Saturday 27 February 2021

Rethinking a disrupted electricity sector

DUBAI, December 30, 2020

By Louis Strydom


The past decade has delivered profound changes in the electricity sector. These changes are creating new roles, business models, and technologies that will overhaul the electricity grid as we know it today. Electricity grids used to be predictable systems with expected daily variations and seasonal patterns. Synchronous generation (rotating equipment with a robust spinning mass) was able to cater for measured and planned load variations. This approach is being dramatically uprooted by the introduction of Variable Renewable Energy using technologies such as solar and wind (VRE).

Over the past five years, VRE has demonstrated that it is one of the most cost-effective and sustainable sources of electricity. This low-cost leadership means that in most places in the world, cheap but unpredictable electricity is now first on the grid’s merit order. Putting VRE ahead of baseload electricity changes the role of balancing in the grid. Traditionally, you only had to deal with demand-side variance. Now, you have the supply variance too, as the VRE fluctuates according to the whims of the weather. This means an exponentially growing volatility in the magnitude and sequence of balancing electrons needed to keep your grid stable. It’s like trying to play roulette at two tables at the same time. The increased importance of balancing will create a paradigm shift in the economics and operating models of the future grid.

Cheap, reliable and sustainable electricity supply remains a critical driver for the economic success of a country. Grids need to change their business models to meet this objective and cope with the paradigm shift. First, asset ownership purpose and roles, and especially the role of public ownership, will change. Second, the nature of public-private sector frameworks must adapt. Third, the changing technology landscape necessitates more flexible and agile approaches to grid management and planning. This article discusses these changes and how grids will need to evolve to meet them.

The changing environment

At low levels of VRE penetration, there is still plenty of ‘wriggle room’ left in current grid adaptability, and it becomes easy to absorb VRE. At higher VRE penetration levels – twenty per cent and above – this picture starts to change. This is the tipping point.

Currently, tipping points occur most often when low electricity demand meets a high renewable generation period. Grid operators then find themselves in a challenging situation as a system’s generation exceeds demand. The conventional baseload is typically “must-run” – think of large Combined Cycle Gas Turbines (CCGTs), coal or nuclear – therefore you can’t go below a specified minimum output. Either you have to dump your excess must-run generation at a discounted price (or even pay someone to buy it from you), or you curtail renewables. In any of these instances, there will be a trade-off – you will waste renewable electrons or money. Germany and other countries in Europe experienced this when COVID created reduced demand paired with high VRE output.

In the near term, curtailing renewables remains the preferred choice for many grids until they begin to implement more effective balancing and storage technologies. This approach satisfices and is not a sustainable long-term solution. How then do grids need to change to deliver better solutions?

How grids need to change

The satisficing approach cannot continue indefinitely. Existing generation fleets need to be replaced as they reach the end of their economic life, and in growing economies, generation assets need to be added. New assets and new generation models allow you to move from a satisficing to an optimising grid model. Optimising the grid means more VRE, less traditional baseload and robust grid balancing. It will also change your planning approach to an agile and flexible 30-year horizon, constantly redefining uses and purposes of assets in the grid. That is easier said than done.

Near-term decisions are more straightforward

To replace or add balancing assets in the next five to ten years, you generally have two best-in-class solutions. For short-duration balancing, firming and dealing with grid intermittency caused by VRE, battery energy storage solutions (BESS) are becoming the go-to option.

Unless you have ideal hydro or pumped-hydro conditions, for the medium to longer duration balancing, you need thermal balancing. The most significant shift is occurring in the type of thermal balancing asset you need. Your criteria have changed – you need modular distributed technologies that can match the fast ramping and high-volume response cycles that a VRE-dominated grid will demand.

Precision response and efficiency across a broader spectrum of utilisation levels are the best ways to match your demand and supply curves. Unwieldy assets with inefficient starts that lag the response curve exacerbate your satisficing dilemma. Naturally, I prefer Internal Combustion Engines (ICE), as the long-term economics stack up well; but open cycle gas turbines also play a significant role. Technologies that are not useful:

1.    only display good economics at high utilisation levels (capacity factors)
2.    cannot economically handle multiple stops and starts per day
3.    cannot deal with the steep ramping up and down.

