The true cost of smart grid capabilities
- Jun 30, 2014 6:00 am GMT
- 591 views
[Editor’s note: This article is excerpted from the book "Smart Grid Hype & Reality -- A Systems Approach to Maximizing Customer Return on Utility Investment."]
The initial multi-billion dollar splurge to modernize the grid as prompted by the Smart Grid Investment Grant (SGIG) component of the American Reinvestment and Recovery Act (ARRA) economic stimulus package is coming to a close. Now regulators, consumer advocates, and customers are increasingly interested in what they are getting for their money. One way to respond to these inquiries is through classic financial analysis, in which direct economic (and non-economic) benefits from smart grid capabilities are compared to their costs. The first step in such an analysis is to determine the cost to deploy a smart grid capability.
Wired Group researchers used publicly-available SGIG data to calculate average capital costs for advanced metering infrastructure (AMI) and distribution automation (DA) projects on a per customer basis. (These efforts were part of a larger project to complete benefit-cost analyses of these capabilities in “typical” and “ideal” deployment scenarios.) Utilities that have already deployed these capabilities can use the average costs the Wired Group calculated to benchmark the cost-effectiveness of their own deployments, while utilities that are considering deployments can use average cost calculations to validate capital budget projections.
Capital costs per customer
The U.S. Congress ordered the Department of Energy (DOE) to oversee how the four billion dollars of ARRA stimulus funds earmarked for U.S. distribution grid modernization and research was spent. The DOE established the SGIG program to solicit, award, administer, and govern grant applications submitted by U.S. utilities to invest in smart grid capabilities. As part of the grant application process, the DOE required utilities to specify how much money they wanted, how much money they were willing to invest (a minimum dollar invested per dollar granted was required), and on what they were going to spend the money. Wired Group researchers mined this information to calculate the average size of AMI and DA investments on a “per customer” basis.
The researchers examined in detail the applications of utilities that were ultimately awarded grants, distinguishing between AMI and DA projects. For each project they noted the customers served and total investment, including both funds from the government grant plus the utility’s own investment. Projects for which researchers could not accurately determine (or reasonably estimate from SGIG data) any one of these data points were removed from the analysis.
Most frequently, utilities were removed from the analysis as a result of researcher inability to distinguish between the AMI and DA details of projects involving both components. As a result, utilities pursuing both AMI and DA as part of the same SGIG project are not represented in the sample. The final sample consisted of 24 AMI projects and 12 DA projects. (Project data source: U.S. Department of Energy’s smart grid website, https://www.smartgrid.gov/recovery_act/project_information.)
Researchers simply divided the dollars invested in these projects by the number of customers served by each to calculate average cost per customer. Though DA capabilities varied somewhat from project to project, most included the standard features (fault location, fault isolation, and integrated volt/VAr control). Utility projects included in the final sample are listed below.
Most DA projects were not comprehensively applied to all of a utility’s feeders, requiring an adjustment to customer counts to get a true capital cost per customer served. In these cases researchers divided each utility’s total customer count by its feeder count, obtaining an average customer count per feeder. They then multiplied by the number of feeders to which DA was added, delivering a reasonable estimate of the number of customers served by a particular DA project. The capital cost per customer Wired Group researchers calculated for AMI (Smart Meters) and DA is presented in the graphic below.
“Per customer” cost caveats
Utilities and stakeholders using this information should consider a few caveats:
- Smaller utilities and utilities with lower customer densities are likely to experience a higher cost per customer, particularly for DA (characterized by higher fixed costs than AMI).
- DA cost per customer drops as the number of feeders to which it is applied increases (utilities in the sample applied DA to 10% of feeders on average)
- Ongoing O&M cost increases (labor, parts, vendor fees, software licenses) are not included in these figures. (Wired Group researchers estimate the present value of such increases at 27% above and beyond the figures presented.)
With these caveats in mind, utilities and stakeholders can use average cost data to evaluate the cost of specific deployments and the projected costs of proposed deployments on a per-customer basis. Understanding actual costs is a critical step to conducting benefit-cost analyses for various smart grid capabilities. Benefit-cost analyses can help utilities and stakeholders assess and increase the value delivered by recent smart grid investments as well as the likelihood a proposed smart grid project will achieve utility, customer, and community deployment goals. Such analyses are therefore a highly recommended component of pre-deployment design phases and post-deployment project evaluations.
Paul Alvarez is author of "Smart Grid Hype & Reality -- A Systems Approach to Maximizing Customer Return on Utility Investment." He credits Jonathan Falk and his team at the Wired Group for conducting the research presented here.