The U.S. Department of Energy (DOE) is trying to help connect more distributed photovoltaics to the grid. Researchers from the DOE's Sandia National Laboratories, the Electric Power Research Institute (EPRI) and the Georgia Institute of Technology released a report describing a new methodology to help utilities better evaluate how PV can affect distribution system operations.
The report, ‘Time Series Power Flow Analysis for Distribution Connected PV Generation,’ demonstrates how quasi-static time series (QSTS) simulation and high time resolution data can be used to quantify the potential effects a distributed PV system might have on the grid. The report also illustrates some strategies for mitigating them.
‘Our overall goal of this report was to provide a new methodology to help utilities better evaluate how these PV systems affect their operations,’ says Robert Broderick, co-author of the report and a principal member of the technical staff in the distributed solar grid integration group at Sandia.
Before a distributed generation system can connect to the grid, it has to undergo an interconnection study process. Based on rules from the Federal Energy Regulatory Commission, some projects can be fast-tracked, and others that are deemed to have high risk must undergo a full system-impact study. Those studies are time consuming, and the methodology has certain limitations that affect accuracy.
The Sandia, EPRI and Georgia Tech researchers offer a methodology that is designed to capture and assess the daily changes in load – and also PV output – on distribution equipment. The standard way to do this, Broderick says, is a snapshot approach. The power flow analysis evaluates the system at two crucial points: a summer peak load and a winter peak load. Broderick explains that the limitation of that approach is that the load can be highly variable throughout those seasons and the rest of the year. Also, the PV on the system can have a highly variable profile.
‘Just looking at it one or two times during the year does not help identify concerns in terms of risk,’ Broderick says. ‘The methodology may be overly conservative in assessing what the impact is.’
An overly conservative assessment might overestimate the potential negative effects of high levels of PV on the grid, which can slow down growth and deployment of PV. Instead, QSTS simulation uses multiple time-consecutive power flow solutions. QSTS requires a larger quantity and more granular level of data, and it offers a more accurate and detailed analysis.
The researchers applied the QSTS methodology to four sample distribution feeders with high PV deployment. The four feeders had different characteristics, such as minimum and peak loads, distance from the substation, and the percentage of residential and commercial load class. The report demonstrated the steps needed to perform the QSTS simulations and also described the data needed to drive the simulations.
‘We could basically determine the impacts on the distribution system, both positive and negative, that would have been hidden with a snapshot analysis,’ Broderick says.
Co-author Roger Dugan, a senior technical executive for EPRI, says the concept of using many data points instead of a snapshot is evolving.
‘We learned back in the mid-90s you didn't get the right answer for all the impacts of solar PV on the distribution system until you get time series simulations,’ Dugan says. ‘So a series of snapshots in small time increments gives us a picture of the impact on utility devices that operate in a few seconds, such as voltage regulators or capacitor switches. The smaller the increment, the better the resolution.’
Dugan, who is the author of Open Distribution System Simulator, or OpenDSS, an early power system analysis software, says future reports will examine smart inverters and smart meters in interconnection studies.
Broderick hopes the new report will help the industry adopt these methods and tools.
‘As deployment levels of distributed PV increase, it's actually driving change,’ Broderick says. ‘It's driving utilities to improve their planning, improve their operations, and it's actually helping utilities to modernize the grid.’
He adds that not only can utilities use the methodology, but he also has seen interest from developers who might need a new way to decide, for example, which of three parcels of land would be optimal for a PV system.
Nora Caley is a Denver-based freelance writer.