by Christopher Clack, Aditya Choukulkar, Brianna Coté, and Sarah McKee, VCE –
Renewable technologies are lower cost than fossil thermal generation on a levelised cost basis, but their variability creates new and unique constraints and opportunities for the electricity system of the next several decades. Superimposed on the changing structure of the electricity system is a damaged climate which will continue to worsen as mankind continues to emit greenhouse gas (GHG) pollution into the atmosphere.
The US electricity system is the second largest in the world (China has the largest). In 2018 it served approximately 150-million customers with over 3859 TWh of electricity from over 1190 GW of generating capacity, routed through 760 000 km of transmission lines (over 69 kV), 55 000 substations and 10-million km of distribution lines (under 69 kV).
This is the executive summary of the study – click here to download the full study
By the end of 2019, there was 86 000 MW of renewable capacity awaiting construction across the US and each year that number continues to grow. The carbon dioxide (CO2) emissions from electricity generation across the US reached an estimated 1659-million metric tons (mmT) in 2019, accounting for approximately 32% of the total US energy-related CO2 emissions 5130 mmT.
The present study demonstrates, quantifies, and evaluates the potential value that distributed energy resources (DERs) could provide to the electricity system, while considering many facets of their inclusion into a sophisticated grid modelling tool.
The weather-informed energy systems: for design, operations and markets planning (WIS:dom-P) optimisation software tool was used for the present study. A detailed technical document of the WIS:dom-P software can be found online.
The modelling software is a combined capacity expansion and production cost model which allows for simultaneous 3-km, 5-minute dispatch and power flow along with multi-decade resource selection. It includes detailed representations of fossil generation, variable resources, storage, transmission and DERs.
It also contains policies, mandates, and localised data, as well as engineering parameters and constraints of the electricity system and its components. Some novel features include highly granular weather inputs over the whole US, climate change-induced changes to energy infrastructure, land use and siting constraints, dynamic transmission line ratings, electrification, and novel fuel production endogenously, and detailed storage dispatch algorithms.
The distribution grid is where most customers connect with the electricity system at large. However, traditional modelling tools ignore its existence almost entirely. Many assume pre-decided buildout rates of distributed solar PV (DPV), energy efficiency (EE), demand-side management (DSM), demand response (DR), and distributed storage (DS).
As the electricity system continues to evolve, customers are demanding more local resources. This creates a problem because the providers of electricity (across utility service territories and RTOs) do not possess integrated modelling tools that reveal the opportunities and costs of changing distributed generation and demand as a decision variable. The opportunities could include reduced utility-scale capacity and generation, high-voltage transmission, distribution infrastructure deferments, utility-observed peak load reduction, and increased utility-observed load factors. The costs could be more distribution infrastructure, more high-voltage transmission, increased DER buildout, and utility-scale back-up capacity and generation to cover the DER buildout.
Vibrant Clean Energy, LLC (VCE) augmented the WIS:dom-P software to improve its representation and computations of the distribution-utility interface. The augmentations enabled a modelling framework that included the distribution grid and DERs that is tractable and akin to traditional utility planning models.
During the entire study, fifteen nationwide simulations were performed. Numerous intermediate simulations were used to determine the sensitivity of the modelling tool to changes in the augmentation created during the study. The model was initialised and aligned with historical data from 2018 and then the simulations evolved the electricity system across the contiguous United States (CONUS) from 2020 through 2050 in 5-year investment periods.
Two main questions
- Can DERs lower costs across the entire electricity system compared with alternatives, while maintaining resource adequacy, reliability, and resilience?
- Can DERs provide support and benefits for clean electricity goals across the entire electricity system?
The four scenarios simulated for the present report were:
- Business-as-usual, traditional (“BAU”): Allow economics to drive the changes in the electricity system, while including existing policies, mandates, and incentives through 2050. Deploy WIS:dom-P in a manner that mimics traditional models.
- Business-as-usual, augmented (“BAU-DER”): Allow economics to drive the changes in the electricity system, while including existing policies, mandates, and incentives through 2050. Deploy the augmented version of WIS:dom-P that includes detailed modelling of the distribution-utility (DU) interface.
- Clean electricity, traditional (“CE”): Enforce a nationwide clean energy standard (CES) that reduces emissions by 95% from 1990 levels by 2050. Deploy WIS:dom-P in a manner that mimics traditional models.
- Clean electricity, augmented (“CE-DER”): Enforce a nationwide clean energy standard (CES) that reduces emissions by 95% from 1990 levels by 2050. Deploy the augmented version of WIS:dom-P that includes detailed modelling of the distribution-utility (DU) interface.
The augmentation of the WIS:dom-P software to include distribution planning co-optimisation results in cumulative system-wide savings of $301-billion by 2050 (“BAU” vs “BAU-DER”), which rises to $473-billion when considering a clean energy standard (“CE” vs “CE-DER”). Interestingly, the “CE-DER” scenario pathway is lower cost than the “BAU” scenario to the tune of $88-billion by 2050.
Figure 1: Cumulative electricity savings.
For the clean electricity system cost savings to materialise, a small amount of additional spending occurs in the first decade, however, for “BAU-DER”, the savings accrue immediately. By 2035, the “BAU-DER” scenario has saved nearly $115-billion over the “BAU” scenario, while the “CE-DER” scenario has accumulated savings of $114-billion compared with the “CE” scenario.
Over the same period, the “CE-DER” scenario is $19-billion more expensive than the “BAU” scenario but has reduced cumulative emissions by 5112 mmT (equivalent to a cost of carbon of about $3,70/mt).
By 2050, the scenario has avoided more than 10,000 mmT compared with “BAU” (as shown in Figure 2), while saving $88-billion in costs.
Figure 2: Cumulative CO2 emissions reduction vs. BAU.
If a clean electricity mandate were imposed by 2035, rather than the modelled 2050 (and the US could deploy enough generation), the DERs would bring forward the cost savings observed by 2050 to 2035, since they enable more clean utility-scale variable generation to be deployed efficiently.
The inclusion of distribution modelling within the WIS:dom-P software drives emergent behaviour. The distribution grid seeks to minimise exposure to the utility grid while maximising its benefits of being connected by minimising system costs that includes infrastructure connecting the utility and distribution grids.
This manifests as increased load factors as experienced by the utility-scale grid, while reduced peak demand. Further, the more local resources can defer some distribution infrastructure costs. The sum of these is net system cost savings, increased jobs, more manageable installation rates, a more reliable and robust system, and more opportunities for private capital investments.
The striking result is that the cost savings come with relatively little change in the macro-scale view of installed capacities and generation stack. This is because a small change in the tails of production and demand can have amplified cost implications throughout the system. Additionally, the distribution cost augmentation facilitates economic trade-off between more resources, which improves competition and reduces costs further.
Click here to download the full study