Electricity is an increasingly complex industry in the midst of transition to renewables and decarbonization. Using my 25 years’ experience as an engineer, policy analyst, and academic, I help my consulting clients think through their toughest technical challenges and formulate their best business strategies.
Energiewende, German for “energy transition,” was the theme of the “Texas-Germany Bilateral Dialogue on Challenges and Opportunities in the Electricity Market” conference held in Austin in late February. The conference was organized by the German American Chambers of Commerce and supported by the German Federal Ministry for Economic Affairs and Energy, in cooperation with ERCOT.
As is well-known, Germany has invested significantly in low-carbon technologies, and their efforts have arguably led to reduction in the cost of production of solar photovoltaics worldwide. Talking at the conference about the transition to a low-carbon energy system were representatives from the German federal government and ERCOT, as well as several companies and consultancies.
I asked about the levels of carbon dioxide emissions in electricity production in Germany and ERCOT. Somewhat remarkably, given the emphasis on carbon dioxide in the Energiewende, there was no direct information about carbon dioxide emissions in any of the presentations. There was information about the shares of renewables, but different speakers had different numbers relating to the total electrical energy and the contributions from various resources.
So I did some calculations. I found about 0.52 metric tons of carbon dioxide emissions per MWh of electricity production in ERCOT and between 0.33 and 0.43 metric tons of emissions per MWh in Germany — assuming 1 metric ton of emissions per MWh of coal generation, 0.5 metric tons per MWh of gas generation, and negligible emissions from renewables and nuclear, and using the energy contributions presented by Falk Bomeke, PhD (German Federal Ministry for Economic Affairs and Energy), Bill Magness (President and CEO, ERCOT) and Arne Genz (German Federal Ministry for Economic Affairs and Energy).
ERCOT per capita consumption of electricity is about double that of Germany, and the emissions calculations indicate that electricity generation in ERCOT emits more carbon dioxide per MWh, resulting in significantly more emissions of carbon dioxide per capita in ERCOT compared to Germany. Clearly, Germany has worked toward reducing its carbon dioxide emissions and has done so without an abundant endowment of low carbon resources.
In contrast, Texas has an astonishing endowment of natural gas and renewables. Although cities such as Austin and San Antonio have set targets for renewable integration, Texas policy has, largely speaking, been indifferent to carbon dioxide emissions. Imagine how much lower emissions would be if Texas policy was re-oriented toward a low carbon future.
Electricity markets have been restructured in many countries around the globe. There is a variety of different designs, and the differences can significantly affect our ability to handle new challenges, such as integrating high levels of renewables. What are the key differences between the US and EU electricity markets? I was able to pursue that question in depth during my research sabbatical at the Florence School of Regulation (FSR) last fall. It was one of the issues addressed in an online debate I had with Daniel Dobbeni, founding president of the European Network of Transmission System Operators (ENTSO-E). Below you can watch the full debate (55:12), hosted by FSR director Jean-Michel Glachant.
The definition of seams. Seams, we agreed, include the technical transmission limitations between regions (for example, between countries in Europe and between ISOs in the US) as well as the regulatory and market differences between these regions. Mr. Dobbeni observed that the long-life of electricity system assets together with the history of development in European countries have played an important part in forming these seams, but that they are evolving with the advent of system changes, including renewables. He advocated for removing the seams, and I agreed.
I argued for a consistent architecture to be applied across as large an area as possible, observing that such consistency was perhaps even more important than, for example, whether or not the market design was nodal (as in the US) or zonal (as in the EU). I mentioned that there were still significant seams at the day-ahead level in the US, particularly in the west, and to a lesser extent in the east, where there are several large ISOs with seams between them. There still remain significant technical seams due to limited transmission between the west, the ERCOT part of Texas, and the east. In the EU, the EU Pan-European Hybrid Electricity Market (EUPHEMIA) has removed seams due to market differences in the day-ahead level across many countries through so-called “market coupling” between the regions.
Mr. Dobbeni observed that seams should also eventually be removed in the intraday markets, which are in place in Europe but not the US, and that he was concerned about the technical difficulty of market coupling in balancing markets. I discussed what I understand is the fundamental philosophical difference between US markets and EU markets: in the US, the real-time market is the “final” market; in the EU, the day-ahead market plays most of this role, with the so-called balancing market at least historically being more akin to the deployment of ancillary services in a US-style market. Moreover, I commented on the need to reach geographical scale to enable real-time management of congestion issues such as loop flow.
Mr. Dobbeni emphasized that congestion management was being complicated by the increase in renewables. In addition, he observed that congestion management across borders was particularly complicated and that US-style ISOs that spanned borders were able to consider overall issues in a way that was difficult for the EU-style markets at the country level. He advocated for the enlargement of regions beyond member states, enabling balancing beyond individual states, with which I definitely agreed!
