TSO & DSO coordination – key for the integration of renewables

Like this post? Let us and others know by sharing it!

Coordination is one of the key challenges for electricity systems based on distributed and renewable generation. In our last post we discussed the coordination problem that evolves between electricity generators and the distribution grid operators. Today, we want to focus on another coordination issue: The coordination between the different network layers.

In general, there exist two network levels: The distribution grids connect the smaller electricity generators and most consumers (e.g. households and commercial consumers) to the electricity system (distribution networks work with 110 kV or less in Germany). The distribution networks are connected to the transmission grids (220 kV and above in Germany), which transport electricity over larger distances. The transmission grids connect the electricity networks across borders (these cross-country connections are called interconnectors) and large conventional power plants (hard coal, lignite, gas and nuclear) are connected to the transmission grid in most cases as well. Renewables, on the other hand, are primarily connected to the distribution grid. An exception are the offshore wind farms, which feed into the transmission grid directly.

The situation today – The TSO focused system

In today’s electricity system the transmission system operator (TSO) has the responsibility to secure system stability. To fulfil this obligation, the TSO uses data on current and projected power generation, the status quo of the relevant network infrastructure, the capacity that is internationally traded and will be transported via the interconnectors, potential congestion and the flexibility potential that can be offered by power plants and large electricity consumers. Different systems are in place to help the TSO to balance the grid, e.g. balancing markets, redispatch and the curtailment of renewables as a last resort.

Redispatch and curtailment – two different congestion measures

Importantly, redispatch and curtailment are not the same. Basically, redispatch allows the TSO to demand a modification of the intended schedule of power plants (via trading) to secure grid stability. One power plant that is located in front of the congested network line reduces power generation and another power plant behind the congestion increases generation. In most cases, the power plant that is forced to reduce production can produce cheaper than the power plant that increases its generation for the redispatch. Therefore, redispatch results in additional costs that are covered by the network charges. Curtailment, on the other hand, reduces the feed-in from renewable generators and the generators are paid a full compensation fee. The curtailment of renewables is labelled feed-in management in Germany.

The electricity system reaches its boundaries

So far, the TSOs manage to operate the system in Germany quite well. Even with renewables, the current system works on a secure level. However, with an increasing share of renewables, the TSOs have to increase the application of redispatch and feed-in management.

Redispatch costs tripled from 2013 to 2015

First, let us take a look at how redispatch has developed. The following data on redispatch and congestion management is taken from the German regulator BNetzA. You can access the data here at the BNetzA. From 2013 to 2014 the number of days on which redispatch was required increased by roughly 50% from 232 in 2013 to 330 days in 2014. Therefore, the costs for redispatch increased by 50 million from 2013 to 2014 (186.7 million). More importantly, in 2015, these costs more than doubled (412 million € in 2015) and the total amount of energy affected by redispatch reached 16.000 GWh in 2015 (compared to 5.197 GWh in 2014). We need to note here that 2015 was a quite special year for redispatch in Germany, as many different factors and developments (lots of wind, unplanned shutdowns of power plants etc) added up and increased the need for redispatch in 2015. In 2016, the overall costs of redispatch fell again. Remarkably, the first quarter in 2017 faced a new record-high demand for redispatch in Germany. In January alone, the need for redispatch equalled more than 40% of the annual redispatch requirement of the whole year 2016 (for details see BDEW 2017)). Primary driver for the significant need for redispatch was the high feed-in from wind energy in Germany.

Feed-in management – curtailed energy increased 8 times from 2013 to 2015

In addition to redispatch, the need for feed-in management, i.e. curtailment of RES, has increased significantly from 2013 till 2015. In 2013, about 555 GWh of electricity production from RES was curtailed. This number increased to more than 4.722 GWh in 2015.

Lost energy in Germany from feed-in management 2009-2015

own illustration based on data from BNetzA

Accordingly, the compensation payments for the curtailed renewables increased by more than seven times from 43 million € in 2013 to nearly 315 million € in 2015.

Compensation payments for feed-in management of renewables in Germany 2009-2015

own illustration based on from BNetzA

Together, the costs of redispatch and feed-in management accounted for more than 715 million € in Germany in 2015. These costs indicate the current inefficiency of the system and the increasing pressure on the systems stability. While the costs for redispatch did not increase any further in 2016, the development of compensation payments for feed-in management points at increasing costs for congestion management in Germany.

The speed of the energy transition, the diffusion of renewables, is the key driver for the increasing application of congestion management. Or, to put it differently, the electricity system struggles to keep up with the fast development towards decentralization.

