Distribution Grid Integration Unit Cost Database
NREL's Distribution Grid Integration Unit Cost Database contains unit cost information for different components that may be used to integrate distributed solar photovoltaics (PV) onto distribution systems.The database is focused on hardware and software costs and contains more than 335 data points from a variety of utilities, PV developers, technology vendors, and published research reports. The goal of the database is to provide a useful, curated, and transparent source of information for assessing distribution grid integration costs associated with PV. It includes information on traditional system upgrades that could be used to mitigate any impacts, as well as additional information on costs associated with advanced solutions including distributed energy management systems, communications infrastructure, and energy storage.
Examples of ways this database could be used by different parties include:
- PV developers: Developing estimates of interconnection costs for specific PV systems
- Analysts: Improving estimates of costs to integrate PV into the power system and identifying cost drivers
- Utilities: Comparing their costs to other utility costs and obtaining additional data about estimated unit cost ranges for emerging solutions that they might not already have access to.
The unit cost data are not intended to be directly compared to determine the cost-effectiveness of a given upgrade, but rather as inputs to engineering studies on specific distribution feeders. These data are also intended for general understanding and estimation of distribution grid integration costs.
Guide to the Database
This database contains unit cost information for different components that may be used to integrate distributed PV onto distribution systems. The total cost of implementing different upgrades on a given system is influenced by the number of units required. This is system specific. Additionally, the efficacy of different solutions to mitigate distribution impacts and expand the hosting capacity varies significantly, and thus unit costs should not be generally compared to determine the cost-effectiveness of different upgrade options. Instead, these unit costs should be combined with an engineering analysis of a specific system to understand which upgrades and how many units of each are required to accommodate different levels of distributed PV. Examples of how NREL is using this data to estimate total costs on different feeders is included below.
The data in this database was collected from a variety of utilities, PV developers, technology vendors, and published research reports. Where possible, we have included information on the source of each data point and relevant notes. In some cases where data provided is sensitive or proprietary, we were not able to specify the source, but provide other information that may be useful to the user (e.g. year, location where equipment was installed). NREL has carefully reviewed these sources prior to inclusion in this database.
The database includes information on the categories of upgrades listed in the table below. The specified party/parties who provided the data correspond to those groups that originated the cost estimate, not necessarily the group from which NREL received the data. For example, data pulled from PV interconnection reports obtained from PV developers is still labeled as utility-sourced data, since the utility rather than the developer came up with the cost numbers in the interconnection reports. Some high-level notes on data sources and data quality are provided here. Additional notes on specific data points are included within the database itself.
This database is focused on hardware and software costs. Soft costs (e.g. permitting, interconnection studies) and other overhead costs are not included and should be accounted for separate in any analysis. Additional engineering costs may also be incurred, depending on the scenario, and thoughtful construction of analysis cases/scenarios is always required to obtain results representative of costs that could be observed in practice.
Some data relates to technologies at an earlier stage than others. In particular, Distributed Energy Management Systems (DERMS) is an emerging technology which currently has no standard definition or functionality. Cost vary substantially depending on the project specifics, existing infrastructure (e.g. communications and enterprise systems) of the utility, the functionality included, and the number of DER being managed. This also varies by vendor, and not all vendors or DERMS functionalities are represented in our database. Finally, there may be some costs associated with DERMS that are not captured here including, for example, the cost of training utility engineers on how to operate the DERMS or any overhead costs associated with re-organization. These can be a substantial portion of cost in some scenarios, but would decrease after the first deployment with a given utility.
Even for more mature technologies, some may undergo innovation to reduce cost. Unit costs in this database are only intended to be representative estimates for a given location or vendor for the date specified.
