Code

Numerical benchmarks and simulation code will be published soon.

UNICORN 4-bus test case

The UNICORN 4-bus test case is a simple distribution line that represents one of the possible configurations available at the Energy System Integration Lab (SYSLAB) at DTU

Despite its simplicity, this benchmark allows to demonstrate the suboptimality of completely decentralized Volt/VAr control strategies, including standard grid codes and standards like the VDE Standard VDE-AR-N 4105 (2018), the EU Commission Regulation OJ L112 (2016), and the IEEE Standard 1547-2018 (2018).

The mathematical analysis of this suboptimality is provided in [1]. The details of the benchmark experiment and its implementation on the SYSLAB facility is reported in [2].

Available measurements

Available actuation

Performance metric

Repository: https://gitlab.ethz.ch/bsaverio/UNICORN4 (in preparation)

[1] Saverio Bolognani, Ruggero Carli, Guido Cavraro, and Sandro Zampieri.
On the need for communication for voltage regulation of power distribution grids.
IEEE Transactions on Control of Network Systems, 6(3), September 2019. [ DOI | preprint ]

[2] Lukas Ortmann, Adrian Hauswirth, Ivo Caduff, Florian Dörfler, and Saverio Bolognani.
Experimental validation of feedback optimization in power distribution grids.
Electric Power Systems Research, 189:106782, December 2020. [ DOI | preprint | video 

UNICORN 56-bus test case

The UNICORN 56-bus test case is a distribution feeder that represents the three-phase backbone of the IEEE 123-bus distribution feeder. Loads have been lumped onto the three-phase backbone and balanced across the three phases. The power demand of each bus has been obtained by aggregating the power demand profiles available in the DiSC simulation framework over a time horizon from 6AM to 6PM. They represent the power consumption of about 1200 individual households from the area around the Danish city Horsens, obtained as anonymized data from the Danish DSO NRGi.

Two large PV generators have been added to the grid. If no volt/VAr regulation is performed, then the active power injection of these generators causes an overvoltage for a long portion of the observation period (see the plot No control). Standard Volt/VAr compensation, as suggested by grid codes and IEEE standard 1547-2018, provides only partial mitigation and does not use all the available reactive power capability of the converters (see the plot Volt/VAr droop). Feedback optimization achieves the goal of voltage regulation by coordinating reactive power compensation and curtailment of power generation (see the plot Feedback optimization). 

Available measurements

Available actuation

Performance metric

Repository: https://gitlab.ethz.ch/bsaverio/UNICORN56 

UNICORN 7019-bus test case

This benchmark represents a portion of the French subtransmission grid which hosts a large amount of wind generation. The controllable area comprises 31 buses at different voltage levels, and is connected to the larger French grid (7019 buses).

The benchmark data include a historical data series representing a quick increase in wind generation over the span of a few minutes.

Different scenarios are included:

Available measurements

Available actuation

Performance metric

Repository: https://gitlab.ethz.ch/ortmannl/online-feedback-optimization-for-transmission-grid-operation