| dc.description.abstract | Peaks in electricity demand shape power system operations at various scales. In large buildings, fees based on monthly demand peaks can comprise 30–70% of electricity bills. In sections of distribution grids, demand peaks can stress transformers, increasing the risk of equipment failure. Thermostatically-controlled loads (TCLs) – electrical loads that regulate temperature – such as air conditioners and heat pumps are key drivers of demand peaks. Therefore, analogously to the way a good driver is aware of neighboring cars, TCLs can coordinate with other TCLs within a building or section of a distribution grid to reduce demand peaks. Although schemes which enable groups of TCLs to limit their peak demand are abound, in most cases, these schemes require a reliable communication infrastructure to properly coordinate TCLs or similar loads. However, inadequate attention has been paid to developing novel or tailoring existing communication strategies to the requirements of the demand-leveling control of TCLs. Accordingly, this thesis takes a holistic approach to develop a low-cost, wide-coverage, and reliable communication infrastructure tailored to the nuances of the demand-leveling control of TCLs.
Although power lines were not originally designed for communication, their low-cost, widespread availability, and ease of connection make them attractive for facilitating the coordination of TCLs for peak demand control. Additionally, in contrast to wireless communication media, power lines are more robust against the risk of cyberattacks and extreme environmental conditions. Unfortunately, loads and electrical switchgear can easily interfere with power line communication (PLC) in both direct and subtle ways. However, due to the infrequent switching needs of TCLs and the long physical time constants associated with their thermal loads, the demand-leveling control of TCLs only requires low data rates for communication. This work leverages the low-data-rate communication requirement to develop a suite of signaling techniques which increases PLC reliability and facilitate the effective coordination of TCLs for peak demand shaving.
First, this thesis presents hardware which enable TCLs to communicate with one another using the power lines in a building or facility. Next, signaling techniques that substantially enhance PLC reliability by responding creatively to the unique requirements imposed by low-data-rate communication are presented. These techniques, which include chirp spread spectrum, distributed repeating, and time division multiple access, leverage so-called “quasi-peak” regulations to improve the reliability of low-data-rate PLC. By way of illustration, this thesis demonstrates a novel low-data-rate PLC system’s ability to facilitate effective demand-leveling control of TCLs in a large building or facility, using a 24-floor apartment building as a case study. The results show that the proposed low-data-rate PLC system provides a low-cost, wide-coverage, and reliable communication infrastructure for peak demand shaving. | |
