Controlling a large constant speed centrifugal chiller to provide grid frequency regulation : a validation based on onsite tests

Abstract

High penetration of intermittent renewables may cause safety and stability problems to the electricity grid. Building thermal loads could contribute to the grid stability in the era of renewable energy and smart grid, because they are high and flexible. This study proposes a model predictive control strategy to control a large constant speed centrifugal chiller to follow grid frequency regulation signals while providing sufficient cooling to the building. Onsite tests were conducted to identify critical parameters in the process including time delay and ramping speed of chiller power consumption. A dynamic platform was built based on the studied system and fine-tuned using onsite tests data. Validation tests were conducted using the 40 min Regulation A and Regulation D test signals provided by PJM (Pennsylvania - New Jersey - Maryland Interconnection). According to the onsite tests, the total delay time and the maximum ramping speed of chiller power were 20–25 s and 2.53 kW/s, respectively. Validation based on the 40 min test signals shown that the composite performance scores were 0.901 and 0.885 for Regulation A test and Regulation D test, respectively. Continuous 12 h validation tests in a working day shown that the composite scores were 0.917 and 0.893 for Reg A and Reg D tests, respectively. And the studied constant speed centrifugal chiller could provide a regulation capacity of 5–7.5% of its nominal power.