The challenge: System stability requires participation of distributed generators from the distribution grid level in the future

As mentioned before, the TSO should use the balancing market to balance the energy system. However, especially the increasing application of feed-in management on the distribution grid level illustrates that the current balancing market does not provide enough capacity to the TSO: Missing capacity on the balancing market forces the TSO to apply congestion management. With the energy transition, generator capacity is moving from the transmission level towards the distribution grid. In Germany, more than 95% of all renewable generators are connected to the distribution grid. Similar tendencies can be observed in Spain and other European countries. Furthermore, as we have discussed in this post, generation becomes more distributed. In addition, most generators on the distribution grid don’t have access or an incentive to participate in the national balancing market. Together, the increasing decentralized generation from RES and the lack of participation of these resources in the national balancing market are the key drivers for the current discussion about the increasing need for coordination between the distribution and transmission network operators.

Why we need more coordination between the TSO & the DSO

As it is, the TSO can make use of flexibility via the balancing market to react to unpredicted deviations from the production or consumption schedule. To do so, power plants and some aggregated distributed generators (mainly bio-gas CHPs with backup-diesel-generators) offer their capacity on the balancing market. The further the energy transition proceeds, the lower the capacity of power plants that are connected to the transmission grid directly, as the total installed capacity of conventional power plants decreases. Therefore, it is necessary that distributed generation participates in the balancing market. Some larger battery storages and virtual power plants are already participating in the balancing market, but still on a proof-of-concept basis.

If we want the distributed generators to participate in the balancing market, we need to secure that the distribution grid operator has access to information about these activities to calculate this into the load projections for the distribution grid. Today, the DSO only knows that one of its grid users is active on the balancing market (via the pre-qualification process). It does not have detailed information what the grid user from its network is offering, nor does it know the TSO’s demand at this specific point in time. Vice versa, the TSO has no information about the current status quo of the distribution grid at the moment of the flexibility request via the flexibility market. In the worst case, it could happen that the TSO requests flexibility from a distributed generator or consumer that puts the distribution grid’s stability at risk (which is not the intention of the TSO). The more generators or consumers from the distribution grid become active on the balancing market, the stronger the potential impact on the grid stability on the distribution grid.

Similar developments can be anticipated in the case of congestion management. Today, most requests for feed-in management in Germany are required by the TSO to address congestion on the transmission grid. If there is no redispatch possible on the transmission grid, the TSO orders the DSOs to curtail renewable generators (in most cases wind farms) to secure system stability. In the future, it is likely that the DSO’s need for feed-in management will increase to avoid congestion on the distribution grid level. Though this has happened only occasionally up until today, it is likely that these events will increase. Again, it is possible that the TSO requests to curtail a wind farm that is currently needed on the distribution grid level to avoid congestion. Vice versa, the DSO could request the curtailment of generators that at the same time are requested by the TSO to increase production to keep the transmission grid in balance.

Potential approaches – How to organize coordination between the TSO and the DSO

The coordination between the TSO and the DSO requires a detailed discussion of different interfaces, like data management, grid codes etc. In the following we want to focus on the increasing need for coordination between the transmission and distribution grid operators when it comes to system stability.

Different concepts are discussed in this context and Gerard, Rivero & Six (2016) provide a nice overview about potential solutions. The following summary of five concepts has been adapted from their work (for details see here):

  • Centralized AS market model: in this model, the TSO operates a market for both resources connected at transmission and distribution level, without extensive involvement of the DSO. This is the closest pattern to the traditional way of doing things.
  • Local AS market model: in this approach, the DSO organizes a local market for resources connected to the DSO-grid and, after solving local grid constraints, aggregates and offers the remaining bids to the TSO.
  • Shared balancing Responsibility Model: here, balancing responsibilities are exercised separately by TSO and DSO, each on its own network. The DSO organizes a local market while respecting an exchange power schedule agreed with the TSO, while the TSO has no access to the resources connected to the distribution grid.
  • Common TSO-DSO AS Market Model: the TSO and the DSO have a common objective to decrease the cost of the resources they need. This common objective could be realized by the joint operation of a common market (centralized variant), or the dynamic integration of a local market, operated by the DSO, and a central market, operated by the TSO (decentralized variant).
  • Integrated Flexibility Market Model: in this scheme, the market is open for both regulated (TSOs, DSOs) and non-regulated market parties (BRPs, CMPs), which requires the introduction of an independent market operator to guarantee neutrality. As a consequence, the boundaries between intraday markets and ancillary services could fade away.

All these models have different strengths and weaknesses (e.g. the development of bid prices in local models, overall costs of the systems ets. ). As a result, there is much disagreement about which model to choose under which conditions. Obviously, interests and expectations vary, which make the discussion even more complex. We will keep an eye on the on-going developments and report on the process here at enerquire soon.

Like this post? Let us and others know by sharing it!