Smart Energy Technologies & Energy Storage
Aim/Objective: Distribution feeders with a high penetration of distributed PV systems can experience voltage variability caused by the changing load flow injections corresponding to the availability of the solar resource. The rapid, uncontrolled voltage fluctuations can affect the system. Tap-changing voltage regulators, switched capacitors, and smart inverter functions have been used to stabilize voltage. However, in some cases, when these options are insufficient or unavailable, dedicated reactive power control devices such as Dynamic Var Compensators (DVCs) can be used for local voltage stabilization. This presentation introduces a statistical framework to characterize voltage variability and stabilization, proposes a data-driven approach to DVC placement, and presents the results of a planning study on a real feeder using real data.
Methods: We define voltage stabilization as a reduction in voltage variability: both dynamic range and fluctuations are considered. To quantify variability and stabilization, the voltage variability index (VVI) is developed based on the variability index for irradiance and PV output variability. We introduced an approach to determine how DVCs can contribute to voltage stabilization by addressing both factors of variability. To fully leverage the capability of a DVC or multiple DVCs, location planning should include consideration of the full range of locations and operating conditions using Quasi-Static Time-Series (QSTS) simulations. The following procedural algorithm is proposed: 1) Collect feeder data 2) Simulate to determine present variability 3) Rank variability at possible DVC locations 4) Simulate DVC(s) to determine stabilization affect 5) Monitor the location(s) and calibrate the DVC(s).
Results: We present case-studies in which three feeders on a Duke Energy Carolinas substation, that has 10 MW of distributed PV installed, are considered for DVC deployment. For each feeder, a QSTS simulation was performed and the VVI was computed at the downstream terminal of each three-phase line in the feeder. Results show that a Voltage Change Reduction (VCR) of 18% to 28% can be achieved. We will also show actual field validation results and quantify the overall improvements due to the installation of 180 KVar and 120 KVar DVCs at two different locations on feeders connected to the same Substation.
Conclusion: Simulations using historical load and PV data can be used to determine the location with maximum uncontrolled voltage variability. This location is sensitive to fluctuations in load and PV generation and is therefore likely to be biased towards the locations of large PV systems, large loads, and remote locations. Simulation tools today do not include models of feedback-controlled DVCs. A droop-controlled reactive power device was used as an alternative. Simulation results show that DVCs can reduce variability and improve grid voltage stabilization. Statistical analyses were required to evaluate voltage improvements. Tuning and coordination will be important for effective DVC performance in the field.