Inertia Enhancement by Grid-Connected Power Converters with Frequency-Locked-Loops for Frequency Derivative Estimation

Renewable energy sources (RESs) have been extensively employed to replace fossil fuels for reducing carbon footprints. Since RESs are normally coupled to the mains grid by power converters without providing any inertia, the system inertia generated by synchronous generators continues to decrease, making modern power systems sensitive to frequency disturbances. To address the inertia concern, this paper proposes an inertia enhancement method implemented by grid-connected power converters. Specifically, the proposed method can be achieved by proportionally linking the time derivative of the system frequency and active power references of power converters. Although the differential operation cannot be applied due to high frequency noises, the presented frequency-locked-loop (FLL) estimates the frequency derivative signal in an alternative and practical way. Through the proposed method, all types of grid-connected converters can easily participate in frequency regulation. Simulation and experimental results are presented to illustrate the effectiveness of the proposed method.

I. Introduction

A paradigm shift in power engineering will transform conventional power systems into more sustainable, environmental-friendly, and high-efficiency smart grids [1]. As an important feature of smart grids, renewable energy sources (RESs), such as photovoltaics and wind, have already been greatly exploited to generate low-carbon electricity. However, the large-scale integration of RESs may bring in stability concerns to modern power systems [2]. This is because RESs are normally coupled to the mains grid through fast-response grid-connected power converters without providing any power system inertia, and hence, the system inertia generated by conventional synchronous generators decreases dramatically. Consequently, the system frequency may greatly deviate from its nominal value under severe frequency events, leading to the undesirable load-shedding, cascading failure, or even large-scale blackout [3].