|Category||Primary Data Sources||Data Quality Notes|
|Voltage regulators||Utilities: publicly available utility cost guides, SGIP reports||Six data sources with some missing data on regulator current ratings.|
|Capacitor banks||Utilities: publicly available cost data from utilities, PV interconnection reports. Range of hardware only costs from a technology vendor are also available.||Nine data sources with some missing data on capacitor specifications.|
|Transformers||Utilities: publicly available utility cost guides, project reports, and PV interconnection reports||10 data sources with a large number of data points.|
|Reclosers and relays||Utilities: publicly available utility cost guides and PV interconnection reports||13 data sources with a large number of data points for a variety of recloser/switch types.|
|Substation Protection Upgrades||Utilities: PV interconnection reports||Six data sources with a large number of data points on 3V0 and DTT.|
|Phase balancing||Utilities: PV interconnection reports||Only two data points|
|Voltage regulating device control modifications||Utilities: publicly available utility cost guides and PV interconnection reports; some data on controller hardware only from technology vendors||27 data points from a variety of geographic locations in the United States.|
|Removal and relocation of existing voltage regulation equipment||
Utilities: publicly available utility cost guides and PV interconnection reports
|Three data sources.|
Utilities: publicly available utility cost guides and PV interconnection reports
|18 data sources.|
Lazard and Greentech Media
|Two data sources. Includes total installed cost data for a range of storage technologies and system types.|
|Communications equipment and sensors||Technology vendors, utilities: publicly available utility cost guides, EPRI||Includes unit costs for different communications components. Infrastructure costs will depend significantly on the type of system used (e.g. fiber optic, wireless, radio) and the existing utility communication infrastructure.|
|SVC, STATCOM, & Related||Technology vendors, utilities: PV interconnection reports||11 data points.|
|Advanced metering infrastructure (AMI)||EPRI, the Edison Foundation||Two data sources.|
|DERMS and Demand Control||Technology vendors, utilities: PV interconnection reports||Most data is on DERMs or SCADA.|
|Telemetry and SCADA||New tab. Previously under ‘SCADA, DMS, DERMS’ or ‘Communications and Sensing.’|
|Service and metering||Utilities: publicly available utility cost guides and PV interconnection reports||6 data sources. Data on specifications (e.g. voltage of connection) is missing for some sources.|
|O&M and component lifetimes||Technology vendors, tax guidelines, private consulting companies, IEEE||Some data is provided to give a sense of the order of magnitude of the lifetime differences between different technology classes only. Extreme caution is required when using any of this data, as O&M and component lifetime data is sparse, the effects of PV on device O&M and lifetimes is not well known or documented, and O&M and lifetimes can vary significantly even without the presence of PV due to differences in operating conditions.|
Organization of Each Page
In order to make this database more machine readable, we have implemented a common structure for each sheet. This structure is as follows:
- Row 1: Sheet title
- Row 2: Blank
- Row 3: Field names/header
- Remaining rows: Data entries
New cost breakdown data have not been added since the previously release of the database; they have only been moved to a separate file titled ‘Cost_breakdown_data_2018.xlsx.’
We include the units for upgrade parameters in the field name (e.g. ‘voltage’ is now ‘voltage_kV’ for voltage regulators).
Dollar Years and Inflation
All costs in the database are in the same dollar year as the date listed. For example, if the date field says ‘2014,’ the costs are in 2014 USD. For analysis or comparison of unit costs, all data should be converted into a common dollar year by adjusting for inflation as needed.
Low-, Mid-, and, High-Cost Numbers in the Database
In some of the sheets, an entry (row) in the database may have a low-, mid-, and high-cost number associated with it. These represent the range of costs from the specific data source associated with that entry or row, not the range of costs for that type of upgrade generally (e.g. across the United States or all data collected). For the tabs where some sources provided a range and others provided only a single data point, the single data points are categorized as mid.
However, these cost ranges do not capture the full range of costs that might be observed for any given scenario.
Data Gaps and Needs
We are seeking additional data on any components related to integration of PV onto distribution systems to include in this database. In particular, we are seeking additional data on:
- Data operations and maintenance (O&M) costs and component lifetimes for different components, and how the presence of PV affects these values
- Data on software and distributed energy management system costs
- D-SVC, D-STATCOM, and power regulators
- Communications and sensing equipment
- Experiences with advanced inverter costs (new inverters, retrofits, enabling existing features)
- Advanced metering infrastructure (AMI)
If you are willing to share any data on these components, please contact Kelsey.Horowitz@nrel.gov. We are accepting data on total installed costs of any component in addition to data on specific cost components (e.g., hardware-only costs). As discussed above, we are able to include data in the database in an aggregated and/or anonymized way if preferred in order to ensure protection of confidential, business sensitive, or proprietary information, and are open to signing nondisclosure agreements as necessary. Our team has earned the trust of our industry partners over the course of years of collecting and handling sensitive information related to PV component and system costs.
This research is supported by the Energy Department's Solar Energy Technologies Office.
How are you using this data? Contact us with feedback so we can better improve future versions of the database.
The Benchmarking Distribution Grid Integration Costs under High Distributed PV Penetrations workshop was held Sept. 19, 2017, on NREL's campus in Golden, Colorado. The workshop shared information and elicited stakeholder feedback on methods and terminology used for analyzing the monetizable grid integration costs and benefits of distributed photovoltaic systems on distribution grids.
Energy Analyst, Markets and PolicyFrancisco.Flores@nrel.gov