Increased investment in transmission and distribution networks (T&D) is a clear near-term decision. VRE demands an excellent T&D system to wheel electrons cost-effectively from the renewable generation sites to the consumption points. Upgrades in T&D infrastructure also enable improved balancing of the grid and enables pairing distinct T&D networks to create a larger pool of electrons to pair excesses with deficits. Further, the traditional unidirectional electricity cycle – generate, transmit and distribute, and consume is at an end. Consumers have increasingly become producers of electricity (prosumers). An essential requirement to accommodate this two-way electron flow is upgrading the T&D infrastructure. Decentralised grid-scale BESS will perform a critical role to enhance this.

Long-term decisions will reframe grid models

This section focuses on the functions a grid needs to reframe to be effective in a high-VRE environment, regulations and public policy are not considered. The policies will either act as enablers or barriers to the successful delivery of the functions but do not alter the need to reframe these functions. The difference is that costs and functions are more transparent and articulated in fully merchant systems, whereas single buyer structures tend to assume these costs more implicitly. These implicit or hidden costs sometimes make it more difficult to quantify the reframed benefits.

How to rethink value in the grid

Reframe #1 – Rethink thermal assets

Public-Private Partnerships (PPPs) are a favoured model for the public sector to engage the private sector in many countries. Grids often procure electricity using them, and increasingly, transmission operations also use PPP models. PPPs offer the following options for the public partner to select from to engage with the private sector: project development, project execution, expertise and training, operations and maintenance, and finance. The advantage of designing good PPPs is that you can identify which skills/risks the private sector can manage and which must remain with the public sector. Independent Power Producers (IPPs) are only one form of PPPs.

So what has changed? The nature of asset use, especially for thermal assets, have changed. Many thermal IPPs are facing project stress as both public and private sector partners face dramatic changes in the generation mix the grid requires. There are many contributing factors at a country-specific level, but a key driver has been the introduction of VRE. VRE is now first to dispatch and thus cannibalises the predictable generation profiles of many thermal IPPs business models.

Most IPPs have a capacity fee which is fixed throughout the lifetime of the PPA and intended to provide security for the lenders, as well as a minimum internal rate of return (IRR) for the private sector. This capacity fee structure works well if you have high utilisation of an asset, but having a high equity return requirement at the tail-end of an underutilised asset return profile, such as caused by VRE penetration, is a terrible business model. Often these situations also give rise to a political backlash

For thermal assets, the ownership model needs to change. For now, they are an essential asset in the grid, not only for the balancing function but also as a strategic public asset to supply guaranteed generation capacity. In the grid, you must have a backup to your VRE generation. Without that, you have no protection against Black Swan events such as the eruption of Tambora did, which could wipe out your solar supply for months. But procuring such an “insurance asset” should not be done under a traditional IPP model.

The public sector is better off taking the asset on their books, as the asset is essential but will likely face decreased use over its economic life. Besides, you can easily shave a few cents per kilowatt-hour off the Levelised Cost of Electricity (LCOE) by paying the asset off through the public sector payments. On large scale plants, that can translate into nominal savings of a few 100 million US Dollars over the asset’s life. One of the objections I come across when having this discussion is that many people believe assets perform better when the private sector has ‘skin in the game’.  The private sector can self-fund plants turnkey up to the commercial operations date, plus a safety period. Likewise, to ensure operational integrity, the private sector can be procured to bid for the operational use or performance of the asset, with sufficient vested interest to protect the public owner. Such mitigants can attract private sector skills and risk-taking yet ensure correct asset allocation to the public sector.

Reframe #2 – Rethink Public-Private Co-creation

One of the more significant challenges remains to find viable economic models for BESS solutions. BESS commercial models work best when you have at least one marginally profitable service, and you then create a value stack of additional (i.e. ancillary services) which provide an upside to developers.

The challenge is that many grids do not explicitly articulate the value of ancillary services, nor is the regulatory and policy environment yet conducive. In the past, there was a minimal need to do this. Still, by introducing VRE into the system, their variable nature now opens up the need for complimentary BESS services to firm, balance and occasionally even shift the VRE electrons.

