Optimal Data-Driven Control of an LCC HVDC System for Real-Time Grid Frequency Regulation

Abstract

Recent advances in data sensing and processing technologies enable data-driven control of high-voltage direct-current (HVDC) systems for improving the operational stability of interfacing power grids. This paper proposes an optimal data-driven control strategy for an HVDC system with linecommutated converters (LCCs), wherein the dc-link voltage and current are optimally regulated at distinct HVDC terminals to improve frequency regulation (FR) in both rectifier-and inverter-side grids. Each HVDC converter is integrated with feedback loops for regulation of grid frequency and dc-link voltage in a localized manner. For optimal FR in both-side grids, a data-driven model of the HVDC-linked grids is then developed to design a data-driven linear quadratic Gaussian (LQG) regulator, which is incorporated with the converter feedback loops. Case studies on two different LCC HVDC systems are performed using the data-driven models, which are validated via comparisons with physics-based models and comprehensive MATLAB/SIMULINK models. The results of the case studies confirm that the optimal data-driven control strategy successfully exploits the fast dynamics of HVDC converters; moreover, cooperation of the HVDC system and synchronous generators in both-side grids is achieved, improving real-time FR under various HVDC system specifications, LQG parameters, and inertia emulation and droop control conditions. INDEX TERMS Data-driven control, dc-link voltage and current, high-voltage direct-current systems, line-commutated converters, linear quadratic Gaussian, real-time frequency regulation. NOMENCLATURE A.

Nonlinear Observer-Based Controller Design for VSC-Based HVDC Transmission Systems Under Uncertainties

“…As α 3 is not the actual control input, we introduce the new tracking error: e 4 = 1 L c C eq C eq x4 − α 3 . Combining ( 11) and (12), equation (11) becomes:…”

“…This implies that if the system's operating conditions change according to the point linearization, the system's performance may decrease. To deal with this issue, nonlinear controls have been proposed such as backstepping control [8], adaptive control [9], feedback linearization controller associated with gain scheduling [10], fuzzy control, and anti-windup action [11], optimal datadriven control strategy [12], model predictive control strategy [13], [14], sliding mode control [15], passivity-based control design [16], [17]. The point is that all the previous studies all designed in d − q reference frame.…”

Employing Machine Learning for Enhancing Transient Stability of Power Synchronization Control During Fault Conditions in Weak Grids

“…Similarly, machine learning has been used in protection of power electronic based systems and determining the possible outage of grid components in [21] and [22]. Further, some data-driven control has been presented to improve frequency regulation and lowvoltage ride through (LVRT) performance of converters in [23] and [24], and the data-driven control of DC power converters has been developed in [25].…”