With increasing shares of renewables, the current electricity system is operating at its limits. Most countries apply balancing markets to secure system stability. So far, the transmission system operators in the EU (or Independent System Operators (ISO) in the US) have been able to secure system stability by relying on the capacity covered on the balancing market. However, in the case of Germany, we can see today that due to the already significant share of renewables in the system, the TSO have to apply non-market based measures to secure system stability.
As discussed in our last post, the TSOs have significantly increased their application of congestion management, the application of redispatch and feed-in management, over the last years. Especially in the case of feed-in-management, we can expect that the trend that is currently increasing will continue to do so within the next years and that similar developments will take place in other countries and US states.
The increasing application of feed-in management is one sign for the increasing coordination problem between distributed generation and network operation (for more details on the coordination problem see this post). In Germany, different measures are currently investigated or applied to address the coordination problem. Among others, the network operators are now allowed to curtail 3% of the annual production of each renewable generator. While this solution might reduce the pressure on the grid, it does not address the core of the problem: Most local flexibility is not available to the DSO or TSO for congestion management. In addition, distributed generators have no information about their impact and potential congestion that might be induced by their own actions. In the near future, we want to address this shortcoming of the market design and secure that
a) local resources get an information about their potential impact on the congestion in the networks (and potential costs) and
b) network operators (both the TSO and the DSO) get access to local flexibilities via regulated or market-based price incentives for the local flexibilities.
How to achieve both aspects is currently discussed in the US and in Europe, especially in Germany (which is due to the fact that Germany already has high costs for congestion management at the moment). In the following, we sum up the first results of a European study which provide a first overview about the European debate on potential solutions to the coordination problem between the TSO and the DSO. In our next post we will compare the discussed approaches on the European level with the models that are under discussion in Germany.
The discussion in Europe: First results of the SmartNet EU project
In Europe, the discussions about potential concepts for regional markets currently intensify. A recent study by Gerard, Rivero & Six (2016) provides a nice overview of potential solutions from a European perspective. Though this is still a theoretical debate, we will dive deeper into the models discussed by Gerard, Rivero & Six. In their report, Gerard, Rivero & Six differentiate between five different concepts to contract distributed resources for ancillary services (AS) from the distribution grid level. Not all of them seem to be applicable in the real world, but some could provide a basis for a more detailed discussion about the potential design of regional electricity markets. All of these models will be tested in three different field tests in Italy, Spain and Denmark in the course of the SmartNet EU-project. As soon as new results are available in these field tests, we will discuss them here on enerquire.
Let us now take a look at the five models investigated in the SmartNet project:
Centralized AS market model
The TSO operates a market for both resources connected at transmission and distribution level. Via these market the TSO directly contracts flexibility from distributed generators on the distribution grid level without taking local network restriction on the distribution grid level into account. The DSO is only involved in the contraction of distributed flexibility if a prequalification process is in place which ensures that the activation of the specific flexibility does not result in network congestion. If the DSO wants to include its grid constraints into the AS market, the DSO has two options in this model: A) The DSO provides the necessary data on potential grid constraints for a specific trading period to the TSO, or B) the DSO gives the TSO full access to the network data on the distribution grid. Figure 1 summarizes the centralized AS market model.
Figure 1: Centralized AS market model: high-level view of roles, market architecture and stakeholder interactions Gerard, Rivero & Six (2016):32
The primary advantage of this system lies within its simplicity. There are not many complex tasks to handle and a very limited number of actors involved. Therefore, the operational costs of this system are likely to be rather low.
On the downside, this model does not secure a coordination between TSO and DSO, except in the case that the TSO gets full information from the DSO about the network constraints. Even if such a data exchange from the DSO to the TSO was established (which seems unlikely at the moment), the potential costs of this data exchange system would probably outweigh the advantages associated with the central AS model. Therefore, network constraints on the distribution grid level are not addressed by this model, and DSOs don’t have access to local flexibility, which raises the question whether the centralized AS model is future-proof at all.
Local AS market model
The second concept delegates the responsibility to operate a local AS market to the DSO. The DSO is responsible to clear the market and, additionally, aggregates the flexibility via the AS market and offers those capacities to the TSO that the DSO does not need to balance its own grid and that do not result in new network constraints on the distribution grid level. The offer from the DSO then is part of the central AS market of the TSO where it competes with flexibilities from other distribution grids and the transmission grid. Figure 2 summarizes this model.
Figure 2: Local AS market model: high-level view of roles, market architecture and stakeholder interactions Gerard, Rivero & Six (2016):33
The strength of this approach lies in the potential for the DSO to reduce the overall costs of network operation by contracting flexibility on a local level. However, this model has to address several challenges. Most prominently, liquidity of local markets is discussed quite intensively. If every DSO operates a separate local AS market for its grid, the number of flexibility providers might be rather small, which increases prices and thereby reduces the efficiency of the local market. While this problem can be addressed by establishing a local market for a larger group of distribution networks, there are other challenges to overcome. For example, it needs to be secured that the DSO’s balancing activities on the local market do not result in new constraints on the transmission level. Additionally, market parties like aggregators don’t have the possibility to directly offer their products to the TSO and the products traded in the different local markets might differ from each other, which again increases costs. Therefore, standardization and harmonization of the different local markets are key issues here.