BESS vendors are best placed to identify storage opportunities that can benefit the grid. To do this, they need access to detailed grid data. This is more likely in merchant grids where many system costs are transparent, making the development of commercial models somewhat easier. But even in Single-Buyer markets creating such opportunities are possible. The key is to rethink the public and private sector engagement.

You are much less likely to have an effective BESS solution on a specified, predefined output-based tender than on a value-based tender. Value-based tenders do not predefine solutions; they allow private sector open access to data to create the best value propositions possible. They also provide an effective way to manage technology and performance risk by placing them with the private sector. Thus BESS vendors can identify and optimise the value stack effectively and create tailored solutions without the public sector facing unknown risks. That is a significant shift for the public sector and will also require proactive regulatory and policy support.

Reframe #3 – Rethink technology – use, disruption, definition and planning

Never has it been so difficult to steer the long-term development of grids. Not only do utilities have to deal with fundamental changes to their business models, but also face unintended impacts of new technologies, such as the intermittency of VRE. Rapid progress on the technology s-curve of VRE and BESS is causing first-mover disadvantages for grids as performance keeps increasing and costs keep dropping.

You need to reframe what and how you procure to redefine the technology use case. One approach is to set a tolerance band for variability, where you expect the generator to address intermittency in say the zero to fifteen-minute variance band, and you absorb the variance above that. This approach will motivate the private sector to move from a VRE only offering to VRE + BESS. This strategy provides you with a more stable supply, reducing your direct balancing burden. Another strategy, to avoid the balancing and integration burden on the grid, is to procure a fully balanced always-on power block of say 500 MW and judge the quality of the block based on sustainability and reliability criteria. This approach completely offsets the balancing burden.

You will be affected by technological disruption. Longer horizon technological changes centre around the conversion and storage of excess VRE either as molecules or electrons. Long-duration storage technology development is continuing apace. Pumped hydro is already well established. However, most technologies have not yet reached unsubsidised full commercial status. Similarly, green fuels present a potential solution to store VRE electrons as molecules. The produced hydrogen or synthetic gas can easily be stored, then later used for power generation in conventional generation units (robust rotating mass=inertia), or as a transport fuel. Every discussion around new or existing thermal generation assets should be cognisant of how capable or upgradeable they are to generate electricity using green fuels. Both long-duration storage and green fuels pose the potential to solve the non-synchronous problem of VRE, and it will disrupt the current grid business models further

You will be facing redefinition of what it means to produce power on the grid. More efficient movement of electrons in the grid, and more sophisticated blockchain and AI technology will also enable more significant innovation in demand-side management and the development of virtual power plants (VPPs). As these technologies reach full commercialisation towards the latter part of the decade, they will play a significant role in the balancing and provisioning of electrons to the grid. These virtual aggregations of producers, prosumers and consumers will change the roles and definitions of various players in the grid.

In short, the one constant the grid will face is continuous change. This change means a substantial rethink of the planning role in the grid. A master plan with one defined solution does not cater to the substantive changes that are coming. Best-in-class planning will need to be probabilistic, analysing different scenarios and selecting a preferred development path. The preferred path will need to be continuously reassessed, adapted and recommunicated to all participants in the grid. Planning departments must continually test the validity of their planning assumptions on generation (engines, storage, turbines, VRE, etc.), T&D, and disruptive technologies to ensure they have the most current information to map the best possible grid scenarios. Further, to navigate these changes, transparent communication in the grid will be essential. Open access to data in grids will dictate which grids survive and which thrive as it will enable new technologies to co-create the agile value propositions that cannot be defined without detailed grid data analysis.


To realise the vision of a decarbonised grid that delivers electrons reliably and cost-effectively will require a rethink of how you approach the management and operation of the grid. A successful grid will rethink:

1.    asset classes and ownership and operating models to maximise economic and sustainable value
2.    what the purpose of public-private engagement is, and which functions of the partnership best reside with whom
3.    how to manage technological change and the evolving nature of grid planning.

About the author

Louis Strydom is Director - Project Development, Middle East, Wärtsilä, a leader in energy transition



calendarCalendar of Events