2 CGM.

Q: Does the FB model in the in the JAO publication tool include the voltage and dynamic limits?
A: Yes, it does, expressed as Fmax on combined dynamic constraints.

Q: In the data to be published, can you identify from the hourly results if a constraint is internal or cross-border?
A: In principle, yes. The data may be incomplete on some CNECs.

Q: What are bidding zones DK1A and NO1A?
A: DK1A and NO1A are virtual areas (DK1A is a sum constraint on the flow on Skagerrak and Kontiskan HVDCs. NO1A is a sum constraint between (NO5 and NO2) and NO1), not having producers /consumers, that are applied under the current NTC.

Q: Will the PTDFs published in the Nordics contain the names or unique IDs (from day to day) for CNECs or combined dynamic constraints? "Decoding” of the PTDFs seemed to be a big topic in CWE.
A: The names and unique IDs of all CNECs are published according the Nordic CCR NRAs’ decision, except the Swedish ones, due to the Swedish national security legislation which covers all Swedish CNECs and all Swedish contingencies associated with the neighbouring countries’ CNEs).

Art 25.3. “Individual Nordic TSO may choose not to identify the CNEC concerned and specify its location when publishing the information referred to in paragraph 2(c) if it is classified as sensitive critical infrastructure protection related information in their Member States as provided for in point (d) of Article 2 of Council Directive 2008/114/EC of 8 December 2008 on the identification and designation of European critical infrastructures and the assessment of the need to improve their protection. In such a case, the withheld information shall be replaced with an anonymous identifier, which shall be stable for each CNEC across day-ahead and intraday capacity calculation timeframes, as well as all long-term capacity calculation time frames. The anonymous identifier shall also be used in the other TSO communications related to the CNEC and when communicating about an outage or an investment in infrastructure. The information about which information has been withheld pursuant to this paragraph shall be published on the communication platform referred to in paragraph 1.”

Q: Will the common grid model be available?
A: No. The common grid model will not be publicly available.

Q: What about information about transmission outages that are published in UMMs today? How will this be communicated in FB?
One of our biggest concerns is how we will be able to assess the effect of grid outages – both in our short term and long term price models. This is both due to outages and other changes in the grid (e.g. new lines).
A: The UMM information is published on the NUCS platform. The Nordic TSOs and the Nordic RCC are developing the NUCS methodology to capture the impact of the outage on the FB parameters. Please refer to the presentation on the NUCS methodology at the stakeholder event on June 29, 2022

Q: Is there a possibility to have 10-14 days forecast for the PTDFs published?
A: No. However, the monthly forecasts, adjusted by information for planned outages which will be published, might provide a simple basis for such forecasts. Please note, the results of monthly FB CC, including the PTDF matrices, will be available 1 year after the DA/ID CCM FB go-live.

Q: With regard to transparency – will you publish the same information as on the continent?
A: No, the data publication in the Nordic context is not the same as the continent. The Nordic TSOs will comply with data publication requirements according to the approved Nordic apporved DA/ID CCM, Transparency regulation, REMIT and national security legislation.

The ATC extraction method

Capacity calculation (parameters/capacity)

Capacity allocation (market results)

Long Term Capacity Calculation

Intraday market

Data provision and tools provided

Other questions linked to CCM

Capacity calculation
(parameters/capacity)

Capacity allocation
(market results)