Shared balancing Responsibility Model
This rather theoretical model is based on the Local AS market model discussed before, but with the important difference that local flexibility can only participate in the local AS market operated by the DSO. The DSO does not aggregate and offer local flexibility to the central AS market of the TSO. Therefore, the TSO can only access flexibility connected to the transmission gird. Importantly, the DSO becomes the balancing responsible party (BRP) for its network and has to use the local flexibility to address deviations from the pre-defined (e.g. via a spot market) schedule. Figure 3 gives an overview of this model.
Figure 3: Shared balancing responsibility model: high-level view of roles, market architecture and stakeholder interactions Gerard, Rivero & Six (2016):35
On the positive side, this model reduces the costs for balancing for the TSO, which is a direct effect of delegating responsibility for balancing to the DSOs. Additionally, the DSO can gain from the fact that it can address local network constraints via its own market. However, there are many challenges for this model as well. Especially, the risk of increasing costs for the DSO is quite high, especially if liquidity is low within the local market: The smaller the number of potential flexibilities available for the local market, the higher the prices. Additionally, the TSO faces the new risk that the system might become instable if one DSO does not fulfil its task as local balancing responsible party. For the market parties, the model has the same disadvantages as the local AS market model.
Common TSO-DSO AS Market Model
The primary feature of this model is that network constraints of the transmission and distributed grid are considered for the clearing process at the same time. Two variants of this model are possible:
A) a centralized approach with both, the TSO and the DSO, jointly operating a central AS market and optimizing flexibility procurement together, or
B) a decentralized variant, where the local AS market of the DSO first runs a clearing process (which is for optimization purposes only, no actual interaction with market participants is taking place at the local level) and then actively coordinates its local optimization with the clearing on the central AS market operated by the TSO. Only at the central market of the TSO, market parties are actively contracted. The decentralized variant requires a continuous exchange of information between the local AS markets of the DSOs and the central market of the TSO. Figure 4 gives a general overview about this model.
Figure 4: Common TSO-DSO AS market model: high-level view of roles, market architecture and stakeholder interactions Gerard, Rivero & Six (2016):37
The primary strength of the common TSO-DSO model is the combined optimization of the costs of TSO & DSOs, which should result in lower grid costs compared to the other models. While the cost allocation between the network operators will become a complex issue, it seems to be favourable from a grid-cost optimization perspective. For variant B (decentralized local markets) the liquidity issue will be a primary challenge as well. Furthermore, the need for communication and data infrastructure will be significant for both variants.
Integrated Flexibility Market Model
The last model introduced by Gerard, Rivero & Six (2016) focuses on a competitive approach. In this scheme the network operators (TSO & DSO) compete with market parties (Balancing responsible parties and others) on a central flexibility market for local flexibilities as well as with flexibility providers that are connected to the transmission grid. Though network constraints from transmission and distribution are the basis of the clearing process, there is no priority of the network operators. Rather, the highest bid wins, which leaves the decision who contracts a certain flexibility to the individual willingness-to-pay of each market participant. Additionally, the network operators is allowed to resell unneeded resources that were contracted earlier. The central AS market in this model is operated by a neutral third party, not by the network operators themselves. Thereby, the current separation between balancing and spot markets will be abolished. See figure 5 for the general idea of this model.
Figure 5: Integrated flexibility market model: high-level view of roles, market architecture and stakeholder interactions Gerard, Rivero & Six (2016):39
On the positive side, the Integrated Flexibility Market Model introduces more competition to the market, which should reduce costs. Additionally, as BRPs can directly access flexibility to balance their portfolio, the need for flexibility procurement by the network operators should decrease. Furthermore, the network operators get access to flexibility that was previously contracted by other network operators via resell. This could increase liquidity and reduce costs as well. On the downside, the competition between the network operators from different network layers (TSO vs. DSO) for flexibility might result in higher prices at certain points in time, increasing the costs for the network operators. From the network operators’ perspective, one downside of this model (central market operated by a third party) is that network operators will have to share network data with the central market operated by an independent third party. As data is becoming a key resource with the increasing digitalization, this might be an important issue for the network operators. Furthermore, the central market authority has no obligation for prequalification of distributed resources, which might result in conflicts with the distribution grid operators. Furthermore, the more parties participate and contract flexibility on the AS market, the lower the liquidity of other markets might become (e.g. the spot market).
Not every model suitable for all ancillary services
Ancillary services differ from each other in many respects. Generally, we can differentiate between four different types of ancillary services:
Table 1: Mapping of ancillary services and coordination schemes Gerard, Rivero & Six (2016):55
Table 1 illustrates that not every model discussed in the SmartNet projects suits all ancillary services. For now, it is an open debate whether we should address all ancillary services via one market platform, or whether it makes more sense to establish (two) separated markets for TSO, DSO and BRP needs.
The European debate currently focuses on market solutions
Though the discussions in Europe are still in flux, there seems to be a stronger case for the TSO-DSO cooperation model. From a theoretical perspective, especially the last two models described above, the Common TSO-DSO model and the integrated flexibility market model, seem to have the overall potential to result in positive CBAs, meaning that these models might reduce the costs of the energy transition compared to the current situation (which in most cases in Europe equals the first model Centralized AS market model). The field tests within the Smart Net project should provide us with additional insights on the strengths and weaknesses of each model and provide a basis for further discussions about the organization of flexibility or AS markets in Europe.