US RTOs and some larger European countries, I observed, are likely at a large enough geographical scale to effectively balance renewables. We agreed that reducing the effect of technical and regulatory borders between regions was desirable to help with balancing renewables, but that it might be difficult to imagine, for example, amalgamating PJM and MISO in the US for various political reasons. Analogous difficulties apply in the EU.
In response to a question from a listener, Mr. Dobbeni mentioned the organizational differences between the EU and US: in the former, the TSO owns assets and operates the market; in the latter, there is a separation of ownership of transmission and operation of the market.
A question was posed about interconnectors, and I responded that building transmission across borders in the US was a challenge, and that this posed particular difficulties in building transmission between renewable-rich regions and population centers in different states. In the EU, security of supply is particularly important in the context of interconnectors between countries, whereas this is less pressing in the US.
We also recognized that energy prices are not bringing forth new capacity, and that prices are more uncertain than in the past. Consequently, even though we both were skeptical about capacity markets, there is evidently a problem. Mr. Dobbeni observed that, while there was overall a very large generation capacity in the EU, there are regional variations in capacity and many uncertainties in the long term. Restructuring in the US, I added, had focused on the generation-side. Now we need more participation by the demand-side in the market as part of the solution to the market providing the right amount of capacity. I also emphasized that the next five years will be the test of the ERCOT energy-only market.
Hurricane Maria has caused huge damage in Puerto Rico, particularly to infrastructure such as the electricity system. My sincerest sympathies go to everyone there, both in PR and in other regions. As my previous work on electricity network interdiction suggests, repair of electricity networks can depend significantly on the long lead-times to order and build extra-high voltage and high voltage transformers. As Puerto Ricans begin to restore services such as electricity, an issue that should be considered carefully is the desired end-point for their replacement electricity infrastructure and whether they should effectively rebuild their previous network or build according to a new design.
Most expansion of transmission networks, and most repair situations, involves adding or replacing equipment in an existing network. This significantly constrains the sort of solutions that can be accomplished.
However, PR is faced with a rather different problem. Although I am not personally familiar with the full extent of damage, the reports in the press suggest significant destruction of most of the network. Repair back to the state prior to the hurricane may involve rebuilding essentially everything. Under such circumstances, and given that future hurricanes may be at least as destructive to a conventional electricity system, it is prudent to step back and consider alternatives.
As an example of an alternative, perhaps a more distributed structure that plans for distributed renewables would be a better approach. Existing electric distribution networks are typically limited in the amount of distributed generation they can integrate. In the mainland US at least, the limits are typically not due to the distribution line capacity itself, but to things like “protection schemes,” typically using fuses, that were designed with the assumption of one-way flow toward consumers. In an existing system, upgrading to allow for net flow from the distribution system into the transmission system can require significant incremental investment to replace protection systems. For a system being fully built from scratch, however, it may be possible to incorporate more flexible protection systems from the start.
This and other issues should be considered carefully before large amounts of money are spent in PR on rebuilding a system according to a design that has already been shown to be vulnerable to the next hurricane.
Despite the end-of-school-year mania, I managed to get away to the 2017 IEEE Innovative Smart Grid Technologies conference in Washington, DC, in late April, to talk about the Smart Grid grad course that I was wrapping up at UT. I participated in a panel, “Innovations in Smart Grid Education,” chaired by Dr. Kenneth Lutz of the University of Delaware, with participants from MIT, the University of Illinois at Urbana-Champain, Wichita State University, and Clemson University.
I talked about the Smart Grid grad course I taught at UT this semester, making the point that “smart grid” discussions in practice are often focused on the distribution system and end-use, despite typical definitions in the literature being more general. I took an expansive definition in this class, including transmission and generation, for example, which also allowed me to invite colleagues from ERCOT and Oncor to participate.
Why do I use an expansive definition in my pedagogy?
Because the phrase “smart grid” implies that the existing grid is stupid. In fact, for many years in North America and elsewhere, operation of the transmission grid has been incredibly sophisticated — far more sophisticated than any other infrastructure system I’m aware of.
When we focus only on making the distribution grid smart, we risk throwing the baby out with the bathwater, by not building on the existing smarts in the transmission system.
In terms of pedagogy, this means students need to be aware of the entire grid, both smart and not-so-smart, in order to avoid a skewed perspective on the electricity system. As we look toward solving problems such as integrating high levels of distributed solar PV, we need to remember that the existing transmission and generation system provides the foundational infrastructure.
Highlights of the course include an overview of architecture of the smart grid, the generation and transmission system, distribution systems, and end-use. The strongest common theme: we are all searching for a good textbook!
Ross Baldick PhD provides strategic consulting to the electricity industry. Professor of Electrical and Computer Engineering at The University of Texas, he is the author of "Applied Optimization: Formulation and Algorithms for Engineering Systems."