Data dependent concurrent storage sizing and control design for frequency support in isolated power systems

As more intermittent energy sources are integrated into isolated power systems, maintaining nominal frequency under the uncertain power fluctuations becomes even more challenging. For that, properly controlled energy storage systems are commonly used to provide frequency support. However, the design of such controllers typically does not rely on system operation data, leading to oversized storage systems and in turn overpriced investments. This paper addresses this problem and presents a methodology for deriving controllers that optimally use a specified storage capability to achieve a target compensation level, given past information of the disturbances. To leverage between uncertainty and actuation (storage) magnitude, the manuscript proposes a data-based approach for deciding alternative combinations of storage size and corresponding control laws that ensure risk constrained robust frequency regulation. The proposed designs are capable of providing additional virtual-inertia services to the isolated system against a guaranteed level of security over all possible uncertainty realizations. An application to an offshore oil and gas platform with onsite gas turbines and locally produced wind power is presented to highlight the numerical properties of the proposed methodology.