Impact of Price-responsive Load and Volt-VAr on Power Quality, Losses, and Customer Economics: A Distribution Grid Case Study

Published

May 17, 2024

Author(s)

David Holmberg, Thomas Roth

Abstract

This report presents insights into the impact of distributed energy resources (DER) price-responsive behavior on voltage for a distribution grid with significant customer photovoltaics (PV) and batteries in addition to price-responsive heat pumps and water heaters. Price response results in more variable power flows and voltages. While peak load is reduced, the very large amount of flexibility (load response) in this distribution grid increases the total power flow and voltage variation across the day while reducing hosting capacity. That is, while the loads on the grid provide significant flexibility to the bulk grid, this comes at the expense of voltage variability and significant power flow changes and even reverse power flows at the substation. The R4-1 distribution grid, with 523 houses representing most of the grid load, and each house with PV, batteries and price-responsive loads, was modeled in GridLAB-D. Simulations were run with GridLAB-D in a co-simulation environment using a NIST-developed device controller. Voltages, power flows, transformer loading, and losses are examined at different points on the grid. Simulations include price-response as well as Volt-VAr control to see how Volt-VAr can support the voltage impacts of price-responsive devices. Devices responded primarily to day-ahead market hourly prices (DAP). Response to 5-minute real-time market prices (RTP) was also examined. Load controllers seek to maximize economic benefit for the customer, reducing consumption (or discharging storage) during high prices and shifting load to low price times. In addition to load response, Volt-VAr control of PV and battery inverters was used to demonstrate the amount of voltage support that could be provided and was shown to successfully pull the mean grid voltage down within ANSI Range A limits. The grid flexibility (amount of load that can be shifted using setpoint temperature adjustment on the heat pump and water heater plus battery charge/discharge) equals as much as twice the baseline peak grid load. The peak load for these experiments was 3 MW. The flexibility on this R4-1 grid moved demand up or down 6 MW when all responsive loads are turning on and off together due to rapid price changes. This in turn can move grid voltage by 4 V (0.03 pu). However, this amount of movement was only seen for the more volatile RTP price signal, not for response to the DAP signal where voltage range increased 25 % over the baseline 2 V range. The observed decrease in hosting capacity is primarily a function of the mismatch of the flexibility response on this distribution grid compared to the grid as a whole. The results here provide a glimpse of challenges that could be encountered with price signals and price-responsive devices when too many devices respond to strong price signals. Recommendations are made for how a utility might craft a price signal that better supports voltage control. Customers' payment for electricity depended greatly on the tariff (price signal) with average electricity charges across two days ranging from $2 up to $10.

Citation

Technical Note (NIST TN) - 2289

Report Number

2289

NIST Pub Series

Technical Note (NIST TN)

Pub Type

NIST Pubs

Download Paper

https://doi.org/10.6028/NIST.TN.2289

Local Download

Keywords

Distribution, dynamic price, load-shifting, price response, voltage control, Volt-VAr

Citation

Holmberg, D. and Roth, T. (2024), Impact of Price-responsive Load and Volt-VAr on Power Quality, Losses, and Customer Economics: A Distribution Grid Case Study, Technical Note (NIST TN), National Institute of Standards and Technology, Gaithersburg, MD, [online], https://doi.org/10.6028/NIST.TN.2289, https://tsapps.nist.gov/publication/get_pdf.cfm?pub_id=957685 (Accessed June 3, 2024)