Date of Award


Document Type

Open Access Dissertation


Electrical Engineering

First Advisor

Roger A. Dougal


This research developed a method for autonomous operation of an inverter-coupled energy storage system for safe islanding- re-energization - and reconnection of the facility-island or microgrid to the utility / distribution circuit in a seamless manner in the event of a fault. This enables the energy storage system to function as an autonomous source within the utility power distribution system without the need to re-engineer the existing protection coordination in the distribution system. Fault detection based on voltage measurements from the inverter terminals are used to detect a fault and isolate the energy storage-powered facility with negligible power feed into the fault. The voltage-frequency modulation capability of the inverter is used to effect a controlled power restoration to the islanded facility which limits the motor starting inrush or the load pickup current to a predefined value. A subsequent reconnection to the main source / utility takes place when the islanded network is phase- and voltage-matched with the utility grid or the main source generator. The method is termed as "Fault-Aware-Soft-Restart" and is shown to be completed within a time period which can be as low as 0.5 s and is able to limit the inrush current to 150% of the nominal value. The controlled energization and synchronous reconnection of the energy storage-powered island, inherently avoids breaker reclosing transients and consequent false trips which degrades equipment reliability and lifetime. Voltage dips experienced by nearby customers and delayed-voltage-recovery effect caused by motor starting inrush resulting from generally

practiced restoration procedures in industrial plants are also avoided.

Decreasing cost of microsources such as microturbines, photovoltaics, and fuel cells combined with the growing demand for electrically powered commodities, against the limited capacity of centralized power generation has caused increased number of installations of distributed energy resources. Large scale (utility) battery energy storages up to 50 MW capable of providing - area power quality regulation to service reliability (UPS), are already installed and are reported in recent literatures. To prevent relay desensitization due to current infeed, and asynchronous reclosing, operational restrictions are imposed which offset the benefits of these energy storage systems. This work shows that by embedding appropriate controls into the energy storage inverter, both the major problems of unwanted fault-feeding and service restoration of "dead" circuits can be avoided. This innovation is the key toward enabling a plug‐and‐play environment in which many micro sources can operate harmoniously with one another with a minimum of expensive, custom site engineering.

Effectiveness of this method is proved for an industrial power network which suffers large financial losses caused by power interruptions. The method ensures personnel safety, and has the potential to avoid costs due to process downtime, motor startup and additional charges due to power quality problems during across-the-line starts. The test scenarios also include a shipboard MVAC power system in which the method is shown to provide a safe, fast and robust service restoration to a load zone. The method is shown to be very effective for such autonomous offshore power system, where improving system reliability and survivability with limited resources is much more challenging.

The dissertation also shows a method to model the electrical characteristics of a distributed generation power system during the different stages of service restoration using the soft-restart method. Assortment of load types in a facility, their startup current behavior during different stages of restoration and their diversity factors are taken into account for the model. Design parameters such as: energy requirement, location and inverter-size, as well as the power quality indices for the distributed generation system can be estimated from the model to devise the appropriate operating algorithm for the energy storage system module.


© 2013, Asif Anwar