This writing reflects the author’s professional views and not necessarily those of Faraday Grid Limited.
Rapid acceleration of changes in the energy sphere renders preparations for their implicit challenges difficult. Bloomberg’s recently published 2018 New Energy Outlook (NEO) highlights key market drivers and integrates insight from a variety of experts to evaluate how the energy market will evolve in the coming decades.
From a pragmatic perspective, the electricity system will have to technologically adapt to enable these trends without inflating costs or dangerously destabilising the system. Electrical Engineer, Alex Kleidaras reviews some of the claims of NEO, with regards to their implications and requirements for their feasibility.
"50 by 50" - Intermittent Renewable Energy Resources (RES) are expected to generate 50% of total annual generation by 2050 .
In traditional electricity system architectures any mismatch between supply and demand has the potential to destabilise the system. Inertia of generators provides a delay (usually a few seconds) until the destabilisation occurs. Without it, small imbalances between demand and supply would initiate the protection system to trigger the disconnection of components and lead to system-wide failures. As such, inertia is essential in maintaining the stability of power systems.
Inertia is inherently provided by the spinning rotors of conventional generators, but not by solar or wind generators (in some cases, they can provide limited artificial inertia at the expense of efficiency). In the German grid, where intermittent renewable generation was over 20% in 2012, there were several hours of them providing ~50% of the total demand. At the same time, the regional inertia (H) dropped significantly due to the temporary lack of conventional generators in use. An estimated drop from H≃6 seconds to H<3 seconds can have detrimental effects to the stability of a well-interconnected system . Australia also encountered serious issues due to systemic instabilities, resulting from a high penetration of renewables , , .
Intermittent RES covering 50% of annual generation could even see instances of power system operation where renewable generation surges to the extent where momentarily no conventional generation is left to provide inertia. To maintain the necessary stability within such a grid, traditional spinning reserves providing inertia or an equivalent service will be needed.
Spinning reserves fulfil this function nowadays. Without an equivalent alternative, these would be required in much larger quantities (multiple times the current ones) to cope with the hours of high intermittent RES generation. “Frequency responsive” spinning reserve, which responds within seconds (usually less than 10) is the most crucial for stability, with operational reserve being the second most important (response within a few minutes) . In a system with very low inertia, even faster response times will be required.
Spinning reserves are designed to be able to ramp up or down their output in short intervals and thus run well below rated (optimal) value. As such, they come with two major drawbacks:
- They operate in an inefficient state, meaning they need more fuel resulting in higher cost per MWh.
- Due to this inefficient operation, they also produce excess CO2 emissions per MWh.
The stability of a national system under very high intermittent RES penetration has yet to be encountered and remains unknown.
"Cheap renewable energy" - Energy produced by RES is relatively cheap and is projected to become even more so, over the coming years .
Advancements in solar and wind generator technologies may indeed enable RES to be a cheaper fuel for energy production. But the cost of production is only a portion of the final price of electricity to the consumer. The cost of transmission and distribution of electricity safely under high penetration of intermittent RES is expected to increase within the current framework.
Most types of renewable generators cannot provide essential services or have limited capabilities compared to conventional generators. To ensure secure and reliable grid operation beyond generation, ancillary services are deployed. Using the current semi-passive grid with low flexibility, additional ancillary services will be needed with higher proportions of RES, adding to the power system’s cost. Moreover, extra capacity will be required to ensure that the system can reliably meet peak demand. This will ultimately increase the end price of electricity.
Estimations of some of the costs associated with high intermittent RES penetration in the UK are summed up in Table 1. . These would all feed into the end price of electricity, meaning that predicted “cheaper renewables” will ultimately result in higher electricity bills.
Reactive Power regulation will also be crucial. Besides Active Power, most loads consume Reactive Power as well. It is supplied (or absorbed in a few cases) by generators and/or special equipment within the network. Reactive Power is important for power systems’ dynamics and transients, even more so in the case of high intermittent RES penetration .
