Smart-Grid-Enabled Distributed Voltage Support Framework

The motivation for this research lies in the use of emerging smart grid technologies, such as smart inverters, to supply reactive power as a means of distributed reactive power support. Power factor compensation closer to the load improves transmission line loading and efficiency. In the distribution networks, reactive power support not only minimizes the system losses, but also improves the feeder voltage profile. The focus of this research was on developing a smart-grid-enabled control algorithm that can determine the amount of reactive power injection or absorption required at each location to minimize the deviation from the control voltage level. Once the voltage level has been achieved, implementing conservation voltage reduction further benefits the system by increasing energy savings and extending equipment lifespan. We examined requirements for a secure communication framework to interact with the large number of devices that would be present. In general, reactive power support occurs at the substation level, whereas the communication advantages and system feedback provided by smart-grid technologies, such as smart meters, facilitate an extensive reactive power support scheme that reaches all the way to the end users. This scenario involves major challenges in ensuring high security to prevent adverse effects on the system; information received by the devices must be trustworthy so they will respond only in an intended way, and communication between the control center and the end users is important. We also looked at the best way to utilize the support from a power system perspective, and investigated the implications for potential contingencies of the system, so that the system can be designed to avoid them. We modeled example power systems, such as distribution feeders, a to show the benefits of local injections of reactive power. We developed algorithms to determine the validity of using distributed reactive power control with different assumptions about the cyber infrastructure. We also used OpenDSS and Matlab to develop algorithms that combine reactive power support, conservation voltage reduction, and on-load tap changer (OLTC) control to find the optimal voltage profile for the feeder system in order to minimize system losses and save energy consumption. Finally, both centralized and distributed minimization problems were solved, and an adaptive alternating-direction method of multipliers (ADMM) control algorithm was developed to minimize the communication overhead.