What is Grid Orchestration? Learn more
Today the Department of Energy’s Loan Programs Office released their latest “Liftoff Report” focused on innovative grid deployments. The report is well worth a deep-dive and inspired our team to debate the most exciting real-world deployments of grid modernization technology undertaken by our partner utilities – who range from 30,000-member rural cooperatives to investor-owned utilities serving millions of homes and businesses.
Which innovative deployments made the cut?
First, let’s highlight our criteria. We kept it pretty simple:
While we’ve seen many impressive achievements and innovative deployments at our partner utilities already this year, three stood out to us. To be able to speak to tangible results, we’ve anonymized these examples – but each comes from an active deployment of grid modernization technologies, including but not limited to Camus’ grid orchestration platform, by a thoughtful, forward-looking utility.
For distribution operators, the era of relying on static load profiles or hours-delayed AMI data to assess what’s happening at the edges of the grid is rapidly coming to an end. The impacts of distributed solar, battery storage, and electric vehicle charging make the use of standard residential or commercial load profiles inaccurate and ineffective. Yet it’s still the standard input for many utilities’ operational systems, including the Advanced Distribution Management Systems (ADMS).
What is an ADMS? It’s a software platform that serves as the “brain” for conventional utility operations, integrating several subsystems. Learn more about ADMS and the “alphabet soup” of electric utility software systems in this blog.
To gain a more accurate view of operating conditions—inclusive of DER impacts—we recently helped one utility partner scale up high-fidelity, meter-level forecasting across its entire service territory, providing a day-ahead forecast for net and gross loading at >1 million meters.
The approach we used combines NOAA High Resolution Rapid Refresh (HRRR) weather forecasts with meter and solar interconnection data and uses machine learning algorithms to rapidly and cost effectively generate day-ahead forecasts for every meter. (Learn more about the technology behind meter-level forecasting in this blog post).
These meter-level forecasts are then aggregated, using the utility’s existing connectivity model, to forecast loading at every service transformer, line conductor, feeder, and other major point on the distribution network.
This approach addresses the common delay in receiving data from advanced metering infrastructure (AMI) and uncovers demand hidden by distributed generation, supporting safe restoration from planned and unplanned outages.
The increased granularity of forecasts also opens the door for more efficient operation of the distribution network. By funneling net and gross demand forecasts at aggregation points into the utility’s ADMS, the utility can enable faster solving of power flow models and increase the effectiveness of voltage optimization and outage restoration workflows, especially on circuits with high penetration of distributed generation.
Our team found this deployment worthy of a top spot because of its use of existing utility data (AMI, interconnection records, connectivity model), public weather forecasts and commercially-ready tools (ML-based forecasting, cloud computing) in a no-regrets approach to improve visibility and ADMS functionality.
Ultimately, we believe this data-driven view into conditions at the grid edge will soon become the norm for utilities of all sizes, taking advantage of the power + cost effectiveness of cloud computing and the accuracy of ML-backed forecasting.
The growth of electric vehicle (EV) charging demand, especially for EV fleets, is a hot topic for utility leaders and system planners. Preparing for load growth—and changing load profiles—is a core responsibility for distribution utilities; delays in interconnecting large EV loads are resulting in unhappy community members and slower electrification. As a result, utilities are keen to find ways to better prepare for EV load growth, proactively investing in their network without overspending.
Recently a Camus utility partner used existing data to conduct an end-to-end analysis of EV impacts on their system, simulating load impacts out a decade into the future.
To do so, Camus first deployed machine learning algorithms to identify where electric vehicles are charging, then used its connectivity model to map those charging sessions to upline equipment. With a fine-grained view of adoption rates within each feeder, the utility was then able to apply growth assumptions on top of the observed localized adoption rates to better estimate the EV charging penetration for each feeder.
The result of the analysis was a feeder-by-feeder evaluation of EV load impacts for the next decade across multiple adoption scenarios, identifying potential hotspots and estimating the costs of subsequent infrastructure upgrades.
We found this deployment to be worthy of recognition because this analysis forms the foundation for a pragmatic and proactive capital planning process, enabling the utility to bring data to regulators and ultimately pursue a more equitable and capital-efficient approach to preparing for EV charging demand.
This innovative deployment leverages data that is readily available at most utilities and provides an example path forward that other utilities can (and we believe, should) follow.
The final deployment on our list breaks our criteria slightly – as the deployment is actively underway and results aren’t yet available. However, we feel its potential to solve a fast-growing pain point for utilities warrants a spot on this list.
For decades, utilities have deployed DER-based “non-wires alternative” pilots, but the era of electrification brings with it a more compelling case than ever before for managing the costs of infrastructure buildout. Many utilities are grappling with the question: can we trust DERs to reliably defer the need for new infrastructure?
A Camus partner utility is deploying an innovative approach to answering this question – bringing together several commercially-available technologies to do so. To narrow the scope, the focus for this deployment is on testing the ability of managed EV charging loads to reliably alleviate the need for 1) service transformer upgrades and 2) line reconductoring.
This deployment includes several technology categories – which are used in different combinations to test different approaches:
A particularly innovative aspect of this deployment is the testing of both different levels of data availability and different approaches to EV managed charging. An explicit goal of this deployment is to identify the granularity, frequency, and latency of loading data required to know when and where to shift EV charging loads to avoid overloading and voltage deviations. Also included will be an evaluation of the efficacy of EVSE-only, EV telematics-only, and EVSE + EV telematics approaches at managing network constraints.
While it’s too soon to applaud this deployment for paving the way for other utilities, we’re excited to see (and share) the learnings later this year.
Each of the above deployments takes an especially innovative angle on applying existing technology for near-term utility needs. However, we’d be remiss if we didn’t briefly touch on a topic that has been top-of-mind for utility conversations recently: flexible interconnection.
Interconnecting large loads and distributed generation into the existing distribution system is quickly becoming a major challenge for utilities across the country. While many utilities have significant unused capacity on their networks, the tendency of electrification loads to show up in concentrated, spiky load pockets is increasing the frequency of interconnection requests resulting in the need for new infrastructure upgrades. Similarly, increasing distributed generation penetration is resulting in areas where backfeed is a pressing concern during minimum loading periods – like a sunny spring day.
Flexible interconnection is a hot topic because it provides a way for utilities to approve interconnection requests while incorporating operational constraints. California’s Limited Generation Profile is a good example of one such approach. We’ll dive into flexible interconnection more in a future piece, but if you’re curious – we’ve enjoyed these quick reads from EPRI and EDF on the topic.
As utilities take action to support sustained load growth from electrification, we believe that utilities of all types will pursue opportunities to serve more load with existing infrastructure – leveraging the nascent flexibility of distributed energy resources. Rather than simply procuring a DERMS, more and more utilities will invest in ways to leverage existing data and software systems for real-time visibility, operational forecasting, and true orchestration of DERs to manage network capacity. We’re eager to see and support the next wave of innovative grid deployments at utilities across the country.
Finally, we want to give a special thanks to the authors of the Liftoff Report for engaging stakeholders throughout the industry on the path forward for managing and investing in our grids. If you’ve made it this far without checking out the report…go take a look!