Intermittent RES have limited to no capabilities of regulating Reactive Power. Replacing half of the conventional generation with them, greatly reduces the systems capability to regulate Reactive Power and thus extra equipment will be required to provide this. Currently available technologies include Flexible AC transmission system devices (FACTS) and Power Electronics devices, which are also costly.
"Coal will shrink to just 11%" & "Gas consumption for power generation increases only modestly out to 2050" 
In addition to the stability requirements outlined above, RES generation’s intermittence will have to be covered by other units. Traditionally, this would entail either conventional generation or energy storage.
Gas-fired facilities are currently used as peak generators and help to balance variable RES due to their ramping capability. Requirements for such facilities is likely to increase as RES penetration does. Similar to spinning reserves, these will not always operate at rated (optimal) point, meaning reduced efficiency and higher costs per MWh.
Partial utilization (only for peaks and intermittence balance) will increase the power system’s overall cost.
"Electric vehicles add around 3,461TWh of new electricity demand globally by 2050, equal to 9% of total demand."
Electricity demand is expected to increase over the next years as the population grows and electricity becomes more widely accessible in developing countries. In addition, new load types are gradually introduced; the power system’s major challenges will include accommodating the needs of electric vehicles (EVs) and those of heat pumps (HPs). Only the introduction of EVs will result in an increased cost of generation of more than 9%.
High penetration of EVs can put a significant strain on the grid. Without appropriate adjustments, these may cause several problems, such as voltage violations and congestion on a local level. Such problems already arose in the Norwegian grid , where parts of the grid have been reaching their operational limits due to charging electric vehicles.
Energy storage has been proposed as an alternative solution. Several energy storage technologies exist which convert electrical energy into some other form and back to electrical energy. They can provide non-spinning reserves and cover times of generation intermittence   . The costs associated with the different storage technologies however should be considered to assess their sustainability from an economic perspective. Specifically, CAPEX, OPEX, and efficiency of the different types of energy storage are relevant factors in network planning.
Efficiency is used commonly under the term ‘roundtrip efficiency’ in power systems, which is the efficiency to convert from electrical energy and back. Roundtrip efficiency of storage technologies is summarised in Table 2. The significance of roundtrip efficiency can be understood when two storage systems with 66.6% & 80.0% roundtrip efficiency are considered. In order to supply 100MWh to consumers, 150MWh & 125MWh have to be produced, meaning an increase in energy demand of 50% & 25% respectively. Subsequently, the cost of energy increases as well, besides the already existing CAPEX and OPEX costs of energy storage . In reality, most types of storage technologies experience some form of leakage, further reducing the stated efficiency .
Some types of storage technologies also degrade with use. One such example is Li-on batteries, which come with a known lifespan, expressed in cycles of operation. At the end of their lifespan their capacity is expected around 70% of the original . Regular use of Li-on batteries (and other on-use degrading technologies) will greatly reduce their lifespan and consequently capacity and value. Using this technology regularly in a large-scale network means that the storage has to be oversized and replaced every few years, something to be considered in power systems planning (usually a horizon of 20-30 years or even more). In the future, technological advancements in storage may enable wider adoption of this solution.
An alternative to energy storage is the deployment of Demand Response (DR) technologies. Demand response is a change in the power consumption of a consumer in response to some event or signal (price related or not), to balance/stabilise the system or better match the supply. Demand response is a relatively old technology which has been getting a lot of attention lately, it is still under development and has high potential.
This could be used to provide balancing services to the grid, whilst also being able to accommodate more loads, such as EVs . DR doesn’t come with the drawbacks of energy storage; roundtrip efficiency, degradation, high CAPEX and OPEX costs, but has its challenges, which stem from the need to coordinate numerous flexible loads .
The current power systems' ability to accommodate intermittent renewable energy is limited by physical and economic constraints. Most available technologies that can stretch its limits are expensive. Only relying on these would dramatically increase the end price of electricity. Moreover, they each add a level of complexity to the network, further limiting their wide-scale adoption within the current framework.
To keep up with modern requirements, Faraday advocates for a rethink of the electricity system. A system design with inherent resilience would not only enable significantly higher levels of intermittent RES, but also provide a platform for other technologies as well – this is the idea behind the revolutionary Faraday Grid. This technology was purposefully designed to take use of existing network assets and reform them into an antifragile system, the Faraday Grid, possessing all the necessary characteristics to make it the energy platform of the future. Read more about the Faraday technology here.
*The cost of additional reserves required to balance electricity supply and demand over the timescales of seconds to hours, and the costs of capacity required to ensure that a system can reliably meet peak demand is referred to as the capacity cost.
 New Energy Outlook 2018, Bloomberg NEF (Online):  https://about.bnef.com/new-energy-outlook/
 Ulbig, A., Borsche, T. S., & Andersson, G. (2014). Impact of low rotational inertia on power system stability and operation. IFAC Proceedings Volumes, 47(3), 7290-7297.
 AEMO releases final report into SA blackout, blames wind farm settings for state-wide power failure, ABC news. (Online): http://www.abc.net.au/news/2017-03-28/wind-farm-settings-to-blame-for-sa-blackout-aemo-says/8389920
 Australia's power grid weakened by wind, solar generation, AEMC calls for plan for security, ABC news. (Online): http://www.abc.net.au/news/2017-03-24/aemc-says-australias-power-system-weakened-by-wind-solar/8381356
 Why the Australian Grid May Be on the Brink of Collapse (and It’s Not Wind’s Fault), Green Tech Media. (Online): https://www.greentechmedia.com/articles/read/why-the-australian-grid-is-on-the-brink-of-collapse-and-its-not-winds-fault#gs.YAOahNI
 Spinning Reserve, Energy Storage Association. (Online): http://energystorage.org/energy-storage/technology-applications/spinning-reserve
 B. J. Kirby, Spinning Reserve From Responsive Loads, Oak Ridge National Laboratory, March 2003.(Online): https://eta-internal-publications.lbl.gov/sites/default/files/spinning-reserves.pdf
 Heptonstall, P., Gross, R., & Steiner, F. (2017). The costs and impacts of intermittency–2016 update. UK Energy Research Centre, London Google Scholar.
 Sarkar, M. N. I., Meegahapola, L. G., & Datta, M. (2018). Reactive Power Management in Renewable Rich Power Grids: A Review of Grid-Codes, Renewable Generators, Support Devices, Control Strategies and Optimization Algorithms. IEEE Access.
 NORWEGIAN GRID STRUGGLING TO KEEP UP WITH GROWING EV FLEET, EV obsession. (Online): https://evobsession.com/norwegian-grid-struggling-keep-growing-ev-fleet/
 Making the Case for Spinning Reserve on the Grid, Renewable Energy World. (Online): https://www.renewableenergyworld.com/articles/2011/05/making-the-case-for-spinning-reserve-on-the-grid.html
 Advanced Energy Storage: What’s the Value of Frequency Regulation?, Renewable Energy World. (Online): https://www.renewableenergyworld.com/articles/2011/05/advanced-energy-storage-whats-the-value-of-frequency-regulation.html
 Round Trip Efficiency, Energymag (Online): https://energymag.net/round-trip-efficiency/
 Calculating the True Cost of Energy Storage, Renewable Energy World. (Online): https://www.renewableenergyworld.com/ugc/articles/2015/01/calculating-the-true-cost-of-energy-storage.html
 Powervault 3 vs Tesla Powerwall 2: It's All In The Chemistry, Spirit Energy. (Online): https://blog.spiritenergy.co.uk/homeowner/powervault-3-vs-powerwall-2-chemistry
 Kleidaras, A., Cosovic, M., Vukobratovic, D., & Kiprakis, A. E. (2017, September). Demand response for thermostatically controlled loads using belief propagation. In Innovative Smart Grid Technologies Conference Europe (ISGT-Europe), 2017 IEEE PES (pp. 1-6). IEEE.
 Kleidaras, A., Kiprakis, A. E., & Thompson, J. S. (2017). Human in the loop heterogeneous modelling of thermostatically controlled loads for Demand Side Management studies. Energy.